YALE UNIVERSITY
ART LIBRARY

MASTERPIECES OF
GREEK SCULPTURE

Printed by GlesecKa & Dovrlent, Leipzig (Germany).

MASTERPIECES OF GREEK * » »
SCULPTURE *
A SERIES OF ESSAYS
ON THE HISTORY OF
ART fr BY ADOLF
FURTWANGLER fr
EDITED BY EUGENIE
SELLERS * WITH
NINETEEN FULL-PAGE
PLATES &P TWO HUNDRED
TEXT ILLUSTRATIONS

NEW YORK : CHARLES SCRIBNER'S SONS : MDCCCXCV

To HEINRICH BRUNN
& ERNST CURTIUS
In Respectful Devotion

AUTHOR'S PREFACE

A'

NY unprejudiced attempt to form, from the
text-books now in use, an estimate of the
state of our knowledge of the history of
art among the Greeks will force us to own with
shame that we appear to know and to utilize the
monuments far less than did Winckelmann in his
day. Winckelmann's History of Art is wholly based
upon a fresh and personal observation of the monu
ments, of which he makes a constant and extensive
use. Our more recent histories of art are wont to
take into consideration only one and the same small
group of monuments, an accidental section of the
mass of what has been preserved. Ever since Brunn
laid with a master hand the foundation for the
history and the characterization of the Greek artists,
so far as these can be gathered from the traditions of the ancients, writers have for
the most part been content to repeat what he established, only perhaps adding an
occasional ' selected ' monument to trim, as it were, the fabric of his weaving. Still it
was quite justifiable to be thus prudent and cautious in dealing with the monu
ments, so long as we only painfully groped our way amidst the wealth of remains,
without knowing how they should be utilized : it was certainly better to limit
investigation to the little which was ascertained, than to venture without stay or
support upon the ocean of what actually exists.
Modern science, however, has afforded us a stay which enables us to keep
steadily in the direction of our goal. Any one who understands how to observe the
monuments, and who is willing, with indefatigable ardour, to test afresh and compare
all forms, may nowadays, by means of photography, which helps to fix the individual
objects, obtain a picture of Greek art far more richly coloured than the pale and
meagre image we have hitherto possessed.
I know that many fellow-scholars are engaged upon these lines, and it is only by
combined effort that we shall be able to advance. Each must bring what contribu
tions he can. The investigations which I publish here are all closely interconnected ;
their ultimate object is to gain from the monuments a new and solid foundation on
which to build a history of statuary among the Greeks, for, before undertaking to

viii AUTHOR'S PREFACE
draw from the monuments — as has long been my ambition — a new presentment of
this history, the way must be smoothed by isolated inquiries.
The first of the essays, that on Pheidias— containing the identification of the
Lemnia— was the external inducement to the publication of the other sections, most
of which had been written previously. Starting from the Lemnia, the inquiry extends
to the whole circle of Pheidias and to the Akropolis, the chief scene of his activity ;
thence it passes on to his more independent contemporary Kresilas, and from him harks
back to Myron. Myron affords a transition to Peloponnesian art, and the inquiry
passes naturally on to Polykleitos. The powerful effect exercised by the creations
of the latter upon the Attic artists of the fourth century leads accordingly to
Skopas, Praxiteles, and Euphranor, and finally to a glance at Lysippos. Some
special tendencies of Skopas and Praxiteles are followed out more in detail in
the next section on the Venus of Milo, while the chapter on the Belvedere Apollo
reverts from the creations of the fourth century to those of the Pheidian epoch.
The method of investigation is throughout the inductive, which passes step by
step from one conclusion to another. Much which belonged together has thus
inevitably been wrenched apart ; but this fault may be rectified by the index, which
will enable the student readily to put the material together in its systematic historical
order. The museographic index will also be welcome to the reader.
The material treated in this book consists for the most part of antique
copies, since, except for the well-known few though splendid exceptions, the
best creations of antiquity survive only in copies. True, to the number of these
exceptions — original works of first-rate artists — I add one at least (Plate XVII.) ;
yet the multitude of other masterpieces whose traces are here followed are still
known only at second hand.
The increasingly rich discoveries of original works on Greek soil have lately
somewhat thrown into the shade the study of the copies, for which we are mainly
indebted to Italy, not to the advantage of our science. The original sculptures from
Greece are, with those rare exceptions to which I have already alluded, works of the
second or even inferior rank. The Roman copies, on the other hand, have preserved
that pick from the masterpieces of the classical epoch which pleased ancient taste and
connoisseurship in the times of highest culture. It is the pick of the best and the
most famous that antiquity possessed. Among these copies it is that we must look
for the masterpieces mentioned by the authors, for the statues that made epochs or
initiated movements. Were we to possess only copies of the noble creations of a
Raphael, a Michelangelo, or a Rembrandt, these would certainly be better worth one's
study than the hosts of other originals of the time.
It is manifest from this that our first duty is accurate examination and criticism
of the extant copies. This kind of study is as yet only in its infancy ; but it is
precisely here that photography is of invaluable assistance, and by its aid we may hope
to make rapid progress. Above all, an almost painful accuracy is required. The
older works dealing with our store of copies suffer almost throughout from obscurity
as to what is to be really regarded as a copy. No adequate distinction is drawn
between copies and adaptations or even figures which are merely similar. Formerly,
indeed, relatively little was known of the existence of actual close copies, and for the

AUTHOR'S PREFACE ix
most part only few late reproductions or variants were believed in. Two mistakes in
particular were very frequently made : either actual copies of one and the same
original were taken for different modifications or variants of one ' type ' ; or else
copies of quite different originals were taken for later copyists' variations of one
original. As regards principles and method in the criticism of copies, many rules might
be laid down — yet I have never been able to see any use in talk about method,
much less in boasting about it. Rules would never embrace, even remotely, the
whole range of possibilities presented by reality. Method can be shown only by
application. The researches in this book show by numerous instances what are the
requisitions I think necessary for the criticisms of copies, and how I think copies
should be dealt with (cf. Index under COPIES). Just a few remarks may however be
in place. In the more delicate appreciation of the copies, all of course depends on a right
discrimination between what is derived from the original and what is added by the
copyist. This point will always be a rich mine of error in inquiries of this kind ; yet
a long familiarity with the monuments, and a sense sharpened thereby, will preserve
one from at least gross mistakes. Further, it is above all important that the
monuments should not be torn out of their setting. Any inquiry concerning the
masterpieces of antiquity must, if we would avoid error, be made only in a wide
connexion. The individual work must be replaced in the environment which
conditioned it, and through which alone it can become intelligible ; and in our
inquiry concerning any single monument we must keep all the others to which it
is akin steadily in view. This procedure alone can guard us from the errors which
must necessarily arise from the dilettante habit of isolating works, and connecting
them arbitrarily with the names of artists. Finally, I may add, I have made it my
first and most natural rule to discuss only those works of art which I have myself
examined. In the relatively few cases where this was not possible I have specially
noted the fact. Ocular examination can, however, be at times replaced by good
photographs ; but the illustrated works, and the large one by Clarac in particular,
are as good as useless for our purpose. It is of course my wish that my readers
should be in a position to compare for themselves as many originals, casts, and
photographs as possible. The illustrations in the book reproduce the most important
and least accessible monuments.
At first many will doubtless think I have been too bold in my attributions of
extant works to celebrated artists. But on deeper familiarity with the actual objects
these doubts will vanish more and more. I can at any rate say for myself that I have,
I believe, been sufficiently critical of my own conjectures, and that I have scarcely
allowed one to stand that has not been practically laid aside and tested by repeated
trials, and has approved itself in a wider connexion. But it may be further objected
that it is not yet time, while we are still so behindhand in the knowledge of the
general development of the separate forms, to inquire into the individualities of the
several artists. The study of these forms, however — in so far as it touches upon the
efflorescence of plastic art and so soon as it enters into more delicate distinctions — is
inseparable from — nay, even identical with — the inquiry into the individualities to
b ~

x AUTHORS PREFACE
whom precisely this or that particular development of form is due. Only the general
pervading features of this development must be assumed throughout as the solid
basis of the inquiry. It were indeed much to be desired that this basis should be
effectively laid down once and for all in some special treatise, since unanimity even
in this respect is still lacking among scholars.
The more deeply we penetrate into that selection of antique masterpieces which
undoubtedly survives in our extant copies, the more forcibly are we impressed by the
individuality of the great artists of the best period. I venture to hope that, beyond
the circle of specialists, the general reader — for the book is intended for him also —
will see reason to modify his conception of the antique, and will grant that it includes
a far greater range of individual development than has hitherto been supposed. It
is true that the great distinction between ancient and modern culture still holds
good : the untrammelled, free individuality, at once the strength and the weakness of
modern artists, was quite foreign to antiquity. The ancient artist clung to established
types of far-reaching influence. He obeyed laws and rules in his treatment of bodily
forms and of attitude — and for this the book affords evidence step by step — which he
modifies and alters, extends and fashions after new inspiration, but which yet impart
to his whole production a something inevitable, typical, known by rule. It is this
that brings about that unique effect of the antique which Goethe sums up in the
words, ' These great works of art have been brought about in the same wise as the
operations of nature. Everything arbitrary, everything self-conscious, disappears :
there is Necessity, there is God.' But to penetrate into the whole mystery of indi
viduality, and to learn to recognize there also the divine necessity— this was reserved
for the moderns. A. FURTWANGLER.
Berlin, October 1893.

EDITOR'S PREFACE

T

'HE task of editing a book like the present,
which has been received almost with ac
clamation by scholars of all schools, has
been a responsible one, and I am quite conscious
that I may have succeeded after all in satisfying
neither the student nor the general reader. The one,
preoccupied with detail, will perhaps complain of the
omissions, while the other, in search only of a vivid
impression, may be repelled by the length and depth.
Two main alterations in the plan of the book must
be noted at once : the two passages treating of
archaic art (pp. 675 — 732, pp. 250 — 257) have been
omitted, partly owing to their fragmentary nature,
and partly because the author contemplates the
publication of a series of essays upon the archaic
art of Greece in which these his first sketches will be worked up and expanded. In
the second place, the long and difficult chapter on the temples of the Akropolis
(for the scholarly translation of which I am indebted to Miss Margaret Alford)
has been printed as an Appendix, for it seemed best not to interrupt the sequence
of the artistic inquiry with an Essay which, though it bears closely upon Pheidias,
is mainly of historical and topographical interest. For the rest, I have made
it my aim, as far as possible, to disengage the author's arguments from all such
controversial matter as might cumber or obscure them. For instance, it seemed
to me that the claims of scholarship would be fully satisfied if the numerous
theories put forward from time to time to discredit the Pheidian authorship of
the Parthenon sculptures, or the fluctuations of opinion with regard to the Kresilaian
Diomede or the Myronian Perseus, were relegated to footnotes which should provide
the learned and the curious with all necessary references. On the other hand,
when Professor Furtwangler crosses swords with champions like Dorpfeld on the
subject of the ' Opisthodomos,' or Loschcke on the date of the Trial and Death of
Pheidias, we feel that contact with such opponents' arguments strikes fire from his
own, so that all passages of this kind have been faithfully preserved. The same
may be said of the chapters on the ' Venus of Milo ' and the ' Apollo of the
Belvedere,' and above all of the whole Essay on the Akropolis temples, for here

xii EDITOR'S PREFACE
again argument and controversy are so closely interwoven that to shorten the
latter would be materially to weaken the former. In these chapters, therefore, the
only alterations are those that have been introduced by the author himself. These
and a number of smaller omissions and additions made by him throughout the whole
of the book call for no special comment ; they will be easily detected and appre
ciated by the reader acquainted with the original. The majority were necessitated
either by subsequent literature or by subsequent discovery. The portions of the
German edition that were printed as ' Nachtrage ' have been inserted in their proper
place in the text, while every effort has been made to give references to the literature
that has appeared since the publication of the German book a year ago. In my
revision of the translation generally, I have ventured upon compression wherever this
was possible without injury to the sense, while in one or two instances I have left
rather more to the imagination of the reader than is usual in a German work of this
nature. Thus, after the exhaustive analysis of the forms peculiar to Myron given
on pp. 165 — 202, it seemed unnecessary to repeat them in detail, in the case of each
single statue or head which the author in a concluding section (XIII.) has grouped
about this artist. Professor Furtwangler has himself found time in the midst of his
various occupations to bestow a general supervision upon the English edition ; nor is ,
it necessary to say that no editorial alterations have been introduced without his
express sanction, while not a few have been planned in consultation with him.
The number of illustrations, which in the portion chosen for translation was
only 162 (including the plates), has been raised to 207. These 45 fresh illustrations
have been selected on the same plan as that already pursued by the author, to bring
into notice new or almost forgotten monuments. In their arrangement I have tried
to convince the reader of what great results might be achieved with the help of a
collection of casts, comprising not merely a few representative works, but all or nearly
all the extant products of classical art whatever their period, and supplemented by a
complete series of photographs. From the three statues reproduced side by side on
page 87, it must surely appear that their attribution to one and the same artist is no
matter of guess-work or of facile intuition, but the reasoned result of such a compara
tive study of form as is possible only in some comprehensive collection of casts as at
Dresden or Munich. In like manner I trust that the full illustrations in the chapter
on the Amazons (pp. 128 — 141) will enable the reader to take in at a glance less obvious
but essential differences which, when we have only memory to trust to, are apt to
become merged and hidden in external resemblances of type and dress. In this
connexion my thanks are due to the Marquis of Lansdowne and to Mr. Astor for
allowing the finest copy in existence of the Polykleitan Amazon to be worthily
published (Plate VIII.)
The very few illustrations which in the German edition were still repeated from
former publications have now been replaced from photographs— of the originals,
wherever this was possible, or at any rate of casts when, as in the case of so many
Italian galleries, bad lighting and other causes often make photography impossible.
The Bologna head (Plate III.) has again been reproduced only from the cast, for in
spite of the trouble so courteously taken by the Director of the Museo Civico, Professor
Brizio, the stained condition of the marble has made it impossible to obtain a negative

EDITOR'S PREFACE xul
sufficiently good for reproduction in photogravure. I had also hoped to replace the
poor illustration of the once celebrated Hope Athena at Deepdene by a plate from
the original. Having failed, however, to obtain from the present occupant of Deepdene
so much as an answer to my' applications for permission to photograph the statue,
there was nothing for it but to repeat the illustration taken from the Ancient Specimens
(Fig. 27). In the matter of illustration generally, I have to thank Mr. A. S. Murray
for the special facilities accorded to me for photographing in the British Museum,
Mr. Barclay V. Head for his assistance in the preparation of the plate of coins (VI.), Dr.
Paul Herrmann of Dresden for the fine new negatives of the Lemnia (Plates I., II.*,
and III.) In addition, Herr F. Bruckmann of Munich and M. A. Giraudon of Paris
have generously allowed me to reproduce a number of their photographs. Of the new
plates there are three to which I should like to call special attention : the superb head
from Beneventum in the Louvre (Plate XIV.), which should rouse us to a sense of
what precious relics of the ancient statuaria may still lie hidden in our museums ;
the 'Aberdeen head,' that exquisite fragment in our own British Museum which
escaped so curiously long the eye of both connoisseurs and archaeologists, and in
which I think it not too bold to recognize an original from the hand of Praxiteles
(Plate XVIII.); finally, the grand and presumably original head of the Skopasian
Meleager, which, though it has already been well reproduced in the Antike Denkmdler,
deserves to become known amid the grace of its Roman surroundings (Plate XV.)
By the courtesy of the author and of his English translator, Mr. James F.
Muirhead, I have been enabled to refer throughout to the English edition of Professor
Helbig's Museums of Classical Art in Rome, and I trust that this book, which will
appear almost simultaneously with the present one, will give a fresh impulse to the
unprejudiced study of the treasures of those Roman galleries to which, as Professor
Furtwangler shows, we must still go if we would find or restore the ancient ' master
pieces.' The late Bishop Wordsworth, during his travels in Greece, likened the
country, from the point of view of what remained and did not remain, to a manuscript
torn indeed and defaced, but ' not yet, like Rome, a palimpsest' Now it is precisely
this fact that it is a palimpsest that still secures to Rome its archaeological pre
eminence, for, with the exception of a fragment recovered of late years here and there,
the record which we are trying to decipher has been forgotten or destroyed in its
native land. The very beauty and uniqueness of these fragments as works of art
blind us to their incompleteness as evidence, but if we wish to gain some idea of
the whole story it is to Rome that we must go, and there accustom ourselves
to spell it out through the mistaken interpretations and ignorant glosses of the
copyists. Per casus varios, per tot discrimina rerum
Tendimus in Latium.
But it would be an error to suppose that Italy, la mere savante de toute
Renaissance, is only fruitful in copies. There is the Ludovisi throne — in which the
infant art of relief seems to have reached at a bound the limit of its accomplishment —
and now at last we are in a position to place side by side with the Hermes itself an
original from the hand of the master whom the consent of ancient connoisseurs
ranked with Praxiteles. In archaeology, unfortunately, we cannot pick and choose ;

xiv EDITOR'S PREFACE
all that we can do is to make the most of what has been thrown up from the wreck
of Time, but ' where are they painted that are lost ? ' It was therefore only natural
that for a time at least we should regard the Hermes as supreme, if only because he
was solitary ; yet whoever will compare him— soft, self-involved, with lips just parting
as vague voluptuous languors steal over him— and the Meleager of Skopas, with that
look all outward and upward of some inspired ' pilgrim of eternity,' will gain some
idea of what we must have lost through the accidental eclipse of this great genius.
Lastly, it is hoped that this book, as it exhibits a picture, will also discover
a process. Of the exact nature of that process it would hardly be necessary to
speak, but for the fact that the copiousness and brilliancy of the achievements of
a single critic in the more popular field of Italian art have thrown us in England
into a state of naive commotion akin to that of the ancient Mexicans, when, having
never seen a horse, they mistook the troopers of Cortes for a new species of animal.
For, as the critic in question never appears without his hobby, the two coalesce, as it
were, in our imagination, until we think and speak of that which is nothing but the
course and condition of all fruitful inquiry as if it were the honorific appendage of a
particular name and the abnormal product of a particular field. On the contrary,
the present book is from first to last an example of the inductive method, which,
though it has never been applied before on so extensive a scale to the art of Greece,
is, in principle at least, as old as Winckelmann. But here observation and com
parison do not end in themselves ; they rest upon a basis of history and philology,
and the result is that we have the reproduction of a development, not merely the
recension of a catalogue.
It cannot be denied, however, that the strength and flexibility of our instrument
are often strained to the full by the very nature of the material it works in. That
material is, as we have seen, not only fragmentary but secondary, while the
limitations of sculpture as an art betray themselves in an external uniformity which
always impedes, and sometimes baffles, our analytic research of variety. If observa
tion comes upon a gap, theory leaps ahead, like a man's shadow that gets in front of
him as soon as he begins to move away from the light. Then there is the constant
temptation to explain too much, to impose a large significance upon minute features,
as to which we might say in words borrowed from Johnson, ' the dull utterly neglect
them, the acute see a little, and supply the rest with fancy and conjecture.' So much
for the defects of our ' method.' They know them best who use it most diligently
and most skilfully ; nor would there be any occasion to insist upon them at all, were
it not for the numbers of those to whom, if we may judge from their attitude of
barren negation, 'willing to wound but yet afraid to strike,' it would seem as if
discretion were the better part of discovery.
It is therefore in a double aspect as matter and method that these Essays are
now offered to the English reader, in confidence that what is not final will yet be
found fruitful ; and — ¦ Was fruchtbar ist, allein ist wahr.
E. S.
November 19, 1894.

ADDITIONAL NOTES.

I. P. 95 sqq. : Diosatri of Monte Cavallo. In a letter which I received a few days
ago, Professor Furtwangler quotes an interesting statement made by Julius Lange
{Thorwaldsen, Fremstilling af Mennesket, p. 9), to the effect that Canova said of the
Elgin marbles in 1803 that only one single antique in all Rome was conceived in their
grand style — namely, the finer of the tzvo Colossi of Monte Cavallo. I also find it
stated by Memes, Memoirs of Canova (Edinburgh 1825), p. 291, that ' of the latter (i.e.
the Colossi), even at a subsequent period of life, he (Canova) made a constant morn
ing study for years, in relation to his own improvement, and to establish the style of
Phidias in the Elgin Marbles! — E. S.
II. P. 132, 1. To the replicas of the Kresilaian Amazon should be added a head
in the Jacobsen Collection (1073 a), poor and much restored, but of value as having on
the right side, in the line of the ear and near the crown, the remains of a rectangular
support, which once connected the head with the right wrist ; this confirms the restor
ation proposed on p. 132. The head may possibly be identical with Michaelis 0.
III. P. 346 .r^. : The Aberdeen Head. Lord Stanmore has the kindness to inform
me that it is certain the head came direct from Greece, and adds : ' All the fragments
my father brought with him from thence were placed together by themselves, and this
head was among them.' — E. S.

ERRATA.
Page 29, line 4 from foot of page (text), for ' 473 ' read ' 470.'
„ 42, lines 7, 10, and 17, for ' archaic ' read 'archaistic.'
„ 78, line 9, for ' Romani ' read ' Romano.'
„ 84, note 1, line 3, for 'bust ' read 'best.'

CONTENTS
l-AGE
PHEIDIAS . . . . . I
I. Discovery of the Lemnian Athena ......  4
II. Site of the Lemnia on the Akropolis. — Date and Dedication . 8
III. Comparison between Lemnia and Parthenos ... .10
IV. Analysis of the Lemnia ...... .13
V. Drapery and Pose of the Lemnia .... 21
VI. Monuments related to Lemnia and to Parthenos. — The Athena Promachos . 26
VII, The Olympian Zeus. — Trial and Death of Pheidias 36
VIII. Other Works related to the Lemnia. — The Master of Pheidias. — Early
Pheidian Works. — The Anakreon. — Pheidian Eros and Aphrodite . , 50
IX. Pheidias and his Pupils. — Alkamenes and Agorakritos . . 73
X. The Dioscuri of Monte Cavallo, and the Elder Praxiteles 95
XI. Pheidian Influences in Sicily and Magna Graecia. — Coins and Vases 104
KRESILAS AND MYRON .... . ... ... 113
I. Literary and Epigraphical Evidence for the Life of Kresilas 115
II. The Portrait of Perikles .... , . 117
III. The Diitrephes .... . 122
IV. The Amazon . . ...... .128
V. The Athena from Velletri . 141
VI. The Diomede . 146
VII. The Medusa Rondanini . . ..... 156
VIII. Statue of an Athlete at Petworth .... ... 161

IX. Relation of Kresilas to Myron.— The Riccardi Head. — The Diskobolos and
kindred Heads. — Pythagoras of Rhegium.— Myronian Portrait-heads
X. Statues by Myron.— Diskobolos and Marsyas compared.— Kindred Works
XI. The 'Cassel Apollo.' — Argive Influences traceable in Myron ....
XII. The Perseus . .

XIII. Myronian Female Head.— The Herakles Altemps and Kindred Works.
Asklepios in the Uffizi . . ....
XIV. The Munich Zeus and the First Argive School ....

1651 So
190
197
202
212

CONTENTS xvii
POLYKLEITOS . . .  P22i
I. Historical and Epigraphical Evidence . . 223
II. The Doryphoros . . 226
III. The Diadumenos . ... 238
IV. The Amazon . . 247
V. The Basis of the Statue of Kyniskos. — Statue of a Boy placing a Wreath
on his Head, and Kindred Works . 249
VI. The Basis of the Statue of Pythoki.es.— Statue of a Boy in Dresden: its
Adaptations and Derivatives . 262
VII. The Basis of the Statue of Xenokles. — The Idolino . . . 279
VIII. The Basis of the Statue of Aristion. — The later Polykleitan School. —
The Beneventum Head . . 287
SKOPAS. PRAXITELES. EUPHRANOR . . 293
I. Skopas. — Lansdowne Herakles. —Hermes from the Palatine. — Statues with
foot raised. — Ares Ludovisi. — The Meleager. — Athena 296
II. Praxiteles. — Date of the Hermes. — Works of the Artist's Early and
Middle Period. — Figures Leaning on a Support. — The Satyr and the
Eubouleus. — Works of the Artist's Later Period ; the Hermes and
Kindred Statues . 307
III. Euphranor. — Imitation of Polykleitos and of the older Argive Types. —
Bonus Eventus ; Dionysos ; Apollo Patroos ; Paris and Aphrodite ;
Athena. — Lysippos . . 348

THE VENUS OF MILO ... . . . -365
I. The Lost Inscribed Fragment : Discussion of the Statue's Provenance 367
II. Restoration of the Statue . . 378
III. Influences that affected the Artist of the 'Venus.' — Skopas. — Historical
Position of the Venus . ... . 384
THE APOLLO OF THE BELVEDERE . . .403
APPENDIX— THE TEMPLES OF ATHENA ON THE AKROPOLIS . . . 413
I. The 'Old Temple' of Athena .... .  415
II. The First Parthenon . . . . .419
III. The Parthenon of Perikles ... . 423
IV. The Erechtheion ....  ... 432
V. The Temple of Athena Nike . . . . . 442
VI. The Meaning of the Pedimental Sculptures of the Parthenon 451
VII. The Earth Goddess entreating for Rain, near the Parthenon . . 46S
INDEX ... . ... - 473

Note. — A number of the plates of the German
edition are noxv reproduced as text-blocks, while
for some of the text -illustrations of the German
book plates have here been substituted. Those
illustrations, whether plates or text-blocks, which
are completely foreign to the German edition,
and now appear for the first time, are marked
by an asterisk (*).

LIST OF FULL-PAGE PLATES
PLATE PAGE
I. Statue of Athena, Dresden (Giesecke and Devrient, Leipzig) . Frontispiece.
II. Statue of Athena, with Cast of Bologna Head, Dresden (Do.) . 8
*II*. Statue of Athena in Dresden (from the cast) (Do.) 12
III. Head in Bologna (from the cast) (Do.) . . 16
*IV. Bearded Head, Louvre (Maison A. Braun, CUment dfCie., Paris and Dornach) 90
*V. Head of a Youth from Parthenon Frieze, British Museum (Giesecke and Devrient,
Leipzig) . .  . . . .96
*VI. Greek Coins (Autotype Co.) ....  . . . 106
1, 3, 5, 6. Thurii, — 2. Sybaris. — 4. Neapolis. — 7. Terina. — 8, 10. Pandosia. — 9. Hyria. —
II. Neapolis. — 12-15. Syracuse. — 16, 17. Naxos. — 18. Herakleia. — 19. Phokaia. — 20.
Lokri Epizephyrii. — 21. Roman denarius. — 22,23. Elis. — 24,25. Amphipolis. — 26. Miletos.
— 27-29. Klazomenai. — 30, 31. Athens. — 32, 33. Argos. — 34, 36. Alexandria Troas. —
35. Heraia. — 37. Coin of Antoninus Phis. — 38. Corinth.
VII. Portrait of Perikles, British Museum (Giesecke and Devrient, Leipzig) . . 118
*VIII. Amazon, Lansdowne House (Do.) . 134
IX. Head of Perseus, British Museum (Do.) . . 138
X. Head of a Diadumenos, Dresden (Full Face) (A. Frisch, Berlin) 240
XI. Head of a Diadumenos, Dresden (Profile) (Do.) . . , . 242
XII. Statue of a Boy, Dresden (Do.) . . 266
XIII. Bronze Statuette, Louvre (Giesecke and Devrient, Leipzig) . . 280
XIV. Bronze Head of a Boy, Louvre (Maison A. Braun, CUment & Cie., Paris and Dornach) 290
*XV. Head of Meleager, placed upon a Praxitelean Apollo, Villa Medici (Do.) 306
*XVI. Head of Eubouleus, Athens (Do.) 330
XVII. Aphrodite, Collection of Lord Leconfield (F. Hollyer, London) . 344
XVIII. The 'Aberdeen' Head, British Museum (Giesecke and Devrient, Leipzig) 346

LIST OF TEXT ILLUSTRATIONS

"4-*5- 6.

FIGURE I. Athena on Gem (Cades, i. H, 17)
2. Athena from an Attic vase (E-lite drain.
i. 80)
"3. Profile of the Bologna head. (From a
photograph after the original)
Athena on a relief from the Akropolis
Statuette of Athena from the Akropolis.
(From a drawing)
'Torso Medici.' (Ecole des Beaux-Arts,
Paris) . .
*7. Niobid from the disc in the British Museum.
(From a drawing)
*8. Statue of Apollo in Museo delle Terme
(Rome) 
*9. Head of Apollo in Museo delle Terme .
10. Bronze Apollo of the Mantuan type from
Pompeii, Naples
II. Terminal bust in the Capitoline Museum :
a, from the original ; b, from the cast
12. Head in Palazzo Barberini, Rome .
13. Terminal bust in the collection at Broad-
lands
Head in the Collection Barracco (Rome).
(By permission of Messrs. Bruckmann,
Munich) . .
Head in the Hermitage . .
Terminal bust of Athena from Hercula-
neum (Naples)
*I7. The ' Anakreon Borghese ' (Ny Carlsberg
Glyptothek, Copenhagen) .
"18. Head with winged fillet, Museo Torlonia,
Rome. (From a drawing)
Head in the Museo Chiaramonti (Vatican)
Double terminal bust in Madrid .
Profiles of the double terminal bust
22. Cameo in Berlin
23. StatueformerlyinPal. Cepparelli (Florence)
"24. Statuette of Aphrodite (Berlin)
25. Two heads of Athena from casts in Dresden:
a, from a lost original; b, head of Athena
Farnese (Naples)
26. Athena Farnese (Naples)
27. Hope Athena in the collection at Deep
dene, Surrey. (From Spec, of Anc.
Sculpture)
28. Profile of the head of Athena (a, 25)
29. Athena in the Villa Albani, Rome

PAGE 6

14

H-

16.

19.
20.21.

28
44 5°
5152
545<5
58

59 59

63
646567
6869 70
7i

7274

75
7779

FIGURE page
30. Head of Athena Albani. (From the
cast) . . . . . 80
Head in Munich . . 81
Head of Herakles (Berlin) . 83
Torso of Herakles, Louvre. (From a pho
tograph by A. Giraudon) . . 85
Head of a goddess (Berlin) . . 86
' Ceres ' in the Rotonda of the Vatican . 87
Apollo Barberini (Munich) 87
Athena in Capitoline Museum 87
38. Conjectural restoration, showing the torso
in Cassel combined with the head of the
Hephaistos Chiaramonti 88
Head in Brescia. (By permission of Messrs.
Bruckmann, Munich) . . 91
Head of Ares, Louvre. (From a photo
graph by A. Giraudon) 93
Ares in the Pal. Borghese (Rome) . . 94
The Dioscuri of Monte Cavallo. (From the
cast) . 98
Head in the Jacobsen collection at Copen
hagen. (By permission of Messrs.
Bruckmann, Munich) 101
Head in the Louvre . 103
'Jupiter de Versailles ' (Louvre) 104
Terminal bust of Perikles in the British
Museum. (From the original) .... 119
Head of a strategos (Berlin) . . . 121
48. White-faced Lekythos (Bibl. Nat., Paris) . 124
49. Gem in Berlin (slightly enlarged) ... 124
50. ' Gladiatore Farnese ' in Naples. (The
restorations are omitted) .... 125
Alkibiades in the Vatican. ( Attempt at a
reconstruction ; old restorations omitted) 127
Amazon in Villa Doria-Pamfili. (Wrongly
restored as an Artemis) ... 129
Amazon of the Capitoline type. (Re
stored) . 132
Amazon head of the Capitoline type,
wrongly placed on the Mattei statue in
the Vatican. (From the cast) . 133
Head of Amazon in Lansdowne House.
(From the original) 135
*56. Amazon type. (Attempt at a restoration) 138
57. Bronze terminal bust of an Amazon from
Herculaneum (Naples). (By permis
sion of Messrs. Bruckmann, Munich) . . 139

31-
32.33-
34-
*35-
*36."37-

39-
40.41.42-43-

44-45-
"46. 47-

Si-

¦52.

53-

54-

*55-

LIST OF TEXT ILLUSTRATIONS

FIGURE 58- 59-60.

61. 62.

63- 64.
65.66. 67.68.69. 70. 7i-72. 73- 74- 75-
76. 77-78.79-
80.Si. 82.S3-

85.86.87.

90. 91-92.93-94-
95- 96. 97.

99.
100.
101.

' Pallas de Velletri ' (Paris)
Head of Athena from Velletri
Diomede in Munich. (From a cast with
the modern restorations omitted) .
Head of the Munich Diomede. (From the
cast)
Cast at Dresden of a replica of the Dio
mede. (Original presumably in Eng
land) . . • •
The Medusa Rondanini (Munich) .
Profile of an athlete (Petworth Coll.)
Head of an athlete (Petworth Coll.)
Head of a hero (Palazzo Riccardi, Florence) 166
Replica of the Riccardi head (Berlin)
Head of Diskobolos (Catajo) .
Head of Diskobolos (Berlin) . •
Head from Perinthos (Dresden) . .
Head in the collection at Ince Blundell
Hall (Lancashire) . . ¦ •
Head in Brescia. (By permission of Messrs.
Bruckmann, Munich)
Portrait-head in the Villa Albani. ( From
the cast) . . • •
Portrait-head in the Hermitage. (From
the original)
Herakles in the British Museum
Mercury in the Vatican
Statue restored as Neptune (Vatican) .
Head of a god (Berlin) . •
Statue restored as Asklepios (Hermitage)
Apollo of the ' Cassel type ' (Louvre)
Head of Apollo
Apollo head of ' Cassel type ' in Barracco
Collection (Rome) . .
Head of Perseus (Rome)
Female head in Giardino Boboli (Florence).
(By permission of Messrs. Bruckmann,
Munich) .
Head in Museo Chiaramonti
Head in the British Museum
Asklepios in the Uffizi (Florence) . .
Asklepios and Hygieia in Palazzo Barberini
(Rome) .... . •
Terminal bust in the British Museum .
Statue in Munich
Head of statue in Munich .
Replica of the head of the statue by Ste
phanos . .
Bronze statuette in British Museum
Mercury, Coll. Oppermann (Bibl. Nat.,
Paris)
Head of Herakles. From Herculaneum
(Naples)
Head of Herakles. From the collection
at Broadlands ( Hampshire) .
Polykleitan statue in Coll. Barracco(Rome).
(By permission of Messrs. Bruckmann)
Diadumenos in Madrid
Statue of a boxer (Cassel) .
Greek gem .
Carnelian in St. Petersburg

AGE

FIGURE

142

102.

143

103.

147

IO4.
150
I05.
106.
151
107.
157
108.
162
IO9.
163
I IO.
166
I I I.
167
112.
169
170
113-
171
II4.
115.
172
116.
117.
174
Il8.
"119.
176
120.
121.
177
122.
179
123-
183
I24.
.85
I25.
187
126.
189
127.
192
128.
193
I29.
195
¦"130.
198
131-
132.
133-
203
134-
205
135-
206
136.
207
137-
138.
208
211
139-
213
140.
216
•141.
""'142
217
H3-
232
144.
145-
233
146.
234
H7-
148.
235
149.
237
150.
241
151.
246
152.
248
153-
24S
*I54
250 251252253 256
Statue of a victorious boy placing the
wreath on his head. (Restored) .
Head of boy (in possession of Sir Edgar
Vincent) . ....
Head of a boy (Hermitage) . . .
Back of the Westmacott athlete (British
Museum) . ...
Figure from Ildefonso group (Madrid) .
Athlete in the collection at Petworth House 258
Apoxyomenos on a gem . . ... 261
Apoxyomenos on a gem (Hermitage) 262
Basis of the statue of Pythokles in Olympia 263
Athlete in the Braccio Nuovo (Vatican) 264
Head of the boy's statue in Dresden.
(From the cast) .... . .
Statue of a youth (Hermitage)
Youthful Pan (Leyden). . . .
The ' Narkissos ' (Berlin)
Bronze statuette (Bibl. Nat., Paris)
Gem (Cades, ii. D, 16)
Basis of Zenokles in Olympia
Bronze statuette in the Louvre (back)
Athlete in Galleria delle Statue (Vatican)
Bronze head from Herculaneum (Naples)
Head of the Idolino (Mus. Naz., Florence) 285
Basis of Aristion in Olympia . 288
Hermes in Lansdowne House
Herakles in Lansdowne House
Heroic statue (British Museum)
Statuette of Zeus (British Museum)
Bronze statuette of Asklepios (Carlsruhe)
Hermes from the Palatine (Museo delle
Terme)
Statue of Athena in the Uffizi
Satyr in Dresden .
Head of Satyr in Dresden .
Eros from the Palatine (Louvre)
Eros in Naples
Head of Eros of Centocelle (Vatican)
' Venus d' Aries ' (Louvre) . .
The 'Townley Venus' (Brit. Mus.) .
Statue in Louvre. (From Clarac, Mus.
deSc. PI. 341) . . 323
Artemis in Dresden . 324
Head of Artemis (Dresden) 325
Statuette of Artemis, from Kition in Cyprus 327
Attic statuette vase . . . 333
Head in Pal. Pitti (Florence) . 335
Hermes in the Uffizi 339
Herakles in Villa Albani . . 340
Herakles with Telephos (Mus. Chiara
monti) .... 341
Head of Herakles (Chiaramonti) . 342
Profile of Aphrodite in the collection of
Lord Leconfield . . 344
Carnelian in the British Museum . 350
Dionysos from Tivoli (Museo delle Terme) 351
Bronze statue of Apollo (Brit. Mus. ) 352
Statue in Dresden . ¦ 352
Apollo ('Adonis'), Vatican 355
. Paris in Lansdowne House . . . . 358
267 269 271273276
277 279280281 284
289297298299300 3°i3063103" 3i3
3M 315
320321
LIST OF TEXT ILLUSTRATIONS

FIGURE *I55- Torso of Aphrodite (Naples)
"156. 'The Faun of Winckelmann' (Munich).
*I57- Athena Giustiniani (Braccio Nuovo,
Vatican) ... . .
158. Venus of Milo (with plinth unrestored)
159. Drawing by Debay showing the inscribed
block adjusted to the plinth ...
160. Ground-plan and projection of the plinth,
with restorations indicated
161. Side view of the extant plinth . .
162. Side view of the plinth. (Restoration
indicated by dotted lines)
*i63. Proposed restoration of the Venus of Milo
164. Aphrodite on a gem (Berlin) .
165. Bronze Aphrodite (Dresden)
166. Statue in Dresden ... . .
167. Bronze coin of the island of Melos (Berlin)
168. Relief from a column in Melos .
169. Aphrodite with the apple. Terra-cotta
from Myrina (Berlin)
170. Venus of Capua (Naples)
171. Head of Venus of Capua
172. Head in Palazzo Caetani (Rome) .
"173. Head in the Capitol. (From the cast)

AGE 358

FIGURE *i74-

360

175-

362

176.

37°

177.
37i
'178.
372
*i79-
373
»i8o.
374
181.
380
380
381
182.
382
382
"183.
383
383
"184.
385
389
39°
185.
393
186.
Head from Tralles in Smyrna. (From
the cast at Bonn)
Statue in Pal. Valenlini (Rome)
Statue in Pal. Valentini. (Restorations
omitted)
Head of Apollo (Brit. Mus.)
The four great temples of Athena on
the Akropolis . ...
Kalathiskos dancers on either side of Pal
ladium. Terra-cotta plaque (Berlin)
Archaistic Artemis from Gabii (Munich) .
Left-hand corner of the south frieze of the
Temple of Athena Nike. (From Lebas,
Voyage Archiol.)
Portion of the north frieze of the Temple
of Nike. (From Lebas, Voyage Arc he'd.)
Western pediment of the Parthenon, show
ing the traces on the floor of the pedi
ment
Eastern pediment of the Parthenon, show
ing the traces on the floor of the pedi
ment
Attic seal
Two bronze coins of Krannon
397 398
399
4114i7
440
446
447
454
464 469469
INITIAL AND TAIL-PIECES.
Profile of 'Venus d'Arles' (Louvre) . title-page
Bronze Pan (Bibl. Nat. , Paris)
Profile of Skopasian Meleager (Villa Medici) . .
Eros on Parthenon frieze
Marsyas in Lateran . . .
Doryphoros in Naples .
3
"5
223
Satyr in Capitol ... . .
Head of Venus of Milo . . .
Apollo of Belvedere
Ganymede in Vatican
Reliefs from Ludovisi throne (Rome) .
PAGE 295 367
4054i5
PHEIDIAS

PHEIDIAS

O'

the works of those sculptors whom antiquity
esteemed as its greatest, one masterpiece, the
Hermes of Praxiteles, has come down to our
times in the undoubted original, while others, such as the
Doryphoros of Polykleitos and the Apoxyomenos of Lysip-
pos, are known to us in good and faithful copies, executed
probably on the scale of the originals. But Pheidias, most
famous of all who wrought the images of the gods, is
represented so far neither by any ascertained original nor
by any efficient copy. For of only two of his works,
the Olympian Zeus and the Athena Parthenos, do we
possess reproductions acknowledged to be such : the Zeus
is reproduced chiefly on coins, and the Athena in statues
and statuettes, so much reduced in size that they are rather
abstracts or resume's than real transcripts ; they differ
totally from the faithful copies alluded to above. Though
sufficient to give a general notion of the composition,
of the arrangement and fall of the drapery, they are inadequate for all finer dis
tinctions. Above all they give no exact idea of the head, which, as being the seat of
intellectual life, is naturally of the highest interest. The various replicas exhibit
differences so marked as to convey the impression that not one of them is accurate.
And this is very natural, for the head of a colossal statue in ivory and gold must
have offered unusual difficulties to the copyist. Those who had access to the
original itself could only make drawings or small models, and in so doing each
artist would follow his own style. Other and freer imitations of widely varying
dimensions were derived from these sketches. It is small wonder, then, that the
finer modelling of the head of the Parthenos has been lost to us in spite of its
numberless reproductions.
Yet even if we possessed an exact copy of the head of the Parthenos or of the
Zeus in the original size, I do not believe that we should thus gain a complete conception
of the best that Pheidias could achieve. An artist cannot show the finest and most
spiritual qualities of his treatment of form in a colossal head. Besides, the complicated
ivory and gold technique must have placed many obstacles in the way of free artistic
conception. An artist was much less trammelled in executing the clay model for a
bronze statue, and even Pheidias must have found in this technique a purer and more
exquisite medium for his genius.
Bronze statues by famous artists were usually, when of normal dimensions, copied
in the original size. The copyists probably made use of casts,1 which was of course
1 The passage in Lucian (/up. Trag. 33) concerning the Hermes Agoraios in Athens shows that it was quite
usual for artists to take casts of famous works of art.

U

4 PHEIDIAS
impossible in the case of gold and ivory images. Yet even so copyists allowed them
selves a freedom which occasionally amounted to almost complete remodelling of
details. Nevertheless, copies exist which can be proved to be perfectly accurate. Such
a copy of a bronze work by Pheidias would be the first step towards an exact appre
ciation of his artistic personality. I. Discovery of the Lemnian Athena.
It is this need — this deficiency in our knowledge — which, I believe, I have been
able to fill. The two statues reproduced on Plates I. and II. can be shown to be faith
ful replicas of a bronze work by Pheidias — a work which ancient connoisseurship
preferred to all others by the artist, which roused the enthusiasm of the subtle
critic Lucian, and which even the matter-of-fact Pausanias admits to be the best
worth seeing of all the creations of Pheidias. This work is the Lemnian Athena.
The two statues belong to the Dresden Museum, and have long been known :
one of them had a head foreign to it, while the head of the other, though genuine,
was disguised by inaccurate restoration. The statue given on PL I. is published in
Becker's Augusteum, i. PI. 14, and in Clarac's Muse'e de Sculpt. PI. 464, 868. The
head is antique, but has been broken and put on again. The whole upper part of
the head, from the fillet, is restored ; an ugly modern helmet was formerly placed upon
it. In 1894 Becker pronounced that it was impossible to decide with certainty whether
the head belonged to the statue or not.1 On the other hand, L. Schorn (1822),2 while
acknowledging that the head had been joined on again and the face worked over,
added that there was nothing to prove that it might not have belonged to the statue.
Hettner, in his catalogue of the Dresden antiques,3 declared that the correspondence
of the marble, and the fact that the turn of the head fitted the remaining portion of neck,
showed the head to be genuine. More recently it was recognized by Flasch that the
head was a replica of the beautiful Bologna head which Conze 4 had published as that of
a young man, while Flasch 5 had interpreted it as an Amazon. It was accordingly
taken for granted that the head did not belong to the statue,6 and therefore, when Treu
began his admirable and useful task of freeing the Dresden statues from their modern
restorations, he removed the head from the statue, set it up separately, and restored it
from the Bologna replica — i.e. he took away the helmet, and supplied the place of the
missing upper part of the head by a cast taken from the Bologna head.
The statue and the head were thus separated when I saw them in 189 1. At first
it did not occur to me that they might belong together. It was only in the course of
a protracted inquiry into the type of head appropriate to this statue of Athena that,
to my own great astonishment, I came to the conclusion that the head which had
been removed actually did belong to the statue, and that consequently the Bologna
head also represented Athena, and came from another replica of the same statue.
The portion of neck still attached to the torso and turned vigorously to the right, as
well as the shape of the nude part of the chest, answered down to the minutest details
to the corresponding parts of the Bologna bust. Further, head and torso are of the
same marble. Probability became certainty when Treu, at my request, placed the
1 Augusteum, i. p. 95. 2 In Bbtticher's Amalthea, ii. 206 sqq.
Bildwerke d. Koniglichen Antikensammlung, 4th ed. No. 69.
* Beitrdge zur Geschichte d. Griechischen Plastik, Taf. I. p. 1. Cf. Friederichs-Wolters, Gipsabgiisse, 519.
" Bull. d. Inst. 1872, 66. Brizio declared the head to be modern (ibid. p. 65 ; cf. Heydemann, Mittheilungen
aus den Antikensammlungen, in Ober- u. Mittelitaliens, p. 60, 206), but was immediately refuted by Flasch.
Cf. Puchstein in Jahrb. d. Inst. 1890, p. 96, note 36.

DISCOVERY OF THE LEMNIAN ATHENA 5
head upon the torso ; the two fitted together fracture for fracture, of course not in
front where the edges are broken off, but in the core of the neck.
PL I. shows the statue in its present state. The missing parts in front of the
neck, the nose, mouth, and upper part of the head are restored in plaster. The
face, as Schorn correctly noticed, has been worked over, and has lost much of its beauty
in the process ; but the hair, as far as it is preserved in the marble — i.e. the portion
under the fillet — is uninjured, and corresponds exactly to the hair of the Bologna
head, except for a few variations in the curls, which show that the copyist in this
case was rather less careful. Yet the correspondence between these two copies
is more exact than is usual in antiquity, and goes to prove that each is a fairly
accurate rendering of a common original.
The other statue (PL II.) is published by Clarac, PL 464, 866.1 The head was
not made in one piece with the body, as in the preceding example, but was worked
separately and inserted, with the nude part of the breast, into the draped torso.
The head has been restored by using up an antique fragment of a replica of the
head of the Farnese Athena in Naples,2 which, placed on a hideous thick neck,
looked curious enough when set upon the statue. This restoration has now been
removed, and a cast of the Bologna bust has been inserted into the empty space cut
out in the torso. PL II. represents the statue thus restored. The Bologna bust
fitted into the hollowed torso, as exactly as if it had been made for it, hardly a
millimetre of alteration being necessary. This bust is wonderfully well preserved,3
the only injury it has suffered being the loss of the eyes, which were inserted
separately. The exact correspondence of the hair with the Dresden head, and indeed
the remarkably careful execution of the whole work, show the accuracy of the copy.
Rather less care has been bestowed upon the hair above and behind the right ear,
the head being so much turned that this part comes at the back.
The right breast and right shoulder of this statue were lost, and have now been
restored by a piece cast from the other figure ; the left arm-stump, likewise, has been
added, in accordance with the other figure where it is antique. The bodies, except for
a few unimportant variations,* are exact replicas. On the whole, however, the statue
reproduced in PL II. deserves the preference for its fresher and more refined workman
ship, though it is evident from the close resemblance of the body to that of PL II. and
of the head to the Bologna head that the statue on PL I. is also a very careful copy.
A third replica in the Museum of Cassel contributes nothing to our know
ledge of the original, in spite of its careful workmanship. The head is missing
(it is replaced by a replica of the Athena Giustiniani), and the body is not
a copy, but a complete transformation of the original of the two Dresden statues.
1 Hettner, Bildw. No. 72. A cast of it is in Berlin, Fried. -Wolters, Gipsabg. 478. Published in Sandrart,
Teutsche Academie, ii. (1679), 2, ee, as being in the Palace Caesii in Rome.
2 Published in Becker's Augusteum, i. cap. 15. B. Graf (Aus der Anomia, p. 65) has already noticed
that it belongs to a replica of the Athena Farnese.
3 It has been supposed that the face was worked over in modern times (Fried. -Wolters, Gipsabg. 519) (cf.
Heydemann, Mitt, aus d. Antikens. Ober- u. Mittelitaliens, p. 50). After repeated examination of the original
I am able to state that this is not the case, though it is true that in certain parts acid has been used in order to
remove incrustation (this was noticed by Brizio, Bull. d. I. 1872, 65), but without injuring the face in the least.
' 4 Slight differences are to be observed in the hair and tongue of the Gorgoneion. The scales of the aegis
also differ slightly : in the one statue (PI. II.) they are lighter, flatter, and all turned downwards, while in the
other (PI. I.) they are harder, more plastic, and do not all follow the same direction. The first rendering (PI. II.)
is the more severe, and certainly the most correct. The ends of the snakes knotted in front also vary somewhat
in arrangement in the two statues. [PI. II. * shows the left side of the statue ; it is taken from the cast exhibited
in the Pheidian room of the Dresden Cast Museum ; the body is that of statue PI. I., on which has been placed a
cast of the Bologna head. — E. S.]

6 PHEIDIAS
The artist's aim evidently was to get rid of the severe simplicity of the folds, and
to substitute for them a rich mass of petty details more suited to later taste.1
This copyist shows himself unable to appreciate the greatness of the older manner ;
he tried to correct and to improve, and only succeeded in becoming feeble, futile,
and artificial.
There is still another copy of the head alone (probably part of a fifth replica);
in the Vatican. The head is set on a female statue, to which it does not belong ;
it was once a good copy, but it is now so broken and restored as to be almost
unrecognizable.2 The position of the arms in the original can be made out from the Dresden
statue (PL I.) The left upper arm is raised horizontally. The goddess must have
held with her left hand a lance that rested on the ground. The right upper arm
is lowered, but it is plain that the right forearm must have been extended somewhat
forwards and sideways, so as to bring the right hand into the direction towards
which the goddess is looking.
It would be impossible to say what this hand held, were it not for another copy
of the original, which is fortunately preserved on an engraved gem (Fig. i), s and
which confirms in the most satisfactory manner our reconstruction of
the Dresden statues. It is quite evident that both statues and gem
are derived from the same original. The gem-cutter has copied, as
faithfully as was possible within so small a space, the bare head
with the short knotted-up hair, leaving the ear free, the fillet,
fig. i.— Athena the neck, the opening of the drapery on the breast, even the folds
i°nH,e™{. a es' falling over the left breast, the transverse aegis, and the raised left
upper arm. He chose to represent the left side of the statue, so that
the breast appears somewhat foreshortened, and the head in profile (cf. PL II.)
In the field in front of the left shoulder is a helmet. Such objects on gems are
never mere ' symbols ' as they are on coins, therefore I think it quite certain that
the artist, who had not room in the picture for the right arm, wished at least to
indicate that the goddess bore a helmet in her right hand. This trait completes
our conception of the original statue.
The composition which we have thus won back had been brought into connexion
with Pheidias even before its design was thoroughly clear. L. Schorn 4 was the first
to associate the statue with the Athena Parthenos of Pheidias, of which at that time
no copies were known. This was no vague hypothesis, but the result of a careful and
penetrating appreciation of the ' grand style ' 5 of the Dresden statue. He describes
the simple treatment and rich effect of the drapery ; the transverse aegis seems to him
to indicate peaceful possession, the tranquil watch over the beloved city. And of the
head he says that it has 'the features and solemn expression appropriate to the
maiden goddess,' and that the beautiful turn towards the right completes the
' exalted image, which captivates the beholder as much by its godlike majesty as by
1 Cf. Schreiber, Athena Parthenos, p. 583 ; my article in Roscher's Lexikon, i. 699 ; and Puchstein's remarks
in. Jahrb. d. Inst. 1890, 93. Wolters's view (Gipsabg. 477, 478) is obviously incorrect.
2 Vatican, Gall, delle Statue, 400.
3 According to Raspe, No. 1651, a nicolo, 'a beautiful engraving'; the owner is not named; badly
reproduced on Plate 25 of the same work. A cast is in the Cades Collection, i. H, 17, from which our illustration
is taken. To the left is the inscription HEIOT, badly written and certainly modern, which made me formerly
doubt the genuineness of the gem (Jahrb. d. Inst. iv. 1889, p. 71). But the inscription has nothing whatever
to do with the stone, which is beyond suspicion, and is a beautiful and carefully executed work of about the time
of Augustus. 4 Botticher's Amallhea, ii. (1822), 206 sqq.
5 W. G. Becker (Augusteum, i. p. 93 sqq.) has likewise well appreciated this style.

DISCOVERY OF THE LEMNIAN ATHENA 7
the freedom and naturalness of the pose.' What judgment would he not have passed
could he have seen the head freed from the ugly modern helmet and the statue
completed by the finer head from Bologna ! Recently the extraordinary likeness of
the torso to the Parthenos caused Puchstein1 to reckon it as distinctly Pheidian.
The head belonging to it was unknown to him ; but since an Athena from Pergamon
(see p. 27) which he showed to be dependent on the Dresden type was bareheaded,
he concluded that the Dresden statues also had worn no helmet, and pronounced
it not impossible that we have in them good copies of the Lemnian Athena by
Pheidias.2 The discovery of the head gives to this hypothesis a certainty which is only less than
absolute in a case where there is neither the evidence of provenance nor of inscription
to shut out the last doubts. The premises which lead to the conclusion that our
statues are copies of the Lemnia consist in the following observations. The original
must have been a famous work of classical antiquity, witness the marble replicas
which have been enumerated and the small accurate copy on the engraved gem.
Again, the body of the statue comes nearest in style, of any known work, to the
Parthenos of Pheidias, and possesses all its most characteristic and personal qualities :
this has already been pointed out by Puchstein, and we shall presently prove it more
in detail. If any work is to be taken as Pheidian, surely this must be. Now the
famous Lemnian Athena of Pheidias also wore no helmet, and was moreover the only
famous bareheaded Athena known to literary tradition. And what is said of this
work 3 corresponds in the most striking manner to our marbles, in fact has been made
clear by them. The Lemnia was noted for exceptional charm ; Lucian describes her as
undoubtedly the masterpiece of Pheidias, and borrows for his ideal beauty the outline of
her face, the delicacy of her cheeks, and the fine proportions of her nose {Imagines, 6).
Applied to the Bologna head these words are no empty phrases (indeed, so far as we can
verify, Lucian is never a mere phrase-maker) : on the contrary, they emphasize peculiar
merits which must strike any one who looks at this head and which distinguish it
from other works of antiquity. The eximia pulchritudo of the Lemnia of Pheidias is
possessed in a very high degree by the head which belongs to the statue acknowledged
as Pheidian. Finally, the Lemnia was a work in bronze, and, as nothing is said to the contrary,
it was probably life-size. Now it is from a life-size figure that our statues derive, for,
since they accurately reproduce their original in other respects, they presumably also
reproduce its scale. It seems likely from the great distance at which the arms are held
out from the body, and from the treatment of the hair, that this original was in bronze,
while the extraordinary wealth of motive in the separate locks and the subtlety of the
execution point emphatically to the same conclusion.4 True, artists before Pheidias
employed for marble the same technique that we consider characteristic of bronze — as
for instance in the Aegina marbles and in the boy's head from the Akropolis 5 — i.e.
they imitated bronze technique in an inferior material ; but within the circle of
Pheidias we can no longer suppose that this was the case, although the complete
separation between marble and bronze technique, which manifested itself especially in
1 Jahrb. d. Inst. 1890, p. 93 sqq., 96, note 36.
2 Michaelis, Athen. Mitt. 1876, p. 287, had previously tried to refer this type to the Athene Hygieia of
Pyrrhos ; his theory was however refuted by the discovery of the footmarks of the statue on the actual basis. Cf.
Ath. Mitth. 1891, 163, Taf. 6.
3 For the most recent discussion of the passages see Studniczka, Vermutungen 2. Kunstgeschichte, p. 1 sqq. , and
Weizsaecker, Neue Jahrb. f. Philologie, Bd. 133 (1886), p. 14 seq.
1 Cf. Flasch in Bull. d. Inst. 1872, 66. ° 'E^/tepls apxaioA. 1888, Taf. 3.

8 PHEIDIAS
the Lair, was not effected till the fourth century. As regards the treatment of hair,
we may compare the head of the Lemnia with the fragment of the head of Nemesis
from Rhamnus,1 by Agorakritos, the closest pupil of Pheidias. The enormous contrast
between them is certainly due to differences not only of date but of material ; the pupil
working in marble having contented himself with the broadest and most general repro
duction of the principal outlines. Bronze technique involves a previous modelling in
clay and chisel work after the casting ; all this brings out the form in a sharp and
detailed manner, and accounts for the totally different treatment of our Athena head.
Lastly, the hollow eyes afford another proof, as they occur principally in copies from
bronze works.
We are now justified, I think, in claiming to possess exact copies after a bronze
work by Pheidias, and to have thereby gained for the first time a full conception of
this artist's achievements in the round. All through the fifth century, apparently,
bronze casting was the method most in esteem ; artists found in it a vehicle for the
finest execution of which their mastery was capable, whereas ivory and gold technique
imposed on them an over rigid restraint, and the secrets of marble work, as Praxiteles
understood them, were as yet undiscovered.

II. Site of the Lemnia on the Akropolis. — Date and Dedication.
The Lemnian Athena of Pheidias was named diro rac avaOivTcov. As votive
gifts of foreign states do not occur on the Akropolis of Athens, at least in the fifth
century, the dedicators were not the Lemnians, but, as is universally agreed, the Attic
kleruchs on Lemnos. As to the occasion of the dedication, Loschcke has made the
plausible suggestion 2 that it was set up by the kleruchs on their departure in order to
secure for themselves, even when far away, the potent protection of the city goddess.
Loschcke cites two votive gifts on the Akropolis with inscriptions which prove
that they were set up by departing kleruchs ; 3 since they belong approximately to
the same period as the statue by Pheidias, they afford a good analogy. The objection
that such a votive gift would not be offered before some success had been attained
is without weight, for according to the ideas of the time ' the permanent holy offering
was intended to remind the god constantly of the pious intentions of the giver, and
of the consequent obligation to grant him protection and help in return for his
gift' i Just as at the departure of the kleruchs a great sacrifice, v-rrep t% diroiKiw},
was offered (as we know from the inscription concerning the kleruchs of Brea),5 so
it appears to have been the custom for the colonists to leave behind some more
lasting symbol of sacrifice in the form of a votive gift. Loschcke's suggestion is,
I think, confirmed by the fact that the two inscriptions were found in the very
part of the Citadel where, according to the periegesis of Pausanias, the Lemnia
must have stood. The inscriptions were found near the Propylaia ; I remember
seeing one of them at the eastern end of the northern half of the principal structure.
The Lemnia stood on the same part of the Akropolis. If these three monuments,
then, have a local connexion, we have the more reason for concluding that they
1 Cf. Rossbach in Athen. Mitt. 1890, p. 64.
' Tod des Phidias in Histor. Unters. dedicated to A. Schafer (Bonn, 1882), p. 43.
C. I. A. 1. 339, Tijs airoi[iclas] Trjs is 'Ep[erpiay ; 340, 'EttoIkwv is UorelSaiav.
4 Emil Reisch, Griechische Weihgeschenke, p. 5.
5 C. I. A. i. 31 ; Dittenberger, Sylloge, 12.

II.
Statue of Athena, with Cast of Bologna Head.

LEMNIAN ATHENA 9
were presented on similar occasions. Plainly the statues dedicated by departing
kleruchs, of which the Lemnia was one, all stood together.
This place needs however to be more exactly defined. Pausanias, whose
evidence we take first, so arranged his periegesis of the Akropolis that after coming
through the Propylaia he went to the right to view the south side of the Citadel and
then by way of the Erechtheion along the north side back to the gates. He begins his
account by describing the Propylaia and all that was to be seen in connexion with
them, and does not mention them again at the end. We only find out that he has left
the Akropolis by his reference to the Pelargikon below the Akropolis and to the
grotto of Apollo. Of the four monuments which close his description of the Citadel,
one, the so-called Promachos of Pheidias, is fixed, if not actually by remains of the
basis,1 yet at least by well-known coins,2 as having stood between the Erechtheion
and the Propylaia, just where we should have placed it in reliance on the account
of Pausanias. Here too must have stood in the time of Pausanias the four-horsed
chariot which he mentions together with the Promachos ; and close by, nearer the
gates, the Perikles and the Lemnia ; Pausanias passed them on his way from the
Promachos to the gate.3 They cannot have been within the Propylaia or in front
of them, for the gates are not mentioned again, and a complete description of them,
including their north and south wings, had been given at the beginning. The
statues more probably stood to the north of the principal road which led from
the Propylaia along the top of the Citadel, and would therefore correspond to the
other statues, south of the same road, which are named at the beginning of the
actual periegesis of the Citadel. These other statues are not, it is true, clearly
marked off from the description of the Propylaia, on account of the eloquent
transition which Pausanias devised from the wisdom of Sokrates to the wisdom of
Peisistratos and Hippias, and so on to the mistress of the murderer of Hippias.
But the exact place of the Athena Hygieia, one of a group of works expressly
mentioned as standing near together,4 is definitely known through the basis found
in situ on the Akropolis outside the Propylaia on the east. Hence we may con
clude that the other works associated with it — i.e. the Lioness, the Aphrodite of
Kalamis, the Diitrephes, and the ' Hygieia ' — were also outside. From the circum
stance that Pausanias mentions the Athena Hygieia last among them, we may
presume that this statue, standing in front of the last pillar to the right, would not
be noticed until the others had been passed. To suppose that these statues were
in . the porch itself, and to reason from this that the other statues — or at least the
Perikles and the Lemnia — stood opposite them in the northern part of the porch,5 is
entirely mistaken. Neither the one series of statues nor the other was within the
porch. No traces whatever of any such monuments have been found on the floor of
the Propylaia. The two bases belonging to the statues were found not within, but

1 Cf. Loschcke, loc. cit. p. 45, note. Lolling in Iwan Midler's Handbuch d. Alterthumswissenschaft, vol. iii.
p. 352. - Imhoof-Blumer and Gardner, Num. Comm. on Pausanias, PI. Z, iii. — vii.
3 The assumption that Pausanias is here, contrary to his usual custom, enumerating works of art without
reference to their locality (Curtius, Stadtgeschichte, p. 300) is entirely without foundation. Cf. Michaelis,
Athen. Mitt. ii. 95.
4 After the express mention of the Hermes Propylaios and the Charites as xa-rh. ri/v iaohov avriiv tJSt] tt)v is
axpinroMv, follows the mention of the lioness (the transition is effected by means of the wisdom), no definite place
being given ; ttapk avTi)v stands the Aphrodite ; irX-t\aiov the Diitrephes, and again TrXnaiov to this the Hygieia
and the Athena Hygieia.
5 Cf. especially P. Weizsaecker, Neue Jahrbiicher f. Philologie (Fleckeisen), vol. 133 (1886), p. I sqq. C

IO PHEIDIAS
immediately without the Propylaia,1 in the locality where they once stood, though
they were no longer in situ. Besides, all the statues, the northern as well as the
southern,2 with the single exception of the Athena Hygieia,3 and possibly also of the
hitherto unidentified ' Hygieia,' are older than the Propylaia, and therefore, supposing
they were within the Propylaia, they must have been moved there from some other
position. No possible reason for such a transfer can be discovered, for not only were
all the statues in bronze and made for the open air, but the corridor of the Propylaia
was the most unsuitable place that could be conceived for the reception of votive
gifts. The narrow space was doubtless kept as free as possible and not crowded up
with monuments.4
The Lemnia of Pheidias, then, stood on the way from the Promachos and the
quadriga to the Propylaia ; close to it were placed other monuments set up by the
departing kleruchies, and here also stood the portrait of Perikles. It is quite evident
how appropriately all these works were placed together, for the kleruchies, so often
a godsend to the poorer citizens, were the work of this statesman.
The position which we have shown the Lemnia to have occupied explains the
motive of the statue very simply. It stood on the left of the road, and would
naturally face it. But as the gates were so near, it was natural that the figure should
look towards them. With a kindly glance which promised peace and prosperity, the
goddess followed the departing kleruchy, while the colonist returning to his native
land and coming up to worship would be met and welcomed by the same protecting
gaze.6 Plate II. gives the view of the statue as I imagine it to have appeared to
any one entering the Citadel.
According to Kirchhoff6 the kleruchy to Lemnos must be dated between
01. 82, 1 and 01. 83, 2, i.e. between 451 and 448 — 7. It was at this time therefore — in
the middle of the fifth century, roughly — that Pheidias made the statue. On the
basis was inscribed not only the dedication, but, as tradition expressly states, the
artist's signature.7 Thus it is an absolutely authentic work by Pheidias, and not
merely one that was ascribed to him, which survives in the beautiful Dresden copies.

III. Comparison between Lemnia and Parthenos.
The chronology which shows the Lemnia to have been a little older than the gold-
ivory Parthenos executed in B.C. 447 — 438 receives important confirmation from a
comparison of the style of the two figures. One external detail, the snake-border of
the aegis, is specially important in this connexion. The Lemnia follows an old
fashion taken over from the archaic period ; the rolled edge of the aegis is worked in
round scollops, each point terminating in the upper part of a snake. In the Parthenos
the system is quite different : here whole snakes form the trimming, the scollops of
The Kallias basis, C. I. A. i. 392, which supported the Aphrodite of Kalamis, and the Hermolykos basis,
C. I. A. i. 402, on which the Diitrephes probably stood (see infra, p. 122).
2 For Perikles and the Diitrephes cf. the section on Kresilas.
3 Wolters (Athen. Mitt. 1891, p. 153 sqq.) has pointed out that this statue is more recent than the Propylaia.
4 The much-disputed passage of Herodotos about the four-horsed chariot need cause no confusion. Hero-
dotos had the old Propylaia before his eyes, and it was only at the building of the new Propylaia that the four-
horsed chariot was set up where Pausanias saw it. Cf. Miller, Amer. Journ. of Arch. 1893, 504 seq.
The Athena Hygieia too, to judge by the footmarks, turned slightly towards those entering from the gates ;
she did not stand exactly in front of the pillar.
6 Abhandl. d. Berliner Akademie, 1875, 33.
7 Lucian, I/nag. 4 : r\v Ar/fiviav g Kal itrtypiipai Tofoo/ia *«5fes rjjieoo-e.

LEMNIA AND PARTHENOS II
the older aegis being approximately imitated by the winding of the snakes, and by
making the tail of the one curl round the neck of the next.1 The statues on which
this new manner occurs are almost without exception later than the Parthenos, and it
is quite possible that the fashion was derived from this statue. The innovation
presupposes an artist of bold and original temper ; it is just what we should expect
of Pheidias, who, it seems, also invented the motive of the girdle composed of two
real snakes tied in a bow in front. We already find this girdle given to the Lemnia ;
it is directly derived from that archaic motive in which two snake-coils coming from
the long aegis at the back are twisted together in front, to keep not the drapery but
the aegis together.2
Another certain mark for the earlier date of the Lemnia lies in the rendering of
the free leg {Spielbein). The first thing that distinguishes it from the Parthenos is that
this foot, instead of being set considerably, farther back than that of the supporting
foot, is on a line with it. One result of this position is that the heel of the foot at
ease is not raised as in the Parthenos, but seems, from a side view, to be placed flat on
the ground, although in reality it presses less heavily than the ball of the foot. This
attitude of the Lemnia is one stage nearer the old severe position (in which the foot is
quite flat on the ground) than is the Parthenos. Again, in the Parthenos, the free
leg is much more emphasized. In the Lemnia the drapery conceals the outline of
the leg ; in the Parthenos it clings closer, revealing the full rounded contour of the
thigh, and in a lesser degree the shape of the leg below the knee. There is a sharper
distinction between body and drapery. The fold which falls down on the outer side
of the left shin of the Parthenos is missing in the Lemnia, where the treatment
on the outer side is excessively simple ; the garment is heavy, almost like leather,
and shrouds the leg without forming any larger folds. The treatment of the
Lemnia is evidently a survival of an older manner, in the Parthenos the transition
is already being made to the development, which ended in the Caryatids of the
Erechtheion, whose light clinging drapery reveals the bent leg, almost as if it
were nude.
A consideration of the manner in which the garment terminates at the bottom
leads to the same result. In the Parthenos it falls over the feet and touches the
ground.3 In the Lemnia the garment is shorter, and its lower hem touches neither the
ground nor the supporting foot, but falls free, thus forming a simple horizontal line
without the richness and variety shown in the corresponding part of the Parthenos.
In this particular the Parthenos has the form which occurs in all more recent statues,
while the Lemnia finds analogies only in earlier works.4
Although these details seem to indicate that the Lemnia is earlier than the
Parthenos, the difference in date must have been slight, because the two statues
correspond so closely in other respects. This conclusion is in harmony with the result
arrived at in another way — namely, that the Lemnia must have been produced immedi
ately before the first designs for the Parthenos (447).
Having thus gained a firm foothold whence to review the development of one of
the greatest artists of antiquity, we may now proceed to examine these two fixed
1 The border of snakes is best preserved in the replicas D. F. G. enumerated by Schreiber, Athena Parthenos.
- Cf. the copies of archaic Athenas in the Villa Albani and in Dresden, Clarac, 462 D, 842 B ; 460, 855 ;
for the bronze see A. S. Murray, History of Greek Sculpture, ii. PI. 10.
3 It does not seem likely that the folds of the drapery were bunched up over the right foot in the original as
they are in the copy of the Louvre (Minerve au collier). The Varvakion statuette, which only shows a slight
bend in the drapery over the instep, is probably correct.
4 Cf. e.g. the Athena of the Akropolis, 'Zip-nfj.. apx. 1887, Taf. 8, 7. See infra, p. 21 sqq.

12 PHEIDIAS
points, the Lemnia and the Parthenos, more in detail.1 To begin with the
dress : there is no essential distinction between them in the treatment of folds.
Too much stress must not be laid on the difference of material in the originals —
bronze in one case, gold and ivory in the other ; its only consequence would be to
prevent Pheidias from weakening the wooden core of the gold-ivory statue by bringing
deep furrows from opposite directions towards the same point, a caution unnecessary
in the casting of bronze. Otherwise in his conception of drapery Pheidias appears in
general to be independent of technique. And, given the precaution mentioned, there
was nothing to check a free and rich arrangement of folds in the drapery of the
Parthenos. Ivory and gold statues were in all probability executed in their complete
form in wood, and the draped parts would then be covered with thin plates of gold,
while for the nude parts flat slices of wood would be removed from the surface
(already completely modelled) and replaced by plates of ivory cut to the same size
and shape.2 The wood technique, while imposing the form, offers not the smallest
obstacle to the execution of deeply cut folds.
Other differences in the drapery of the two statues have nothing to do with
technique ; they may arise partly from the difference in date, but most of them exist
on internal grounds, and express the different character of the two statues ; the
Lemnia is as a whole simpler and more severe, the Parthenos richer and more
majestic. The drapery of the Lemnia gives an impression of magnificence by its
severe reality, by the large, almost coarse simplicity of its rendering ; there is
no trace of the artificial or the intentional ; the folds fall as they must, without
regard to symmetrical arrangement, and equally unsymmetrical is the aegis
thrown obliquely across the breast. The Parthenos is in festal attire, all is
arranged carefully, soberly, and of intention ; especially so are the folds above and below
the girdle. Here the goddess steps forward to meet her worshippers in her holiday
robe disposed with delicate precision, but the Lemnia is in her everyday dress. This
distinction must arise from the different conception which underlies the two statues.
In the temple the goddess appears in full majesty and splendour, as a queen
giving audience in her throne-room ; the figure in the open air has laid aside
outward marks of dignity, and meets her worshippers unadorned, on the footing
of everyday life, ready with friendly sympathy, advice, and aid.
This accounts for the stiffness and regularity of the folds in the diplois of the
Parthenos, in contrast to the more natural and fresher rendering of the drapery of
the Lemnia. The difference in the lower part of the body may be explained much
in the same way. In the Parthenos this lower portion consists of two distinct halves,
owing to the different treatment of the free leg, and of the supporting leg which is
covered by regular perpendicular folds resembling the flutings of a pillar. Pheidias
found this type ready to hand ; 3 he elaborated it, and in the period after the
Parthenos it came into great vogue. But what we see in the Lemnia was never a
' type ' either at the time it was produced or afterwards ; it is something quite
individual and unique. The artist has broken away from tradition and scheme, from
rule and symmetry ; in a happy hour he has gone straight to nature, and expressed
1 My remarks on the style of the Parthenos are based exclusively on the good copies, from which we must
exclude the torso from the Akropolis taken by Puchstein, Jahrb. d. Inst. v. (1890), p. 85, as the starting-point of
his inquiry. P. was further misled in his observations by an inaccurate cast of the Varvakion statuette, in which
the depth of the folds had been smoothed down, so that his section (loc. cit. p. 84) is not exact.
2 I owe this convincing argument to an experienced sculptor.
It occurs for instance on the Athena of the Olympian metope representing; the cleansing of the Augean
stables.

Printed by Giesecke ft Devrient, Leinzic (G«

LEMNIA AND PARTHENOS 13
what he felt in a large and forcible manner. This work keeps the spontaneity of the
first conception : it might be copied, but could not serve as a type or be further
developed, while statues in the type of the Parthenos might exist in numbers.
What we have learnt from a comparison of the bodies of the statues doubtless
applied in a far higher degree to the heads. But an exact comparison of these is
impossible, because, as we remarked before, none of the extant copies of the head
of the Parthenos are anything like as reliable and exact as the copies of the Lemnia.
They differ among themselves so completely in the finer details of form and expression
that they afford no fixed basis for argument. We can only judge of the superficial
effect and the general character.1 The head seems to have differed markedly from
that of the Lemnia. The Parthenos appeared (as was appropriate to the conception)
fully adorned, with necklace and earrings, with richly decorated helmet, and stately
ringlets falling on the breast in severe symmetry. She was standing upright and
looking straight out with joyful and victorious countenance. Carrying out this funda
mental idea, Pheidias gave the goddess a somewhat broad, full face ; it was to express
radiant health and well-nourished vigour, for the goddess was the personification
of the great and brilliant Attic Empire. This was no place for the severe and
thoughtful Athena, with the delicate cheeks. The deeper intellectual expression
must make way for the representation of joyous wealth and splendour.
All this is an admirable testimony to the fact that the great artists of antiquity
did not repeat the same type again and again, as modern critics would fain have us
believe, but, like artists of more recent times, indefatigably created new forms to
embody new ideas. IV. Analysis of the Lemnia.
It is now time to examine more closely the individual peculiarities of the
Lemnia, and in particular of the head, as being the grandest and most noteworthy
part of the statue. The Bologna replica is reproduced alone on PL III.
The two most striking qualities of our Athena are the absence of helmet and
the short hair combed up at the back. The goddess has wavy and curly hair parted
in the middle and confined with a fillet tied in a simple knot behind ; in front, above
the forehead, the full hair is combed back towards the ears in a manner which occurs
only on female heads, and which is alone sufficient to decide the sex. The hair
leaves the ears free, or only just touches their upper edge, and disappears under the
fillet ; behind the ears the wavy ends -are combed straight up, and likewise tucked
into the fillet.
Was this conception of Athena, with bare head and short gathered-up hair,
and carrying her helmet in her hand, an invention of Pheidias, or did he only
follow an existing model ? A glance at the monuments will show us that the
latter was the case.
If we look back into early archaic art we shall become aware that side by
1 For heads of the Parthenos see Loschcke, Festschrift d. Vereins v. Altert.-Fr. im Rheinlande z. 50 Jubil.
1 89 1, p. I sqq. ; the list omits, however, the best of the heads, that of the statue by Antiochos in the Mus. Boncom-
pagni (Helbig, Museums, 870). On the other hand, the head in Naples, published by B. Graf (Aus der Anomia,
p. 61) cannot, on account of its great variations from the ascertained copies, be reckoned among them. The
most important copies of the Parthenos head are : the Varvakion statuette, the Mine>~ve au collier, the statue
by Antiochos, the Cologne head, the Berlin head ; of smaller monuments, a marble head from the Akropolis
Museum (No. 647, unpublished) ; also the Aspasios gem and the corresponding head on Athenian bronze
coins (Imhoof-Blumer and Gardner, Num. Comm. PI. Y, xxiv. Specially good example in the collection
of U, Kbhler, Berlin), The St. Petersburg gold medallions are only free imitations,

H

PHEIDIAS

side with the ordinary warlike conception of Athena there was another equally
independent and original type, which presented the goddess entirely unarmed.1 This
conception, it is true, fell gradually into the background ; it did not disappear, but
it took a secondary place beside the dominant image of the armed warlike goddess,
the only one recognized by the national poetry. To represent the peaceful Athena
later archaic art invented a new motive : the goddess takes off her helmet and
holds it in her hand. She appears frequently in this attitude on the later black-
figured Attic vases, e.g. in peaceful colloquy
with Herakles or in procession to the unwar-
like competition before Paris on Ida.2 On the
earlier red-figured vases, too, this type some
times appears ; it seems to be a special
favourite on vases belonging to the transition
from the severe to the fine style or to the
earlier fine style, i.e. to the period circa B.C.
480 — 450. Here Athena appears holding her
helmet in her hand,3 sometimes alone, some
times in a peaceful assembly of divinities.
From these paintings we 'gather that just in
the period immediately preceding the Lemnia
of Pheidias the Athenians had a fancy for
representing their goddess with her helmet off
and armed only with the lance. Probably the
kleruchs who emigrated to Lemnos had ex
pressly ordered just such an image, for their
undertaking could be successful only if times
were peaceful, and therefore a goddess of peace
must be their patroness.
In the period after the Lemnia this type
outlived its popularity : it occurs rarely,4 and after a time disappears altogether.
By a kind of extension of this type the helmet is sometimes entirely left out,
i.e. it is not even carried in the hand. This modification in the representation of their
goddess also enjoyed its own popularity among the Athenians just in the decades
preceding the Lemnia, as we learn from vases.5 It occasionally occurs later, but

Fig. 2. — Athena from an Attic vase {Etite dram. i. 80).

1 Roscher's Lexikon, i. 687 sqq. ; 693, Z. 63 sqq.
2 Elite dram, i. 81 ; Millingen, Vases Coghill, 34 ; Roulez, Choix, PI. 7, 2.
3 Especially on the slender ' Nolan ' amphorae, with as a rule only one figure on each side ; cf. Elite
Ciramogr. i. 76 and 76 A, style about 480. Somewhat more recent, ibid. PI. 80, 82, 86 ; cf. Mus. Gregor.
ii. 52, 2 ; 21, 1 ; Klein, Lieblingsinschriften, p. 63, 80 ; Berlin, 2377, 2378 ; Samm. Sabouroff, PI. 61 ; Roulez,
Choix de Vases, PI. 8 ; Mon. d. Inst. 6, 58 ; Benndorf, Vasenbilder, PI. 27, 3 (Berlin, 2251). The seated
figure on the coin (Gardner, Types, PI. 3, 44) is certainly not Athena, but the nymph Thebe (so also Head,
Historia Num. p. 296).
4 Cf. the fine Attic relief of the end of the fifth century, in Lansdowne House (Michaelis, Anc. Marbles in
Gt. Britain, p. 450, 59). The gem Brit. Mus. Catal. 649, Cades, i. K, 66 (for the forged inscription cf. Jahrb.
d. Inst. iii. 214), and a similar one in Berlin (4651), probably reproduce a prototype of the later fifth
century. The same is probably true of the beautiful Cyprian terra-cotta (Ath. Mitth. vi. 250; J.H.S. PI. 16;
A. S. Murray, Hist, of Gr. Sculpture, ii. PI. 17).
5 Athena pouring libation, Berlin vases, 2162 ; in an attitude of triumph with a ship's akroterion, Elite
dram. i. 75 ; pouring wine for Herakles, Wiener Vorlegeb. Ser. A, 2 (Kylix in style of Duris) ; Millingen,
Vases Peints, ii. 41 ; receiving Erichthonios, Mon. d. Inst. i. 10, x. 38 ; receiving a fillet from Nike, Elite
dram. i. 68 ; before Paris, Elite dram. ii. 28 ; with Perseus, Dumont-Chaplain, dram. Gr. ii. PI. suppl. A ;
with Pandora, /. H. S. xi. Taf. ii. ; Roscher's Lexikon, i. 205 1 ; alone, leaning on a lance, on a white lekythos
from Eretria (Athens), similar to the relief (infra, fig. 4). Even in a fighting scene Athena appears without a

BAREHEADED ATHENA I 5
apparently only on monuments belonging to the fifth century or deriving from fifth-
century originals.1 Thus we find Athena without a helmet in the peaceful festal
assembly on the Parthenon frieze and on reliefs inscribed with treaties or decrees.2 It is
interesting to note that the head of Athena without a helmet and wreathed with olive
occurs in the first coin-types of the city of Herakleia in Lower Italy, founded about
433 — 432 (Plate VI. 18). An aegis is given as background to the head in order to
characterize it as Athena.3 The founding of the city was the result of an understanding
arrived at after long conflicts between Thurii and Tarentum concerning the Siritis : the
Tarentines were the actual founders,4 but the choice of Athena in the attitude of
peace for the coin-type was evidently a concession on their part to the Attic
colony of Thurii — a concession typical of the peaceful understanding between the two
states. At this period even Tarentum for a short time stamped on her smaller coin
age (on the obols) the head of Athena without helmet, against an aegis.6 This peaceful
Athena, then, like the Lemnia, is closely bound up with historical events. Some
later monuments which represent the goddess bareheaded are apparently derived
from Attic prototypes of the fifth century,6 so that on the whole the type may
be said to belong to the great period of Attic supremacy within which it took its rise.
The Athena who holds the helmet in her hand or has no helmet at all
almost always wears a fillet or diadem, which is frequently of stately dimensions.7
This is her festal ornament as goddess of peace. In giving the Lemnia a fillet
Pheidias was again following a firmly established tradition.
helmet on the krater, Millingen, Div. Coll. 49. These vases all belong to circa 480 — 450 B. c. The Olympian
metope with the bareheaded Athena (Overbeck, Gesch. d. Plastik, 3rd ed. i. 442) belongs to the same period,
and so does a Greek bronze of the Iran Collection in Vienna.
1 The so-called Kodros Kylix is contemporary with, or only a little later than, the Lemnia.
2 Studniczka, Vermuthungen z. Kunstgeschichte, p. 1 2, A. B. The head of a statue representing the bareheaded
Athena seems preserved in a copy : see Mon. d. Inst. ix. 49 ; cf. Annali, 1872, 5 sqq. (Flasch). The gorgoneion
and the snakes on the diadem seem enough to justify the interpretation as Athena. The head probably repro
duces an Attic original from the second half of the fifth century. The snakes have nothing to do with Hygieia.
3 Good illustrations Gardner (who wrongly interprets the head as Nike), Types, PI. 5, 22 ; Head, Guide, PI.
24, II. The type appears on the obol of the city as well as on the stater. From their style these coins must be
dated immediately after the founding of the city. The usual assumption that they belong to the fourth century
(Head, Hist. Num. p. 59, dates them circa 380 — 300), and that only the small obols with the Herakles head and
the inscription HE (according to Head the only coins stamped between 432 — 380) are older, is untenable. The style
of these obols is exactly the same as that of staters ; especially is this the case in the fight with the lion (also
with the inscription HE) which occasionally appears on the reverse. The inscription HE affords no proof of date,
for among these coins are some inscribed HP (Berlin) which differ in no other respect from those marked HE.
The variation in epigraphy only marks a period of transition ; HPAK... and V HPAK... occur side by side on the
staters. The- staters with the Athena type of Thurii follow very closely on those with the bareheaded Athena.
4 Cf. Antiochos,/?-^. 12, Miiller (Strabo, 6, p. 264); Diod. 12, 36, 3; Busolt, Griech. Gesch. 2, 592.
5 Examples in the Berlin Cabinet ; reverse Kantharos.
6 So the Lower-Italy Kyknos vase, Bull. Napol. n. s. i. 6, from an Attic prototype of the end of the fifth
century (cf. Roscher's Lexikon, i. 2231, 54) ; Mon. d. Inst. ix. 32. The Ficoroni Cista (where Athena
appears among the Argonauts without a helmet) is derived, I believe, from a mural painting of the Polygnotan
cycle (cf. infra). The Athena of the Tarentine vase, Mon. d. Inst. vi. 71, I ; Wiener Vorlegeb. Ser. A, 10, 1
(Judgment of Paris), seems to be dependent on the Lemnia, e.g. the hair, the helmet on the hand, the transverse
aegis, and the drapery correspond. A Florentine gem which belonged to the Medici family and is undoubtedly
genuine (Gori, Mus. Flor. ii. 55, 1 ; Miiller- Wieseler, Denkm. ii. 200a ; Impression, Cades, i. H, 30) is
remarkable : the goddess appears here bareheaded with short hair. A plume and snakes on the drapery
designate her as Athena ; no aegis ; free fifth-century style. An analogous work is the Berlin paste of Solon,
Jahrb. d. Inst. iii. Taf. 3, 8. Both are free ' fantasias ' by artists of the Augustan age on themes of the Pheidian
epoch. 7 Long fillet twice bound round, Elite dram. i. 76. For flowers or zigzags on the diadem cf. Elite dram. i.
68, 75. For Pegasoi on diadem see the krater Millingen, Div. Coll. 49. Cf. in addition to the Parthenos the
earlier coin of Methymna (Head, Guide, PI. II. 27) ; from this evidence it follows that the magnificent head on
a fragment of pottery, known only from Tischbein's drawing (Elite, i. 29), may be named Athena ; the goddess
wears a diadem crowned with Pegasoi, and no helmet.

16 PHEIDIAS
Nor was the short bound-up hair invented by Pheidias. It is quite unknown
in the archaic period, but in the transition style, when hair was more plainly dressed
and more closely tied, the change of fashion influences the type of Athena. We
find it for the first time in some helmeted Athenas belonging to the early decades
of the fifth century. These show the simple round roll of hair on the nape of the
neck,1 introduced by Peloponnesian art and freely adopted for male and female
figures of every kind.2 It seems, then, that these monuments are Peloponnesian in
style and perhaps in origin. In Attic art proper the bound-up hair occurs almost
exclusively in the bareheaded Athena, just in the period immediately preceding the
Lemnia, but the mass of hair is richer and fuller than in the Lemnia.3 The coiffure
of the Lemnia in its extreme simplicity and restraint recalls the plain roll of hair
of Peloponnesian art. In the period after the Lemnia the bareheaded Athena with
bound-up hair sometimes occurs.4 The short hair, not rolled up, worn by the goddess
on the Parthenon frieze, is practically unique.5
Another emblem of peace is the transverse aegis of the Lemnia : it is not intended
as a protection, and therefore leaves part of the breast free. The girdle passes above
it. Pheidias found the essential part of this arrangement ready to hand : it occurs
more especially in the artistic cycle to which the Olympian sculptures belong ; on one
of the Olympian metopes, for instance, representing Athena, bareheaded, seated on a
rock and receiving from Herakles the birds he has slain,6 the goddess wears the
transverse aegis fastened, just as in the Lemnia, on the right shoulder. The only
difference is that the aegis on the Olympian metope covers the left shoulder and the
left arm, which is pressed close to the body, while in the Lemnia the left arm is raised
and the aegis passes under it. The same transverse aegis and manner of wearing it
occur on an Athena rather under life-size lately found in Rome ; the work is an
original belonging to the same cycle as the Olympian sculptures.7 The transverse
aegis frequently appears in later art, but only in the narrow contracted form which it
assumed in the second half of the fifth century. Even the Parthenos no longer wears
the mighty aegis retained by Pheidias for the Lemnia, while the Athena on the west
pediment of the Parthenon wears a very narrow transverse aegis.
Thus the external characteristics of the Lemnia — the uncovered head, the closely
bound-up hair, the festal fillet, the helmet carried in the hand, and the transverse aegis—
1 I. Staters from Corinth ; severe style. A poor example in Brit. Museum, Catalog. Corinth, PI. 2, 20 ;
three better in the Berlin Cabinet.— 2. Staters from Side ; fine severe style, hair short and curly or rolled up
(Berlin) ; Hunter, PI. 49, 3, quoted by Head, Hist. Num. p. 586, is a different type.— 3. Bronze statuette from
Magna Graecia in the British Museum, Coll. Castellani, badly reproduced in Mon. d. Inst. Suppl. Tav. 26, 6 ;
good severe style ; Athena striding. -4. Small silver coin of Kyme in Italy ; severe, not fine (illus. Brit. Mus.
Catal. Italy, p. 88).— 5. Relief from the Akropolis (infra, Fig. 4). In all these monuments the helmet has the
Corinthian shape. The same simple roll of hair, but with the Attic helmet, is seen on 6, the Olympian metope,
Augean stables (Ausgr. von 01. ii. 26 A), and 7, Athena head of Severe style, wrongly placed on a statue
of the Capitol (Jahrb. d. Oesterr. Kunsts. xii. p. 73).
2 Cf. 50th Berl. Winckelmannsprogramm (1890), p. 128.
3 Cf. Elite dram. i. 75, 80, 82 ; Mon. d. Inst. x. 38 ; Overbeck, Gallerie, Taf. 10, 3 (Berlin, 2536) ; Benndorf,
Vasenbilder, Taf. 27, 3 (Berlin, 2251) ; also Berlin, 2378 ; /. H. S. xi. Taf. ii.
4 Cf. the coins already referred to Herakleia and Tarentum, the Kodros Kylix, the Kyknos vase, the Ficoroni
Cista, the gem in the Brit. Museum Catal. 649, the similar one in Berlin, 4651, and the Pergamene statue,
Jahrb. d. Inst. v. 1890, p. 95 (infra, p. 27). The Berlin carnelian (Tolken, iv. 245), showing head of Athena
(not Achilles) wearing helmet, with hair rolled up behind, is influenced by an earlier work.
6 The hair, now much defaced, is wound round with a broad fillet as on the Kyknos vase. Similar short
hair for Athena on the Anesidora Kylix (Roscher's Lex. i. 2058, date circa B.C. 460).
6 Overbeck, Plastik, 3rd ed. i. 442; Fried. -Wolters, Gipsabg. 273. The aegis is absolutely certain, and
the figure (sometimes interpreted as a nymph) is certainly Athena.
7 In the Museo delle Terme ; Parian marble ; head missing.

HEAD OF THE LEMNIA IJ
were borrowed by Pheidias from works already in existence. But the working out of
these characteristics and the whole inner conception of his Athena is the master's
very own. His intention is sufficiently clear. He wishes to present the ideal maiden,
in her purity and strength. The forms of the body are slightly masculine in contour,
the hips are narrow and the abdomen is almost flat, the chest is broad and powerful but
the breasts undeveloped. The head was formerly supposed to be male ; the mistake
was recognized even before the head was found to represent Athena, for not only the
dressing of the hair but the forms, especially of the throat and cheeks, are undoubtedly
female ; yet a right instinct lay below this wrong interpretation, for there is in the
head a strong dash of boyishness combined with feminine charm. The maiden still
shows the delicate austerity of early youth, the woman in her is not yet awake, and she
is still intent on the occupations and sports proper to boy and girl alike. Modern art
has tried to reach a similar goal by similar means ; the type of the full-grown angel
shows the same mingling of boyish and maidenly characteristics. Finally, we may
remember Preller's beautiful comparison,1 made as if in anticipation of our statue,
between the Lemnia of Pheidias and Schiller's Maid of Orleans, who lays down her
arms when war is over.
We now pass from general impression to a close examination of the head and
body of the statue.
The profile of the skull (Fig. 3) forms a perfect curve from the forehead to the nape
of the neck. On the frieze and pediments of the Parthenon this line rises
more perpendicularly above the forehead, and runs more horizontally on the top, as
in the Doryphoros of Polykleitos. The skull of the Lemnia is rather elongated, and
remarkably uniform ; looked at from above it has the shape of a regular oval
equally developed in front and at the back — a shape usually supposed to betoken
an harmonious disposition.
The face viewed from the front forms an elongated and regular oval to
correspond with the skull. The forehead, cheek-bones, and chin are not prominent,
but are bounded by softly curved lines. This is what Lucian means when he praises
the ' contour of the face ' and ' delicacy of the cheeks ' in the Lemnia, and adopts
them for his ideal beauty. The nose (no less admired by Lucian), although marked
off from the forehead only by a very slight depression, is yet set at a more marked
angle to the forehead than in the heads of the Parthenon frieze and other works of
the same period ; the profile line finds its nearest parallel in the Olympian sculptures,
the Spinario and the Massimi Diskobolos, as well as on vases of the earlier
fine style. This is a fresh confirmation of our dating of the Lemnia at about
450 B.C. The lower part of the face is delicate, and recedes rather than projects,
without however appearing weak ; the chin, seen in profile, forms an exquisite
curve. The forehead does not, as might appear from a cursory glance, form a simple
plane surface. Placed in a good light or felt with the hand, it proves to swell
gently and gradually towards the middle ; shallow depressions separate this central
part from the rise over the eyebrows. The modelling of these parts is executed
with extraordinary measure and restraint. The eyebrow itself has a very sharp
edge and is finely arched, the eyelids are delicate, the bridge of the nose is narrow,
the sides steep, the nostrils are full and living, though treated with moderation.
The transition from the nose to the cheek is effected with peculiar beauty. The
nose of the Lemnia differs totally from the Polykleitan nose, the sides of which
1 Hallische Encyklopddie, iii. Vol. 22, p. 186. D

iS

PHEIDIAS

slope much more gradually.1 In the Bologna head the furrow between nose and
upper lip is strongly accentuated, in evident imitation of the sharp contour of the
bronze original. The lips are closed, a feature of frequent occurrence in works of
the first half of the fifth century, but rarer afterwards. The mouth is modelled
with indescribable grace ; every trace of hardness and severity such as still clings
to the Olympian sculptures has disappeared ; and, further, Pheidias has succeeded
in representing the very breath of life in a manner at once refined, measured, and
reserved. The unknown artist who made the head of Antinous Mondragone in
the Louvre2 for Hadrian seems to have attempted to bring some of the charm

Fig. 3.— Profile of the Bologna head. (From a photograph after the original.)

and beauty of the Lemnia into the face of the Emperor's favourite. It was
probably just this Pheidian quality in the Antinous that aroused Winckelmann's
enthusiasm.8 The ears of the Lemnia are worthy of independent study : they are carefully
pi /^^[-'f ^ n°Se iS alS° t0 be Seen on the bronze head from Beneventum in the Louvre (infra,
Plate XIV.), which offers many other analogies to the Lemnia.
2 Fried. Wolters, Gipsabg. 1661 ; Brunn-Bruckmann, Denkm. No. 70. Cf. mouth, cheeks, and parts
below the eyes. r
3 History of Am. Art, xii. 1, 17 (tr. Lodge, vol. ii. p. 335).

HAIR OF THE LEMNIA 19
worked, and only just touched by the hair at their upper edge. The car is slenderly
formed with a long narrow lobule distinctly marked off from the upper cartilage.
The copies of the Parthenos head, as far as they allow comparison, seem to have a
similar ear : we may therefore conclude that this shape was the one preferred by
Pheidias. This is interesting, because on Attic monuments of the Periklean epoch
we find a broader, shorter type of ear, with a wide hollow and a short lobe ; in this
type the distance from the end of the lobe to the upper edge of the hollow is usually
very small, if at all greater than the whole width of the ear. This is the shape of ear
in the Massimi copy of Myron's Diskobolos, and Myron probably learned it from
Kritios ; x we also find it employed by the artist of the east pediment of Aegina,2
but it does not occur earlier. Later it appears in the Aphrodite of the Louvre 3
(probably a copy after Alkamenes) ; on the Parthenon frieze 4 and in other works of
the period 5 it is the common type, although the other more slender form also
occurs ; 6 in the Hegeso-relief, for instance, the ear of the mistress has the one shape,
and that of the maid the other.
The hair of the Lemnia, finally, deserves a closer examination. It is parted
in the middle from the crown (not merely in front as in earlier times), and combed to
either side straight from the parting. The Lemnia appears to be the earliest female
head with this arrangement of hair ; following immediately on it come the famous
Amazon heads. In the severe style the hair was invariably combed forwards from the
crown without any parting, and only parted to the sides over the forehead.7 The
hair is thick and full, rising in a plastic ridge on each side the line of parting ; just
towards the back of the head it lies somewhat flatter. The smooth fillet presses
deep into this soft abundant mass, which is subdivided into a multitude of
separate meshes, each with a movement of its own. Not least charming are the
little curls that cannot follow the track of the rest, but lie separate in front of the ear
and at the back against the neck. Pheidias, however, was not the first to employ
this motive. It was known to the artist (Kritios, as I believe) who made the statue
of a boy, from the Akropolis,8 and to the artist of the Apollo of Mantua ; 9 in whose
works the motive, which seems quite foreign to archaic art, apparently occurs for the
first time.
Despite the marvellous wealth of form and movement displayed by the hair of
the Lemnia on forehead and cheeks, the treatment distinctly follows an earlier
tradition. The general form of the front hair is that of an undulating mass, with
a regular rise and fall. Upon this full mass and in subordination to it the separate
meshes of hair are carved. They start from the parting in almost parallel lines, but
soon lose themselves in a rich tangle. This manner had its rise in a fashion, already
invented in earlier archaic art, of long front hair10 rendered by a solid undulating
1 See the head, 'Eipriii. dpx. 1888, PI. 3 (cf. 50th Berl. Winckelmannsprogr . p. 150).
2 This pediment belongs to the early fifth century, as can no longer be doubted after the discovery of the
Athenian Treasury in Delphi.
8 Gazette Archiol. 1887, PI. 30. Cf. infra, p. 82. « Cf. the beautiful head of the so-called ' Peitho.'
5 E.g. the horseman-relief in the Villa Albani (Hebig, Mttseums, 759) ; the statue of the Coll. Jacobsen
(Arch. Anz. 1891, p. 70) ; the Barberini Suppliant, etc.
6 E.g. in the head of Iris on the Parthenon frieze.
1 This older manner still appears on the Barberini Suppliant, which in other respects is a work of the
free style, and at a much later date on the Artemis head wrongly placed on the Artemis Colonna in Berlin
(Skulpt. 59).
8 Cf. 50th Berl. Winckelmannsprogr. p. 150. The head, 'Etpyfi. dpx- 1888, PI. 3.
9 The replicas are quoted 50th Berl. Winckelmannsprogr. p. 139, Note 61.
10 Cf. e.g. Ath. Mitth. 1889, PI. 3 to the left. Sometimes only the mass was given plastically, and the
lines of hair were painted upon it, e.g., Ath. Mitth. 1879, PI. 5, 6, 1.

20 PHEIDIAS
mass with parallel inner wave-lines. In the later archaic period, this type of headdress
was supplemented by one more artificial, but only to be taken up again with greater
energy in the period of the severe style at the beginning of the fifth century. It
remained very popular for female figures both then and down to the early stages of
the fine style, about the middle of the century, as is testified by numerous Attic
vases,1 Greek terra-cottas, bronzes,2 and coins.3 Among larger works of art are to be
named the foremost of the Charites on the famous Akropolis relief of severe style and
the Athena seated upon a rock on one of the Olympia metopes ; 4 the undulating
mass, with its regular rise and fall, is perfectly distinct in the Athena, though the
separate lines of hair, which once were painted, are now obliterated. This whole system
disappears from Greek art in the second half of the fifth century ; the rendering of the
front hair as a solid mass was gradually given up, and the separate meshes were made
freer and more independent. Undulating front hair was still in favour, but it was
rendered quite differently ; it was compounded of loose separate masses, as may be
seen on heads of the Parthenon frieze, in the Aphrodite of Alkamenes, the Barberini
Suppliant,5 and others. The earlier formation was retained only, as on certain coins,6
when older types were expressly copied.
Later artists must have studied the hair of the Lemnia, for its influence makes
itself felt in many works, as for instance in the hair of the beautiful bronze Camillus
of the Capitol,7 which would be executed just at a time when Pheidias appears to
have been held in special honour.
The proportions of the body still have to be touched upon. They have a
peculiarity, which has already been accurately observed by L. Schorn,8 who re
cognized that the legs of the Dresden statue were somewhat too short. According
to normal proportions (in the Doryphoros, for instance) the length of leg from the
upper edge of the knee-cap to the sole is just three times that of the face, measured
from the root of the hair ; thus, while the upright leg of the Lemnia ought to
measure o'S82, it is, as the measurements show, some centimetres too short. Schorn
has already suggested that the statue was originally placed rather high, so that, the
upper part being seen foreshortened, the statue would appear in the right proportion.
This supposition is borne out by the forward bend of the head, which would produce
an excellent effect if viewed from below. It must therefore be assumed that the
1 In the severe style exceedingly common ; e.g., Wiener Vorlegeblatter,v. 2, 5 (Athena); 7, I (Hetaira) ;
vii. 6, Benndorf, Griech. Vasenb. PI. 19, 3, 36, 8. In the earlier fine style, e.g. in the bride of the Sabouroff
marriage vase (Coll. Sab. PI. 58), the Athena of the Kadmos-hydria (Berlin, 2634), and others.
2 Examples in terra-cottas of the severe style are very common. For bronzes cf. B.C.H. 1891, PI. 9, 10;
Frohner, Bronzes Griau, PI. 27.
3 E.g. Syracuse the Demareteia (Head, Guide, 17, 33 ; Coinage of Syracuse, PI. I. 10), and the succeeding
severe and severe-fine coins of the so-called transitional style, Head, Coinage of Syracuse, PI. 2, 2, 6, 7 ;
3, 2 = Guide, 17, 37. Also the female heads of the severe-fine style in Segesta (Wiener Numism.
Zeitschr. 1886, PI. 7, 11), Pandosia (Head, Guide, PI. 15, 11 = our Plate VI. 8), Metapontum (where a
beautiful Bacchic head with this hair also occurs. Friedlander v. Sallet, Berl. Munzkab. No. 731), and
Terina (Berlin) example of the severe style (loc. cit. No. 688) ; then the earlier type of the Athena head of
Thurii (Head, Guide, 25, 17 = our Plate VI. 5), and its numerous repetitions in other towns of lower Italy
(Herakleia, Velia, Neapolis, Cumae, Hyria, Nola, and others). All these types of coins originate in the middle
decade of the fifth century (cf. infra, p. 104).
4 The somewhat earlier relief of the flower-girl from Pharsalos (Friederichs-Wolters, Gipsabg. 41) may
also be quoted.
6 Notably also the female heads upon coins, which follow immediately on the types enumerated above
in note 3.
6 Especially the numerous repetitions of the earlier type of Athena from Thurii in the towns mentioned
above (cf. infra, p. 106).
7 Helbig, Museums, No. 607.
8 In the Amalthea, quoted supra, p. 6. Feuerbach and Hettner agree with him.

DRAPERY AND POSE OF THE LEMNIA 21
Lemnia did not, like many of the statues in the Akropolis, stand on a low pedestal,
but upon a higher base in the form of a column or pillar, such as was customary in
earlier times both on the Akropolis and elsewhere. Excavations on the Akropolis
have brought to light numerous bases of this kind from pre-Persian times, the most
celebrated being that supporting the statue by Antenor.1 Later this mode became
more rare, and if the Lemnia was really raised upon a pillar or column, Pheidias was
in this again submitting to an earlier tradition.2
In the best copies of the Parthenos the leg is also rather short, though less so
than in the Lemnia. It is probable that, in a colossal image made to be looked at
relatively close, Pheidias kept the lower part a little short, because to a spectator
standing near the upper part would naturally appear considerably reduced.3
Thus the extant works confirm the traditional reputation enjoyed by Pheidias
for his delicate knowledge of optical laws. And the statues of Pheidias rank with
the Parthenon, which as is known surpasses every other antique structure in its
optical effects. V. Drapery and Pose of the Lemnia.
We still have to discuss, in the dress and the position of the legs, the most
important factors in the general effect produced by the statue.
Even the type of dress Pheidias found ready to his hand, and actually in use for
Athena, as is clearly proved by a relief and a small statue, both from the Athenian
Akropolis 4 (Figs. 4 and 5). These are at the same time the closest antecedents to
the Lemnia of Pheidias that can be pointed out. The dress is the same on all three
monuments ; it consists of the Doric peplos, made of a simple piece of strong woollen
stuff fastened upon the shoulders.5 It falls over from the neck, so that the upper
portion of the body is covered with a double piece of drapery. On the right side it
is open, but is prevented from gaping by the girdle which is put on over the whole, so
as to confine the diplo'is also. A mass of pictorial material preserved on Attic vases
enables us to specify with tolerable precision the date at which this dress was intro
duced into Athens.6 It is absolutely unknown in the period of the severe style, which
from the find of the Persian debris must be dated before 480 ; at this time the women
wear exclusively the Ionic linen chiton. To help us to date the succeeding stages of
Attic vase-painting, we have, in addition to the finely graduated development traceable
in the paintings themselves, and to epigraphical criteria, certain definite points
afforded by dated sculpture, such as the Tyrant-slayers, the metopes and the frieze
of the Parthenon, and, finally, the reliefs of the Nike temple and its balustrades.
Grouping the vases upon this basis, it results that even in the transitional vases of ' fine
1 Jahrb. d. Inst. 1887, p. 139. Cf.J.H.S. 1890, 215 ; Ath. Mitth. 1890, 126.
2 One is reminded of an anecdote told in Tzetzes, Chil. 8, 340 seq., 353 seq. (Overbeck, 5. Q. 772, 810),
which, though a late invention, must contain a kernel of truth : it certainly implies the existence of a statue
by Pheidias set upon a column and especially constructed for its high position. For the competition with
Alkamenes which is the second element in the anecdote cf. p. 84 and note.
3 Cf. Brunn, K. G. i. 195.
4 Relief: AeKriov apx- 1888, 103, 123; Ath. Mitth. 1890, 22, 9. Statue: Sybel, Katal. 5003; Roscher's
Lex. i. 695, 35 ; Studniczka, Zur Gesch. d. Altgr. Tracht. p. 142 ; 'Ecp. apx- 1887, PI. 8, p. 148 seq. ; Ath.
Mitth. 1890, 22, 8.
6 Cf. Studniczka, loc. cit. 141 seq.
6 For the change in dress at Athens in the fifth century cf. Bohlau, De Be Vestiaria Graeca, p. 56 sqq. ; Stud
niczka, Zur Gesch. d. Altgr. Tracht, p. 26 sqq.; Rom. Mitth. ii. 54 ; Winter, Die Jung. Attischen Vasen, p. 27 sqq.;
Benndorf, fahrb. d. Oesten. Kunstsamml. xii. I, p. 53.

Fig. 4.— Athena on a relief from the Akropolis.

DRAPERY AND POSE OF THE LEMNIA

23

severe ' style, dating from 480 to approximately 465, the Doric peplos was still but
little known in Athens. Only rare examples, on young girls and on Athena, occur at
this time: in these the peplos is not girt over the
diplois, but it is either left loose or girt out of sight
beneath it; in the latter case it is drawn over
the girdle so as to form a kolpos that falls more
or less low.1 Now these are the two modes in which
the Doric peplos makes its appearance in Pelopon
nesian art, where it occurs frequently in the severe
style about the end of the sixth century. As a rule
it is ungirt ; yet it is frequently sewn together,
instead of being open at the side. Some important
artist of the period, circ. B.C. 500 — Hagelaidas perhaps
— must have adopted this dress for a statuary type of
which we have the reiterated echo in numerous works.2
One of its distinguishing marks is the severe sym
metrical arrangement of the folds of the diplois ; the
edge forms a level line in the middle both back and
front, and then falls down to either side in corre
sponding folds. The attitude allows the one leg to
be just a trifle relaxed.3 This Peloponnesian type
had an immense influence, extending far beyond the
narrow Peloponnesian school ; 4 figures reproducing it
have been found even as far as Lykia,6 in Pantika-
paion,6 and among the Phoenicians.7 The artists of
the Olympian temple sculptures were also under its
influence : they make use both of the ungirt peplos
and of the peplos with the kolpos under the diplois?
Parian sculpture seems to have contributed especially
to the propagation of the type.9 It is also found in the west. A superb bronze
statuette from Magna Graecia represents Athena clothed in this manner ; 10 other
bronzes of the same provenance show Athena striding, but clad as in the type where
she is standing tranquilly.11 The artists of the Hestia Giustiniani and of the

Fig. 5.-

-Statnette of Athena from the Akropolis.
(From a drawing.)

1 Ungirt : Mon. d. Inst. xi. 14, quite young girl.— Potsherd of the Akropolis, found by S.E. corner of the
Parthenon, but certainly not among the Persian debris ; Nike with Akrostolion.— With the kolpos : Gerhard,
Auserl. Vasenb. 300, 301 girl; Mon. d. Inst. xi. 19, Athena.— The same two kinds, ungirt and with the
kolpos, appear on two of the Charites in the well-known relief (Roscher's Lexikon, i. 881), the original of which
is about contemporary with the Tyrant-slayers, and must therefore have been made about 470.
2 Especially in the female supporting figures of the Corinthian mirrors and similar little Peloponnesian
bronzes. Cf. Olympia, iv. Die Bronzen, p. 21, 56; 27; Athen. Mitth. iii. PI. B. Arch. Ztg. 1881, PI. 2, 2.
3 For the corresponding male type in Argive art, cf. 50th Berl. Winckelmannsprogr. p. 124 seq.
4 Cf. e.g. the Boeotian terra-cottas of the type Dumont-Chaplain, Cdram. ii. PI. 3.
5 Three torsos in the Brit. Mus. in Parian marble, from Xanthos. A. H. Smith, Catal. of Sculp, i. 96—98.
6 Gilt wooden figure from Kertsch, in the Louvre, about 50 cm. high, without head.
7 Phoenician sarcophagus from Sicily, Perrot-Chipiez, Hist, de I' Art, iii. 189, Fig. 134.
8 Standing women on the east pediment ; Hesperides and standing Athena on the metopes. Cf. also
Athena in the Terme Museum at Rome (mentioned above, p. 16, note 7).
9 Cf. the close proof which I have endeavoured to give in Arch. Studien, H. Brunn dargebracht 1893,
p. 69 seq.
10 Gazette Arch. 1881, PI. 7, an undoubted Greek original. On the other head, the Pozzuoli bronze, Arch.
Ztg. 1881, PI. 2, is only a late imitation, with the folds much weakened.
11 In the Brit. Mus. from the Castellani Coll. Cf. above, p. 16, notes I, 3. In Berlin, Inv. 6242. Both are
originals of 470 -460; the striding Athena from the Akropolis, Arch. Ztg. 1873, PI. 10, is probably pre-Persian.

24 PHEIDIAS
originals x of the ' dancers ' from Herculaneum take a more independent line : they
give up the symmetrical folds, and aim at a more faithful reproduction of nature.
The Herculaneum figures come at the end of the series ; in them the relaxation of
the one leg begins to be emphasized. Possibly Kalamis and Pythagoras treated
standing female figures in this manner.
Athens at first fought shy of this widely spread Peloponnesian type ; traces of
its influence can certainly be detected,2 but, as vases show, the Doric peplos was not
popular in Athens until somewhere about B.C. 465. From that time, however, it
became more and more frequent, though it is transformed in the independent Attic
manner : it became customary to gird the peplos, open at the side, over the diplois —
a plain, homely fashion, which elsewhere was only employed occasionally for atten
dants,3 but was at Athens exalted into a dress of distinction. At first the Ionic
chiton was still worn underneath, as in the Athena of the Akropolis statuette, which
is one of the earliest monuments that reproduce this new fashion, and in the Athena
on some vases.4 From the first the dress was by no means reserved to Athena : it
was given indifferently to any young girls, divine or human. And by the side of this
new mode the old ungirt Doric peplos now appears quite frequently. The dress,
however, was evidently looked upon as a foreign innovation : wherever a company of
young girls, Muses or Nereids, for instance, is represented, at first only one or the
other wears it. Of the two Eleusinian divinities, Kore, the maiden, is the first to wear
the Doric garment ; and only later was it given also to Demeter.5 The dress only
became really naturalized after 440, on the vases which correspond to the Parthenon
frieze, and then upon the later vases, contemporary with the Nike temple and the
Erechtheion. Alongside the special Attic fashion of girding over the diplois, the old
fashion with the under girding and the kolpos now came again into vogue.6
Returning now to the Lemnia, we see that the dress in which Pheidias repre-
1 Comparetti de Petra, Villa d'Ercol, PI. 14. These statues are certainly only late copies, and not, as has
been supposed, Greek originals. This is convincingly shown by such external marks as the plinths, the technique,
the buttons on the shoulders, etc.
2 As in terra-cotta dolls, certainly Attic, of the type Dumont-Chaplain, dram. ii. PI. 4. The Attic origin
is less sure in some bronzes from Athens (cf. supra, p. 23, note 11) ; it may probably be admitted however for the
beautiful statuette, Frohner, Coll. Grtau, Bronzes Ant. PI. 27, which may be dated circ. 460.
3 Cf. the two Olympian pediments, where it is worn by the attendants (these two figures, it is true, are
only preserved copies), while the mistresses wear the established Doric types. Cf. also the maid behind
Odysseus on the Polygnotan vase with the murder of the suitors.
4 Mon. d. Ins.'. x. 54, and the Polygnotan Argonaut vase, Mon. xi. 39.
5 Cf. Arch. Anzeiger, 1891, p. 118, 114.
6 The following is a selection of characteristic examples : I. Peplos girt over the diplois : vases of the time
circ. 465—450 ; Elite dram. i. 64 (Eileithyia), ii. 36, 41 (both Artemis) ; Mon. d. Inst. xi. 40 (Artemis) ; Berlin,
2381 (Nike, front view), 2521 (Maiden); Frohner, Burlington Fine Arts Exhibition, No. 51 (Maidens playing
musical instruments, the one in front view).— Rather later, from about 450 : Gerhard, Auserl. Vas. 243, 291, 305 ;
Mus. Greg. 11. 19, 2. 21, 2; Dumont-Chaplain, dram. ii. PI. suppl. A. (Athena); Sachs. Ber. 1875, PI. 3 ,.
(Athena); /. H. S. xi. 11 (Athena) ; A. Z. 1881, PI. 15, 16 ; Millingen, Div. Coll. 57; Mon. d. Inst. x. 53 ; Wiener
Vorlegebl. E, 12 ; Heydemann, Vasenb. i. 1 ; Overbeck, Atl. d. Kunstmyth. PI. 12, 2 (Amymone) ; Monum.
d. Inst. x. 39 (Aglauros, not Athena).-About 440 : Elite dram. ii. 26, 62, 72, 86 A (one muse out of seven) ;
)'J^ I75 ,r fab0Ur°ff' "J MiUin> Vases Peints, i. 54; Overbeck, Atlas, PI. 15, 31 ; Mus. Gregor. ii. 82, I
(Medea) ; Mon. d. Inst. 11. i5.-Towards 430 and later, on vases in the style of Aristophanes and Meidias, the
dress has become the prevalent one. -II. Ungirt peplos. Earlier instances, from circ. 465 : Mon. d. Inst. viii.
S' 2 ''t 6!.37' 38 ! ™fmf"' D™- Coll. 60; Elite dr. ii. 57 (Artemis) ; Mus. Greg. ii. 24, I.-About circ. 450-
440 : Berlin, 2388 (Muses), Mon. d. Inst. 5, 37 (Muse) . Mus. Greg. ii. 13, 2 ; 15, 2. 19, 2. 20, ,. 63. 2 ; Gerhard,
Auserl. Vas 302, 5; Overbeck, Atl. d. Kunstm. PI. 15, ,3. „. 23; 18, ,5 ; Elite dram. 1, 83. 9X ; Coll.
Sabouroff, PI. 63, 55 ; Millingen, Div. Coll. 55, 57. What Studniczka (Z. Geschichte d. Tracht, % seq.) has
endeavoured to ascertain from literature with regard to the dress accords on the whole with the vases; he is
however mistaken in holding (loc. cit. 142 seq.) that the ancient image of the Polias already wore the peplos like
he Parthenos. The prize amphorae, on which he relies, belong only to the fourth century, when the dress of the
1 arthenos was an established type ; the garment never appears on the old representation.

DRAPERY AND POSE OF THE LEMNIA 25
sented her had already been given to Athena in Attic art, though its invention was
still quite recent, and by no means established and generally accepted. The
individuality that lay in the thick simple Doric garment was still felt in all its fresh
ness. The image of the pure and powerful maiden received thereby a material
completion, while it was only with the girding over the diplois that the
dress became properly expressive of Attic sobriety and compactness. On vases
of the time about 460 — 450 there are many figures that clearly reproduce this
character. These are the vases that on other grounds may be brought into close
connexion with the painting of Polygnotos.
The two marble works, of the Akropolis, named above, cannot belong to a
very much earlier period. The relief must be dated at the earliest about 460, and
the statuette somewhere circ. 465, the time when the first examples of the new dress
appear also upon vases. It has been unnecessarily supposed that the statuette came
from the Persian debris : this seems to be most unlikely. It is however probable,
from the circumstances of their discovery, that both works came to be buried when
the Periklean Parthenon was built.1 A summary treatment of the past is customary
in periods of great progress ; and the excavations on the Akropolis show how truly
this applies to the time of Perikles. Style and workmanship as well as the Athenian
fashion of wearing the girdle prove both works to be purely Attic.2 The Doric
garment of Peloponnesian art with its symmetrical folds, adopted also by the
Parian sculptors, is completely given up, and replaced by independent fresh
observation of nature.
The promise contained in these two works was more than fulfilled by Pheidias.
The hard, constrained, and timid touches that may still be detected in statuette and
relief disappear in the Lemnia. A freer, more majestic air pervades the whole.
Even the diplois is no longer so short and straight, but falls in greater fulness, and
wherever on the relief the folds, with their round backs, lie side by side in uniform
monotony, on the statue the intermediate spaces are made to vary in breadth, while
the backs of the folds no longer present a round, almost padded appearance, but
begin gently to sink and flatten in the middle.
Turning now to consider the way in which the Lemnia stands, the statue will
be found to differ in this point from the type prevalent in the first half of the
fifth century, according to which the leg in action is only slightly set free, and is
brought either close to the other or in front of it, with the foot flat on the ground.
In the Lemnia the leg is considerably more to the side, so that the foot, though not
raised, yet rests solidly only on the ball. Even in this, however, Pheidias is not
entirely without precedent ; the Apollo of Mantua, which has already been mentioned
on account of the hair (p. 19), differs from the prevalent type, in having the same
broad posture with the free leg very much to the side.3 Some of the Herculaneum
maidens also come near to the Lemnia in this respect. Pheidias has, however,
known how to remove all constraint and hardness from the posture. In the
Parthenos he ventures upon a stronger innovation ; the free leg is brought not
merely to the side, but also drawn somewhat back, so that the heel is raised. The
leg, released from the weight of the body, is not completely at ease, as in the case of
a person standing still, but is slightly drawn up towards the other, as when a pause is
1 The relief appears to have been built into a structure of the Periklean epoch, which has been taken to be
the Ergasterion of Pheidias. The statue was discovered when the foundations of the Museum were laid.
2 Studniczka, loc. cit. 'E</>. dpx- 1887, 153, unnecessarily explains the Athena to be the work of a Pelopon
nesian artist working for Athens.
3 Cf. 50th Berl. Winckelmannsprogr. p. 140. E

26 PHEIDIAS
made in walking. The creation of later Argive art, with the leg still drawn back
in the walking attitude, is exactly parallel to the pose introduced by Pheidias.1 The
Argive scheme was not adopted in Attic art till the period following upon the
undoubted works of Pheidias. The pose adopted by Pheidias for the Parthenos is
not found in earlier works, and grew out of the stage represented by the Lemnia ;
it was intended to give to the figure freedom of movement without detracting from
its majesty, which, according to the Pheidian conception, would have been the case
had the leg been quite drawn back in the ordinary walking attitude.
Our inquiry has shown the Lemnia to be closely connected with earlier fore
runners, and yet to be an absolutely independent work, pervaded with a powerful
personal spirit. In none of the earlier statues do we find anything that can be even
remotely compared with the broad energetic throw of the draperies, or with the
eminently individual character of the superb head. Everything in the statue betokens
a fresh and genial spirit, that enters upon its course with conscious strength. The
Lemnia is its earliest manifestation that is known to us, and the Parthenos follows
close upon it. Yet, according to the received view,2 which places the birth of Pheidias
at about B.C. 500, the artist must have been already fifty years of age when he created
the Lemnia, so that not only his youth but the prime of his manhood, as well
as the greater number of his works, must have fallen within "the preceding period.
Were we to judge only from the monuments, we should rather take the Lemnia
to be a work of the artist's prime. This seems confirmed on the one hand by the
bold freshness which even in the Parthenos had already somewhat faded, and on the
other by that fidelity to earlier traditions which has been observed in sundry
particulars.

VI. Monuments related to Lemnia and to Parthenos. — The Athena Promachos.
At this point it becomes important to note the existence of a number of
monuments which may be grouped with the Lemnia and the Parthenos. Some are
of the same date or only a very little earlier; others again are later, but they all
bear more or less distinct traces of the personal style of Pheidias : thus it is about
B.C. 450 — though not earlier — that we come upon Pheidias everywhere.
A statue of Artemis 3 in the Villa Albani, of which the head is unfortunately
missing, may serve as an example of a work closely akin to the Lemnia, though
of slightly earlier date. The coarse woollen stuff is treated with the same vigour
and naturalness, but the folds are still convex and show no depression. The
majestic figure is clothed in the Doric peplos girt in the Attic manner. The peplos
is closed at the side, and the part folded over is unusually long, falling below the
knee. The right arm was raised, and the left hand carried a young roe, in archaic
fashion. The leg in action is placed to one side, as in the Lemnia, though its
movement is rather less energetic. Two heavy folds fall perpendicularly from
the knee. Another work very closely related to the Lemnia has been preserved in a torso
1 For the apparently contemporary introduction of the ' walking ' motive in Peloponnesian art, see below, at
the end of the chapter on Myron. Cf. also Winter, Die Jung. Att. Vasen.
Brunn, K. G. i. 164 ; after him Overbeck and others.
3 N°' ,662' Gerhard> dnt. Bildw. Tf. 12; Roscher's Lexikon, i. 562; Clarac, 678 F, 1621 B. ; Helbig,
Museums, 11. 856. The head, right arm, and right foot are restored. The right arm was raised.

ATHENA PROMACHOS 27
of the Louvre 1 and a replica in Madrid. It also represents a young girl wearing
the peplos girt above the diplois. The simple, firm, almost rough treatment of the
drapery recalls the Lemnia, especially in the folds under the girdle. The attitude
is even more vigorous than that of the Lemnia, owing to the feet being very much
turned out ; further, the left foot, although not drawn back, is placed more to the
side. The girdle consists of a wide band of stuff tied in a bow in front. The
whole figure is so unconventional, so full of fresh individuality and natural simplicity,
and through these qualities so closely allied to the Lemnia without yet being in
the smallest degree an imitation, that I imagine it to represent another creation of
Pheidias. A statue of the Capitol 2 akin to the work of Pheidias but not bearing the stamp
of his individuality may be placed, owing to its pose and the treatment of the folds,
somewhere between the Lemnia and the Parthenos.
A small statue in the Lateran 3 corresponds very closely to the Parthenos, and the
same may be said of a beautiful statue in St. Petersburg,4 in which, however, are to be
seen some slight indications of a later style. This figure also represents a youthful
goddess ; she wears an Ionic chiton under the peplos.
Many more examples might be added,5 such as a whole series of statues
which reproduce the main features of the Parthenos combined with all sorts
of later traits : in most of these one foot is drawn back in the walking position.6 It
is not however my intention to discuss these statues here. Mention need only
be made of an. Athena from Pergamon which is very closely related to the Lemnia.
The statue has wrongly been taken for an original,7 whereas it is merely a copy
executed in the loose Pergamenian manner after an original by some artist of the
fifth century, who utilized the Lemnia for the body of his Athena, but in the treat
ment of the head betrayed his affinities with the school of Kalamis, which we shall
have to discuss in detail later on (p. 8l).
We turn from this ugly Pergamenian Athena to a work which has every claim
to be mentioned in the present connexion — namely, the Torso Medici in the Ecole
des Beaux- Arts in Paris (Fig. 6).8 Its peculiarly Pheidian character has been generally
recognized,9 and it has been rightly placed in close relation to the Parthenos ; its
affinities to the Lemnia are no less marked. It may be said that a common character
closely unites all three works. They represent the same conception of the maiden
goddess, with the narrow hips of a boy, and the broad though undeveloped breast of
a young girl. In all three the folds of the peplos, which is girt, are rendered with
1 (a) Louvre, Gal. Denon. No. 2903, rather under life-size. Pentelic marble ; head and right shoulder were
put on separately, and are now lost, (b) Madrid No. 70, Hubner No. 43.
2 Capitol, Salone No. 29, restored as Hygieia ; the portrait head is foreign to the statue. Weight on the
left leg, the right leg to the side and a little drawn back, the foot flat on the ground.
3 Benndorf-Schone, Catal. No. 6. (Phot, in the German Institute at Rome.)
* Stephani, Compte Rendu, 1881, PI. VI. p. 130. Cast in Dresden. Cf. Wochenschrift f. Klass. Philol. 1885,
p. 292 ; Jahrb. d. Inst. Anzeiger, 1889, p. 10. The head does not belong to the statue.
5 Thus a statue in the Villa Mattei (Matz-Duhn 1375) is very similar to the Parthenos, though the weight is
thrown on the left leg.
6 Some instances are mentioned by K. Lange, Arch. Ztg. 18S1, 197, n. 2.
7 Conze, Sitztmgsberichte der Berl. Akademie, 1893, xvi. p. 207. Puchstein, Jahrb. d. Inst. 1890, p. 95.
Kalkmann's theory (Prop, des Gesichts, p. 66), that the head did not originally belong to the statue, is de
monstrably false (see my remarks in the Berl. Philol. Wochens. 1894, p. 1142).
8 Cf. Friederichs-Wolters, Gipsabg. 476 ; Gazette des Beaux-Arts, 1890, i. 281 ; Brunn-Bruckmann, De/ikm.
No. 171.
9 Puchstein alone (Jahrb. d. Inst. 1890, p. 90), starting from his mistaken conception of the Parthenos, which
he places in the period of the sculptures of Olympia, has disputed the Pheidian character of the torso.

Fig. 6. — 'Torso Medici.' (Ecole des Beaux-Arts, Paris.)

ATHENA PROMACHOS 29
the same massive energy, and the coarse woollen stuff is indicated in the same way.
The simple straight line at the lower edge of the diplois and the converging folds at
the girdle are points of special similarity. Any one who has an eye for essentials
must admit that the three form a close group that stands out distinctly from other
extant creations of the fifth century. There are of course many differences in detail.
We have already characterized the Parthenos as a later work than the Lemnia.
The Torso Medici will be found to be the latest of the three. It displays a greater
richness of motive, more splendour, but less simplicity and restraint.
The folds of the peplos are crowded closer together. In the Lemnia the groove
like depressions on the backs of the folds practically only appear on the diplois below
the girdle, but in the Torso Medici they also appear on the large perpendicular folds
over the leg which bears the weight of the body ; and they are more regularly and
deeply hollowed. Further, in the drapery of the Lemnia a small round tube-shaped
fold occurs sometimes in the furrow between two projecting ridges : 1 in the torso it is
of much more frequent occurrence. We have already observed that, as compared
with the Lemnia, the Parthenos produces a stronger effect through the greater
definiteriess in the scheme of its draperies. The Medici torso represents a still more
advanced stage of the same tendency. The Lemnia charms more than either by the
greater naturalness and truth with which the drapery is rendered. In the torso the
regular perpendicular folds over the leg at rest, and the manner in which these folds
rest on the foot (a motive not found in the Lemnia), mark an immediate connexion
with the Parthenos.
In addition to the Doric peplos the Medici figure wears the Ionic under-garment.
This in itself is no argument for a later date. On the contrary, this double garment
seems to have found great favour in Athens as a transitional fashion 2 just at the time
when the Doric peplos was introduced ; it appears frequently on Attic vases between
450 — 440, and particularly in representations of Athena.3 But the artistic treatment
of this garment on the torso, and the carefully thought out contrast between the fine
clinging linen chiton on the right leg and the heavy woollen folds which cover the
left leg, while leaving the right free, make it probable that the torso is a later work
than the Parthenos. Already in the Parthenos there is, by comparison with the
Lemnia, an attempt to emphasize the leg in action by means of the clinging drapery.
By the arrangement of drapery adopted in the Medici torso the desired contrast
between the two legs becomes still richer and more effective.
The artist's principal aim in the stylistic treatment of the linen garment was to
express the nature of the material. This he did by means of fine parallel lines
running downwards. Actual folds are few and broadly rendered. This treatment
bears distinct traces of the archaic tradition. These small wavy stripes are only a
freer form of the archaic method of representing linen. The archaic method, however,
as is well known, did not combine with this technique any attempt at the rendering
of real folds.4 The linen chitons of the Charites on a relief which must be dated
about B.C. 473 are still without folds, but completely covered with fine wave-lines.
In the torso Medici we find this linen technique combined with a broad treatment
of folds. The next step in development was to abandon the fine lines and to
replace them by real folds. On a metope of the Parthenon (XXIX. S. side) the
1 Cf. Puchstein's section, loc. cil. p. 94. This observation, which is emphasized by Puchstein himself,
loc. cit., ought to have sufficed to prevent him from separating the torso Medici from Pheidian work.
2 As in the Athena from the Akropolis, Fig. 5-
3 Cf. Arch. Ztg. 1876, Tf. ii. (Berlin 2354).
4 Cf. Studniczka's excellent remarks, Rom. Mitth. iii. 1888, p. 287 seq.

3o

PHEIDIAS

same method is employed as in the torso, but on the frieze and the pediments 1 the
linen of the Ionic chiton is scarcely even characterized by parallel wave-lines. In
their place we have a multitude of small real folds, which though not strictly true to
nature produce a very rich effect. The artist of the torso combined the wave-lines
with a few large folds, and thereby attained an infinitely more natural effect.
Hence it may be dated as contemporary with the metopes, and earlier than the frieze
and the pediments. In the latter two the woollen garments are already less heavily
rendered, and the contrast between the wool and the linen is less emphasized. On
the other hand, the famous Amazon statues belong to the same stage as the torso.2
True, the treatment of the linen stuff in each of the three Amazon types varies con
siderably in detail, but the principle is the same, and in the Mattei type not only the
principle but the whole manner is that of our torso. The linen stuff is indicated by
the same closely drawn parallel wave-lines, and only the main folds are represented.
Although the wave-lines almost tend to become real folds, thus showing the
Amazon to be the later work, the stylistic affinities of the two statues make it
probable that they are nearly connected. But we shall have to speak of the
Amazon again.
The rich drapery of our Athena is completed by a mantle thrown over the left
shoulder. It is of the same strong woollen stuff as the peplos, and is treated in the
same manner.
The torso is a copy of Roman date,3 as is evident to any expert from the style
of the work.4 Though good on the whole, it is, like most copies, not without dull and
lifeless parts. Other but much inferior copies of the same original are in existence :
these are a small statuette, a torso three-quarters life-size, and a relief, all three in
Athens.5 The relief is specially important : on it an olive-tree with an owl appears
beside the goddess, whence it is probable that the original stood on the Akropolis ;
the relief also shows that the statue carried the shield raised in the left
hand. This agrees with certain indications on the colossal torso itself. The left
upper arm is lowered and held somewhat away from the body ; the forearm
was extended sideways and slightly raised ; the hollow where it was let in is still to
be seen,0 together with a broad contact-surface at the back of the shoulder, which
proves that some large object, which can only have been the shield,7 was fastened on
at this spot. Held in this way, a considerable portion of the inner side of the shield
would be visible from the front, as is also the case on the relief. The mantle filled
the space between shield and body, while shield and mantle together formed a
magnificent background for the side of the figure represented at rest.
The other side was more animated. The right arm was not close to the body, but,
as the torso and the relief show, it was placed somewhat to one side. As the goddess
held the shield in her left hand as if in readiness for the fray, it is reasonable to suppose
that in the right hand, which is also missing on the relief, she held the spear, her
weapon of attack. The statue stands like the Parthenos, the foot in action being
set back and the heel raised, not flat on the ground as in the Lemnia. This foot
1 Cf. the so-called ' Peitho ' on the frieze, and the female figures in the right angle of the east pediment.
2 For the three types see Jahrb. d. Inst. 1886, p. 14 ; cf. infra, chapter on Kresilas, p. 128 sqq.
3 Puchstein, loc. cit., is of the same opinion.
4 Moreover the marble seems actually to be Italian, as Nibby first noticed.
s Von Sybel, Ath. Mitth. 1880, 102, Taf. 5.
6 Both arms and the head were made of separate pieces.
7 Just the contrary of what Wolters maintains in the note to Friederichs's Bausteine, 82 (Fried. -Wolters, 476).
It is impossible to suppose that there was nothing here but a nude forearm and a lance.

ATHENA PROMACHOS ' 3 1
moreover is much more turned out than in either of the other statues, and in harmony
with this movement the head is turned to the right. This point, not visible on the
coarse and mutilated relief, is made quite clear by the small torso in Athens 1 and by
the Torso Medici. In both the bunch of hair at the back of the neck is pushed towards
the left shoulder, and in the large torso the hollow for the insertion of the upper part
of the knot of hair (now lost with the head) plainly proves that the head was turned
to the right.2 On the head rested, as the relief shows, the helmet with its splendid
triple plume, similar to the one worn by the Parthenos.
Thus restored, the statue, although standing so firmly and tranquilly, becomes
instinct with life and almost with animation. An Athenian coin, struck in Imperial
times,3 shows that the right hand did not lean on the lance, but grasped it low down as
if to raise it for the attack. The similar pose of the figure, the raised shield, and the
turn of the head prove beyond a doubt that this coin reproduces the original of
our torso.4 This conception of the goddess is quite different from that of the pacific
Lemnia who grasps the spear high up in order to lean on it, gently bending her head
and carrying her helmet in her hand. It differs also from the stately festal Parthenos
who has laid aside her lance and shield and grasps a figure of Nike. The Athena
of the torso is the warlike maiden looking about her with courage and resolve,
ready for defence or for battle.
Finally we have to remember that the original of the torso Medici must have
been of colossal size. All the premises adduced point to the conclusion, bordering
on certainty, that this original was the statue of the Akropolis known as the
Promachos, an opinion which Konrad Lange expressed long ago.6 The Pheidian
style of the work, the warlike conception of the goddess, the presence of the
traditional attributes of the Promachos such as lance and shield, the position of
restrained activity testified to by the coins,6 the colossal size, and the fact that the
statue stood in Athens, seem to me, when taken altogether, absolutely convincing.
Final confirmation is found in the turn of the figure, which, as Lange has pointed
out,7 is appropriate to the place occupied by the Promachos on the Citadel ; the
figure towered high above the Akropolis wall and looked towards the city, while
its front was turned in the direction of the great gates. Like the Lemnia, this
statue also was constructed with careful appreciation of the spot it was to occupy.9
According to Pausanias, the Promachos was a work of Pheidias. Neither Pliny,
1 Ath. Mitth. 1880, Taf. 5, 2 ; p. no.
2 Curiously enough K. Lange did not observe this confirmation of his theory ; cf. Arch. Ztg. 1881, 203.
3 Imhoof-Blumer and Gardner, Num. Comm. PI. 2, I. II. , p. 128 ; Arch. Ztg. 1881, 197 ; Collignon,
Pheidias, p. 15.
4 The omission of the under chiton and of the cloak and the simpler form of the helmet are mere simplifi
cations introduced by the coin-engraver.
6 Arch. Ztg. 1881, 197 sqq. ; Studniczka, Verm. z. Gr. Kunstgesch. p. 10.
8 The tranquil attitude of the Promachos is plain from the coins which give the view of the whole Citadel,
though, as K. Lange (Arch. Ztg. 1881, 198) and Imhoof-Blumer and Gardner (Num. Comm. p. 129) have
already shown, these coins are no guide to the actual composition of the statue.
7 loc. cit. 200.
8 This remains true whether the statue stood upon the basis whose existing remains were formerly supposed
to belong to it (Loschcke, Tod d. Phidias, p. 45, is of contrary opinion) or close beside it (Lolling, Geogr.
Griechenl. 343, 352).
9 The attempt lately made by W. Gurlitt (Analecta Graeciensia, Festschrift z. Wiener Philologenversamml.
I^93, p. 101 seq.) to obtain certain knowledge of the 'Promachos' from Byzantine sources rests on absolutely
untenable suppositions. — The zealous Byzantine scholar Arethas, writing in the ninth or tenth century A.D., made
the following marginal note in his copy of Aristeides opposite the passage referring to the ivory and bronze
Athena of the Akropolis — 'This is probably the Athena which stands in the Forum of Constantine.' Niketas

32 PHEIDIAS
however, nor any other ancient authority mentions it amongst the works of this
artist.1 A scholion to Aris-teides (Overbeck, Schrift. Quellen, 640), on the other
hand, ascribes it to one Praxiteles, expressly distinguishing it from the Parthenos
of Pheidias. If we wish to criticize the question impartially, we must not, as is
usually done, neglect this last testimony as worthless. The scholion doubtless
goes back to some authoritative source which carefully distinguished the three most
important statues of Athena on the Akropolis — viz. the old Polias, the Parthenos,
and the Promachos, giving in each instance the material, the size, and the artist's
name.2 Now there are three other known instances in which Pausanias assigns to
Pheidias himself statues ascribed by Pliny and others to pupils or assistants of
Pheidias. Such is the case with the Nemesis at Rhamnus, the Mother of the Gods
at Athens, and the Athena at Elis : in all these instances Pausanias gives, as every
one acknowledges, the less trustworthy tradition. We must therefore allow for the
possibility that the information of Pausanias, in the case of the Promachos also, may
have been incorrect, and that one of the pupils or colleagues of Pheidias may have
been called Praxiteles ; further, that common tradition wrongly assigned the statue
to Pheidias, while it was in reality by another artist, perhaps even that the name of
this artist was inscribed on the statue, as in the case of the Nemesis of Rhamnus.
At any rate it seems evident that the Promachos was not signed by Pheidias, from
the passage of Lucian 3 in which it is related as something remarkable that Pheidias
thought the Athena Lemnia worthy of being inscribed with his name. This state
ment would be quite meaningless if the name of Pheidias could have been read close
at hand on the Promachos.
The copies of the Promachos do not help us to a decided answer to this
question. They, however, make it absolutely clear that the statue bore a very
close relation to the authentic works of Pheidias. If a Praxiteles made it, it is
evident that he worked in the manner of Pheidias. The differences which mark
off the Promachos from the Parthenos and the Lemnia are of such a kind as may
(thirteenth century) states that this Athena was represented drawing up her drapery with her left hand. Gurlitt
concludes therefore that the figure which Niketas saw was an archaic one in the Ionic costume of the well-
known ' pre-Persian ' maidens of the Akropolis, and that, like these, the left hand held the drapery. More than
this, he gives unlimited credence to the suggestion of Arethas, and identifies this figure with the so-called
'Promachos ' of Pheidias. He bases this opinion on the presupposition, which we have already shown to be
false, that Pheidias was a semi-archaic artist, who soon after 480 B.C. was in a position to be intrusted with the
most important commission the Athenians had in their gift. Now the hypothesis of Arethas appears entirely
groundless and without authority. But even supposing him to be right, it is still doubtful, first, whether Niketas
understood correctly the motive of the left arm, and, secondly, whether Gurlitt was right in identifying this motive
with that of the archaic statues. The motive, which is very rare for Athena (it occurs only in quite archaic art,
and apparently never in statues), contradicts the traditional type and character of Athena in the periods preceding
and following the Persian wars. Therefore we must assume either that the statue seen by Niketas was an
archaic work, and that Arethas was mistaken, or else that the Byzantine scribe of the thirteenth century misunder
stood, not only the movement of the right arm (as Gurlitt admits), but also that of the left. Nor is it necessary
to suppose that the dainty archaic motive gave rise to the description of Niketas. Just as the right hand had lost
the lance it once grasped, so too the left hand may have lost a shield. If there was any drapery hanging over
the left arm, Niketas might easily mistake the outstretched left hand, robbed of its shield, for a hand holding
drapery.— But we need neither the confused account of Niketas nor the hypothesis of Arethas; the evidence
for the Promachos derived from the monuments themselves is infinitely more reliable than any conclusions
based on these Byzantine writers. Petersen (Rom. Mitth. 1893, 350) also opposes Gurlitt's hypothesis; he
supposes that the statue seen by Niketas may have been an Alexandrine work in the style of the archaistic
Athena published by him ibid.
1 It is not likely that the Cliduchus (Plin. xxxiv. 54) is identical with this statue.
2 The Scholion to Demosthenes c. Androt. 13, p. 597 (Ov. S. Q. 642, 646), goes back to the same source :
the artists' names only are omitted.
3 Imagines, 4.

DATE OF THE PROMACHOS 33
be regarded as necessary developments of the style of Pheidias. Only the tendency
to greater restlessness and animation and the lack of severe simplicity could out
weigh these considerations and make it possible to assign the figure to any other
than to the creator of the Parthenos.
Before discussing this question further, the date of the Promachos must be
fixed more exactly. The work, as we have seen, is probably later and certainly
not earlier than the Parthenos. If this be so, then the usual assumption that the
Promachos was a monument of the Kimonian period must be false. The assump
tion is indeed a pure conjecture which must be abandoned in face of more exact
knowledge. If the torso Medici is a copy of the Promachos, the Promachos can
not belong to the Kimonian epoch. K. Lange rightly drew the same conclusion.
From the earliest testimony concerning the Promachos, that of Demosthenes (Fats.
Leg. § 272), it follows with certainty, as I believe, that the statue, which he only calls
the large bronze Athena, was a votive gift for the Persian war. He describes it
as apicrrelov tov 7t/oo? tov? fiapfidpow; troXefiov, and the stele with the curse upon
Arthmios, who had brought the money of the great king, was set up beside the
Athena, according to Demosthenes, precisely because the Athena was the monument
of the Persian war. The further assertion that the statue was erected out of the
money contributed by the Hellenes is obviously mere rhetorical exaggeration and
inaccuracy.1 Now supposing that the Promachos was a votive gift for the Persian
war, there is still a large margin within which to date it. It has long been
acknowledged that the assertion made by the late authors, such as Pausanias,
Aristeides, and the Scholiast to Demosthenes, to the effect that the statue was a
votive gift from Marathon, cannot be relied on.2 Probably the dedicatory inscription
described the statue merely as a votive gift airb liepawv or d-rro MijSav. A shortened
inscription of this kind would be quite in the fifth-century manner,3 and would best
explain the comprehensive wording of Demosthenes.
The extant copies show the Promachos to be later than the Lemnia and the
Parthenos. The design for the latter must have been executed about 447 B.C. ; 4
and as we need allow only a short interval before the making of the Promachos
(which is earlier than the frieze and the pediments of the Parthenon), we thus
obtain the years 445 — 440 B.C. as the approximate date of this work. Now it
seems to me far more likely that the statue was erected about this time than
during the administration of Kimon.
The career of Kimon was one long conflict with the Persian power ; any
monument therefore celebrating the close of the national struggle would have been
unsuitable to the spirit of his time, for it was his aim rather to keep animosity
alive. Memorials of isolated exploits, such as the gilt Athena and the bronze palm-
tree dedicated at Delphi- after the battle on the Eurymedon,5 were appropriate, but
1 Otfried Miiller (De Phidiae Vita, i. § 10) suggests that there is also an allusion to the tribute of the allies.
Preller, Phidias, p. 165 (Hall. Encycl. iii. vol. 22), suggests the distribution of the booty after Plataia. But if
Demosthenes, as Wachsmuth (Stadt. Athen, i. 542) assumes, really means the money which was brought by
Arthmios, his statement is a purely rhetorical invention ; the general connexion of the statue with the Persian
wars was a well-known fact which Demosthenes takes for granted and works up into a rhetorical period.
2 So already by Otfried Miiller, De Phid. Vita, i. par. 9.— For the inscribed fragment C. I. A. i. 333 which
Kirchhoff referred to the Promachos, cf. Michaelis, Ath. Mitth. ii. 92, and Wachsmuth, Stadt. Athen, i. 542. It
probably comes from a smaller anathema erected in the lower city soon after the battle of Marathon.
3 Cf. especially the inscription of the golden shield in the temple of Delphi, 'kBitvdioi avb UiiSwv ko! &nfaluv
(Aeschin. in Ctesiph. 116), also the existing inscriptions, 'A8r)va!oi curb TleKoirovvqoiwv (Rbhl, Inscr. Ant. 5),
@oipioi airb TapaVT'tvwv (ib. 548 seq.), Me davioi airb AaKetiaifwvlaiv (ib. 46), etc.
4 Cf. U. KShler, Sitzungsber. d. Berl. Akad. 1889, p. 225.
6 Paus. x. 15, 4. F

34

PHEIDIAS

if we are to trust Demosthenes the Promachos was not an offering of this kind, and
the indirect testimony of the later witnesses confirms the account of Demosthenes,
for they would never have given Marathon as the occasion of the offering had the
inscription mentioned any other victory.
It was not until the death of Kimon that the foreign policy of Perikles com
pletely asserted itself. Its first aim was to bring the Persian war to a close so as to
concentrate every effort on the attainment of Athenian supremacy in Greece. Negotia
tions for peace were accordingly begun. Kallias brought the Athenian proposals
(a record of which was probably set up on the Akropolis) to Susa. This occurred
about 445 B.C. Although it is unlikely that the peace so triumphantly celebrated by
the orators of the fourth century ever took the shape of a formal treaty, yet the
practical result of the negotiations was a complete cessation of hostilities between
Persians and Greeks for a long period.1 It would naturally be to the interest of
the Periklean policy not to let the close of the Persian war pass unrecorded, and no
better means of commemorating it could be imagined than a colossal votive gift to
the patron goddess of Athens bearing some such inscription as 'Adtjvaioi atrb MtfSwv.2
This votive gift was, I believe, the Athena Promachos.
The period (445 — 440) to which, for stylistic reasons, we have assigned the statue
thus becomes probable on historic grounds also. Just such a gift is what we should
expect from Perikles, and certainly Pheidias from his relation to Perikles would have
some share in the work. Perhaps he made the first sketch, and handed over the
execution of the large model and the casting in bronze to one of his assistants —
in a word, to the Praxiteles named as the artist by a tradition which was probably
derived from the actual inscription on the statue. According to Pausanias, the reliefs
on the shield were made by Mys from drawings by Parrhasios. At least one ancient
inscription must have testified to the fact that these two artists worked together.
This inscription, which may actually have been on the shield of the Promachos, gave
rise at a later date to the epigram preserved in Athenaeus (p. 782 B).3 From what
we know of Parrhasios independently (the date of Mys depends on his) it is by no
means impossible that he was working as early as 440 B.C., and if so the reliefs
on the shield would be, as is most natural to suppose, contemporary with the
erection of the statue.4 The statement of Pausanias as to the reliefs not being
by Pheidias affords further proof that the statue was by another artist, for Pheidias
appears to have always executed the reliefs on his large works with his own hand,
and, having himself been a painter, to have dispensed with the aid of painters or
engravers. The main reason for the widespread belief that the Promachos belonged to the
period of Kimon lies in the assumption of a ' Kimonian Pheidias.' 5 The only evidence
however for the assumption is that of Pausanias, who says that the group dedicated
1 For the so-called ' Kimonian treaty ' cf. Busolt, Gr. Gesch. ii. 512 sqq. ; Holm, Gr. Gesch. ii. 201 sqq.; Curtius,
Gr. Gesch. 6th ed. ii. 183, 832; Stadtgesch. v. Athen, 205 ; Kbpp, in Rheinisches Museum f. Philol. vol. 48
(1893), p. 485. The date of the embassy of Kallias is to be found in Suidas, sub voc. KaWlas (in the year of
the invasion of Pleistoanax). Krateros incorporated the decree in his collection. The theory of Curtius,
Stadtgesch. 205, that at the beginning of the fourth century a fictitious inscription was cut in stone and set up on
the Akropolis, is scarcely tenable ; Isokrates at least in 380 B.C. considered the record genuine. [Cf. Grote, «.
195-1 2 Cf. supra, p. 33, note 3.
¦J Preger, Inscr. Gr. Metr. No. 185, dates the epigram either in the first century B.C. or in the first
century a.d.
4 Cf. Brunn, K. G. ii. 97 seq. He supposes that they are contemporary, although he places the Promachos in
the period of Kimon (i. 165).
5 See especially Loschcke, Tod des Phidias, p. 45, n. 1.

THE 'KIMONIAN' PERIOD OF PHEIDIAS 35
by the Athenians x in memory of Marathon at Delphi was by Pheidias. But this
statement, given only by Pausanias, must surely be wrong, even if we admit a Kimonian
Pheidias. On the bathron of the group there was an inscription to the effect that
it was a heKa.Tr, for the battle of Marathon ;2 hence the group must have been set up
very soon after the battle, at a time when Pheidias, even if we suppose him to have
been born as early as 500 B.C., could not possibly have been at work. Every trust
worthy record of a monument dedicated as a tenth after a victorious battle shows it
to have been set up immediately after the victory it commemorated.3 Only in cases
where the votive gift was a building it might take longer to complete it.4 The group
at Delphi would certainly be dedicated and carried out immediately after the victory.
In the days of Kimon there would scarcely be any occasion for dedicating a Se/caTr) for
Marathon, which had been thrown into the background by other important events.
Finally, the character of the work, a series rather than a group of separate figures
standing together without any indication of action, is more suited to the older period,
while the conception of Miltiades as the central figure surrounded by the patron
divinities of Athens, Athena and Apollo, and the heroes of Attica, is quite in the spirit
of the days when the name of Miltiades was in everybody's mouth.5 It was also
characteristic of that older period to place this large and magnificent bronze group in
Delphi instead of on the Akropolis. No doubt Kimon wished to pay personal honour
to his father's memory, but the customary way to do so was to set up some private
dvddrjfia or other work of art such as the historical painting dedicated in the Stoa
Poikile by Kimon's brother-in law.6 The group sent as a tithe to the god of
Delphi by the Athenians must have been made directly after the battle. When the
Lakedaimonians overthrew the power of Athens in B.C. 404 they set up in Delphi
without delay a group designed to be in every particular the counterpart and rival
of the one offered by the hated Athenians. Lysander, like Miltiades, formed
the central figure of a group of divinities. Strangely enough excess of pride
brought about the downfall of both leaders, and the greatness of Lysander, like that
1 It should be carefully noted that neither Pliny nor any other author says anything of this, the most
comprehensive work by Pheidias.
Paus. a. 10, I : TijJ fidOpcp . . . iniypa/x/xa p.iv iariv airb SeKa-rjjs tov Mapadcoviov epyov TtBijvai rds el/coVas.
Therefore Pausanias calls it aAi)de'i Xiyip SeKaTii -rrjj fidxns in order to distinguish it from the votive gifts which
were connected with the battle of Marathon not by an inscription but only by a legend.
3 For a list of public votive offerings connected with events in war, cf. Ziemann, De Anathematis Gr. ,
Kdnigsb. Dissert. 1885, p. 10 sqq.
* Yet the portico of the Athenians in Delphi, or example, was built immediately after the naval victory over
Aegina, probably in B.C. 488. Cf. Kohler, Rhein. Mus. 1891, p. I seq.
5 The Pinax by Epiktetos with MiAticSStjj koXus must be older according to vase chronology ; it may
refer to Miltiades as a irais or young man, but not as a victorious general in the prime of life ; in the same
way the 'Barbarian' statue published by Studniczka, Jahrb. d. Inst. 1891, p. 238, can have nothing to
do with Marathon. Not because the style is too archaic, for this objection is no longer valid after the
discovery of the metopes of the treasury of the Athenians in Delphi, but because it seems impossible
that this proud mounted Persian should have served as a monument of his country's defeats. Studniczka
compares the Mardonios which stood in front of the Persian porch at fc'parta, but Pausanias (iii. II, 3)
says that the statues of Persians, Mardonios and Artemisia among them, stood on the pillars eVI t&v kiovosv
of the porch ; they are usually, and correctly, supposed to have been supporting figures (Curtius, Arch. Ztg.
1881, 18, 20); they probably served as supports for the beams of the upper storey of the porch, and were
evidently decorative. Pausanias further implies that this decoration belonged to a later period. In the Persian
rider of the Akropolis I incline to see a votive gift belonging to the end of the period of the Peisistratidai,
when these rulers were trying to come into closer connexion with the Persian king. Against Studniczka's view
see also Hartwig, Meisterschalen, p. 10, and Percy Gardner, Ashmolean Vases, p. 31, No. 310.
6 The most probable hypothesis is that Peisianax erected the porch out of his own private means. Cf.
Wachsmuth, Stadt. Allien, ii. 501. If he had only been ' Vorsitzender der Baukommission,' as Robert assumes,
Hermes, xxv. 422, the stoa would never have been called by his name.

36 PHEIDIAS
of Miltiades, was already over a year after his victory. It follows that each monument
must have been made directly after the victory for which it was a thank-offering.1
It is not, however, surprising to find that in later times the Delphic group was
supposed to be by Pheidias. It was known to commemorate Marathon, and Pheidias
was held, at least by Pausanias and his informants, to be, by reason of the Promachos
in Athens, the maker of Marathon votive gifts kcit efo^K. We have shown above how
little Pausanias is to be relied upon where Pheidias is concerned. In three instances at
least Pausanias has ascribed to Pheidias works which were undoubtedly by other artists.
The Promachos is a fourth such instance, and the Athena at Pellene is presumably a
fifth, for the statement of Pausanias that this gold and ivory image was older than the
similar images by Pheidias on the Akropolis and at Plataia certainly arose 2 from the
archaic style of the image, and this simply means that the image was not by Pheidias,
though it was a gold and ivory statue of Athena. Thus we are led to be sceptical
with regard to the Athene of Plataia also (a marble statue in gold-ivory technique
which Pausanias again stands alone in ascribing to Pheidias, and again wrongly
associates with Marathon).8 According to a reliable authority the temple and the
image were erected with the eighty talents received by the Plataians after the battle.4
Probably the building was begun at once, and the ornamentation of the Pronaos with
frescoes by Polygnotos and Onasias would be left to the last.
We know therefore of no ascertained work of Pheidias older than the Lemnian
Athena, and the ' Kimonian ' period of Pheidias is a mere myth. As a fact, literary
tradition tells us only of the artist's relation to Perikles, and nothing of a relation to
Kimon. Not only did he first become famous through the patronage of Perikles, but
in an age where party opinion ran so high it is hardly likely that Pheidias, who, accord
ing to the most reliable tradition, was a close friend and, we may be sure, a political
adherent of Perikles, should have previously belonged to the party of Kimon.

VII. The Olympian Zeus. — Trial and Death of Pheidias.
The immediate and most important result of the previous investigations is that
the Zeus of Olympia cannot be earlier than the Parthenos. Since Pheidias had no
' Kimonian ' period at all, and since no certain work of his can be proved to be
earlier than the Lemnia, neither can his masterpiece, the Olympian Zeus, be placed
before 450 B.C. The Zeus would be unintelligible both on aesthetic and on historic
grounds at so early a period. If the Eleians intrusted to an Athenian artist the
most important commission that had ever" been in their gift, it must have been
because that artist had already won world-wide renown. Now Pheidias only won his
fame through the works intrusted to him and to his colleagues by Perikles. The
artistic achievements of Periklean Athens were, in fact, the conditions which
necessarily preceded the call of Pheidias to Olympia. In the year 456 there was
no Athenian art great enough to account for an Athenian artist being invited to
1 Fulvius Ursinus published a terminal bust (now lost) inscribed 'Miltiades.' The head, which possibly
does not belong to the term, can scarcely be referred to the Delphic offering, for it wears no helmet. Besides, it is
certainly not older than the Parthenos. The beautiful head (Louvre, No. 1608) infra PI. IV. which Visconli calls
Miltiades (Iconogr. Gr. PI. 13) is purely Pheidian in character, but it is impossible to name it exactly, for it
resembles the Ursinus head in style only. A lion is seen opposite the bull on the neckpiece of the helmet, therefore
the bull cannot be ' the bull of Marathon ' as Visconti supposed.
2 Klein, in Arch. Epigr. Mitt, aus Oesterr. 1883, 69.
3 Paus. ix. 4, I.
4 Plut. Aristeid. 20. Herodotos says nothing about it : cf. American Journ. of Arch. 1891, 400.

OLYMPIAN ZEUS AND THE TRIAL OF PHEIDIAS 37
make the temple statue itself, especially when we remember that the Lakedaimonians
had just been allowed to celebrate their victory over the Athenians by erecting an
akroterion over the completed temple. The only direct ancient tradition we have
about the date of the Olympian Zeus places it, as is well known, after the Parthenos.
Yet the theory that the Zeus is older than the Parthenos, and that the date of
Pheidias as given by Pliny (OI. 83=448 B.C.) refers to the unveiling of the Zeus,
is one of long standing. It was held by Winckelmann,1 though he afterwards
abandoned it,2 and Chr. G. Heyne attempted to prove it in detail.3 Lately Loschcke
revived the theory with great acuteness ; 4 and in spite of the lively opposition of
other scholars he has again quite recently defended his point of view.5
We shall take Loschcke's thesis for a point of departure : he maintains that
there were two ancient and contradictory traditions concerning the date of the
Olympian Zeus between which we have to choose. This view is, however, inexact :
we have only the one tradition mentioned above, for Plutarch's narrative, which,
according to Loschcke, affords ' convincing ' although ' indirect ' proof that the Zeus
was made before the Parthenos, is no evidence for this at all, either direct or indirect.
There is not a word in Plutarch to indicate that the trial of Pheidias, which he
relates in detail, took place immediately after the dedication of the Parthenos, so that
Pheidias would have no time to execute any other commission. On the contrary, the
whole context, combined with the express statement that the trial of Aspasia took
place about the same time, distinctly shows that Plutarch thought of the trial as
taking place just before the outbreak of the Peloponnesian war. Thus Plutarch's
narrative does not contradict the tradition that the Zeus was made immediately after
the completion of the Parthenos. It is not about the date of the Zeus that we have
two traditions ; evidence is only divided as to the date of the trial and death of
Pheidias. The narrative of Plutarch places the prosecution of Pheidias, as we have
seen, just before the war, and makes Pheidias die in prison at Athens after his con
demnation. On the other hand, the account preserved in the Scholia to Aristophanes
is based on the presumption that the trial with all its consequences occurred directly
after the dedication of the Parthenos, these consequences being that Pheidias
escaped from prison, made the Zeus in Olympia, was accused a second time by the
Eleians, and finally was put to death. It is between these two traditions that we
have to choose.
Philochoros, quoted by the Scholiast to Aristophanes,6 after giving the date of
the dedication of the Parthenos, adds some information about the fate of the artist,
but by using the expression XeyeTat seems to decline being responsible for its
veracity.7 The Scholiast accordingly takes it for granted that the events which
Philochoros relates concerning Pheidias, Ta irepl QeiBtav yevop,eva, i.e. from the trial
1 In the first edition of the Geschichte d. Kunst, 1764, p. 332.
2 Geschichte d. Kunst, book 9, cap. 2, § II (Werke, ed. Meyer and Schulze, vi. I, 39; ed.
Eiselein, v. 358).
3 Antiquarische Aufsdtze, i. (1778), p. 203 ; he dates the making of the Zeus 01. 81, I — 83.
4 Phidias Tod und die Chronologie des olympischen Zeus, in Hist. Untersuch. A. Schaefer, gewidmet.
5 Festschrift z. 50 Jdhr. Jubil. d. Vereins v. Alterthuiusfr. im Rheinland, 1S91, p. 16 sqq., where
the latest literature on the subject is referred to. Curtius, Griech. Gesch. 6th ed. ii. 884, should be added
to the list.
6 After the investigations of Scholl and von Wilamowitz it may be considered certain that the long and
trustworthy scholion from QiXoxopos to Xiyovai 8e rives is taken from Philochoros, and that nothing ought to
be subtracted from it. Cf. also Loschcke, loc. tit. 20 seq.
7 This may be concluded from the word kiyerai, although it does not directly prove the uncertainty of the
tradition.

38 PHEIDIAS
to the execution of the artist in Elis, took place in the same year as the dedication
of the Parthenos, to which they form the sequel in the narrative of Philochoros.
From this passage the Scholiast concludes that the misfortune of Pheidias happened
seven years before the Megarian Psephisma and the outbreak of the war, and
consequently could have had nothing to do with either. Now the Scholiast is
evidently in error. Philochoros assigns a date to the dedication of the Parthenos
only: what he adds about Pheidias, including the first prosecution, is undated,
for, as he certainly does not place the completion of the Zeus and the second
prosecution in the year of the dedication of the Parthenos, we need not assume
that he wished to assign that date to the first prosecution either. The fact is that he
dates neither prosecution, but merely makes the ultimate results of an event follow
immediately upon it, probably because he had nothing else to date them by.1
Plutarch's chronology of the prosecution is confirmed by testimony from other
quarters, and notably by Aristophanes himself. In the Peace (605), Hermes, with that
cunning of the townsman which can trace below each event the personal motive
hidden from the stupid peasants, Trygaios and the chorus, alludes to the calamity
of Pheidias, and to the fear felt by Perikles lest he should be involved in it, as being
the causes of the Megarian Psephisma and of the Peloponnesian war. The point of
the witticism would be lost had Pheidias died six years before the Psephisma, at a
time when Perikles stood at the zenith of his power without the slightest cause
for fear of any sort.2
On the other hand, as regards the death of Pheidias, we have the version reported
with reservations by Philochoros, and the version of Plutarch. It is now almost
universally acknowledged that the trial in Elis, ending with the execution, is only
a 'reflection' of the first trial.3 It is due to rhetorical inventiveness, working
upon the favourite theme of the world's ingratitude towards its great men and
the ill luck which befalls them. In order that the story of the second trial might
be coherent it was necessary to make Pheidias escape from the prison in Athens and
flee to Elis. Plutarch, who gained his information, as is generally admitted, from
the best sources, and from sources substantially older than Philochoros, ignores
that version. His statement that Pheidias died in prison in Athens is incomparably
more credible than the other story, and admirably fits in with the allusion in
Aristophanes to the misfortune of Pheidias, an allusion which would certainly not
be consistent with the escape from Athens and honourable reception at Elis.
Besides, no one would have dared to invent the story of the second trial, which
must be at least as old as the third century B.C., if the Olympian Zeus had not been
confessedly a more recent work than the Parthenos. Thus the only direct testimony
to the date of the Zeus remains uncontroverted.
We assume, then, relying solely on the best-accredited tradition, that Pheidias
began the Zeus immediately after the Parthenos in 438, and finished it for the
eighty-seventh Olympic festival ; thereupon followed the trial and death in prison.
There is nothing improbable in this assumption. If Pheidias could complete the
Zeus in six years, though he required nine for the Parthenos, it was probably
because, in making the Zeus, he had two assistants, Kolotes and Panainos. Nor
is it necessary to suppose that Pheidias was at Olympia all the time : when once
the design and the models were made, the master's presence was not always
1 This is a common proceeding of Philochoros ; cf. Bockh, Plan d. Atthis d. Philochoros (Abh. Berl.
Alad. 1832), p. 7.
2 Cf. Nissen, Histor. Ztschr. N.F. xxvii. 1889, 406. ' Cf. Loschcke, loc. cit. 21.

PANTARKES 39
indispensable ; in the great workshop or ergasterion 1 numberless workmen would
be put in charge of the several parts ; and the assistants of Pheidias would suffice
to superintend the work. The master himself must have had ample time for visiting
his home in Athens, and for setting other tasks on foot. Thus I consider it not
only possible but probable that, even after the year 438, he continued to superintend
the Periklean works, and more particularly the completion of the Parthenon.2 Not
till this temple was quite finished and the position of Perikles began to be insecure
did the accusation of embezzlement, the particulars of which we do not know,3
take definite shape against Pheidias. His death in prison followed upon the trial.
It was a most natural resolve on the part of the surviving members of his family
to abandon their home and settle in Elis. There the memory of the great artist
was held in reverence, and we have evidence, at least in later times, that his
descendants were invested with the honourable office of (paiUpwiai 4 of the image
of Zeus. This chronology of the Zeus fits in with everything else we know about
the statue, and first with the legend of Pantarkes, which, when its tangled threads
are once unravelled, yields the following facts. The story must of course have arisen
from some definite circumstance. This was the inscription on the finger of Zeus,5
in real fifth-century Attic fashion, of the 'love-name' JlavTdpKrji; Ka\6<;. We
cannot be sure whether the inscription was from the hand of Pheidias, as was believed
in antiquity, or from one of his colleagues or assistants, but it seems at least likely
that the Pantarkes of the inscription was identical, as the ancients also believed,
with the Eleian boy Pantarkes who was victorious at Olympia in 436 B.C., and
whose statue stood in the Altis (Paus. vi. 10, 6). We know absolutely nothing
about the artist or the motive of this statue. On the other hand, the ciceroni of
Olympia were merely inventing when they told Pausanias that a likeness to Pan
tarkes, the favourite of Pheidias, was to be seen in the face of a youth binding a
fillet6 about his head which adorned the throne of Zeus. It is only in modern
times that the statue of an Anadumenos, unnamed, which Pheidias made and set
1 In the time of Pausanias this was still shown ; it may probably be identified with the long
narrow building immediately south of the Byzantine church. At Epidauros too, as the inscription 'EtpTi/x. dpx-
1886, 145 sqq. , shows, was a carefully built ipyatrrrfpiov for the temple, provided with Koviaais. The building on
the Akropolis south of the Parthenon may have been the workshop of Pheidias. There is no reason to doubt
the Ergasterion in Olympia as Robert does (Hermes, xxiii. 453).
2 In the year 433 — 432 the works on the Parthenon were still going on ; in 434 begins the inventory of the
contents of the interior, which must then have been complete. Cf. Foucart, B. C. H. xiii. \"]s,sqq. It will be
remembered that Plutarch reckons the Propylaia, which were erected between 437 and 432, among the buildings
under the superintendence of Pheidias.
3 The reasons given by different authorities are at variance. Scholl has shown that Plutarch's stories
about the gold used for the Parthenos and the portraits on the shield are mere inventions ; what Philochoros
says about the ivory, although it sounds more probable, is assuredly another invention ; Diodoros, who
relies on Ephoros, only says ttoXkb. ran UpZv xpvit*Ta"/ > it is quite evident that the ancients had not
mastered the numerous and complicated details of the accusation ; Ephoros gives the general drift, and his
account is probably correct.
4 Paus. v. 14, 5.
D Cf. Robert in Hermes, xxiii. 447 ; von Wilamowitz, Comment. Gramm. iv. 16, 1. The evidence is
reliable and in itself quite credible ; the cathedrals of the middle ages afford examples of far greater licence.
Clement of Alexandria, who usually derives his information from the best sources, has the story, and it is quoted as
a well-known fact in the passage in Photios (which is probably derived from Polemon) about the Nemesis of
Agorakritos. Pantarkes is here called 'kpye'ios, but that is probably only a mistake of the person who made the
extract, or perhaps of the copyist. The correct word is probably 'HAeTos. There is no trace here
of any ancient controversy about the native place of Pantarkes, such as has been imagined. The
testimony of Gregorius Nazianzenus and of Libanius, who connect the inscription with other statues, is
naturally valueless.
6 Paus. v. II, 3 : ioiK^vat rb elSos Tlavrdptiti \iyovffiv.

40

PHEIDIAS

up in the Altis, has been wrongly brought into connexion with this question. It
is a separate work, and has nothing to do with Pantarkes.1 The presence of the
'love-name' is best explained by assuming that the Zeus was still unfinished in
436, for it is extremely unlikely that such an inscription was added when the
statue was once complete and dedicated for worship.
More important than the Pantarkes question are the observations made by
Dorpfeld on the temple of Zeus and on the Parthenon.2 A piece of very decisive
evidence is afforded by the fact that the basis of the Zeus, as well as the pavement
immediately in front of it, were made of black Eleusinian stone. This was well
calculated to set off the surrounding pavement of white Pentelic marble and the
golden reliefs of the basis itself. Dorpfeld lays stress on the circumstance that
Eleusiriian stone was employed in a similar way in the construction of the Propylaia
and of the Erechtheion, but not in any older buildings, and notably not in the Parthenon,
He is no doubt right in considering the square space in front of the Zeus to be a
conscious imitation of the similar larger space in front of the Parthenos. The further
observations that the Zeus and the Parthenos stood at the same distance from the
cella door, that the cella at Olympia had to be altered before the image could be
set up, and that the Zeus was too large in proportion to the cella in which it stood,
while the Athena completely harmonized with its surroundings, in themselves neither
prove nor disprove the priority of one or the other image. But it seems most
natural to suppose that Pheidias imposed upon Olympia the conception which had
belonged by right to the Parthenon, and that the Eleians submitted to such an
alteration in their building at the hands of the celebrated artist of the Parthenos,
because they were anxious to possess a statue as magnificent as the one in Athens.
In the year 456, immediately after the completion of the temple, the date to which
Loschcke assigns the beginning of the Zeus, all this would be quite unintelligible.
Loschcke, however, thinks it strange that the temple should have been left
eighteen years without an image. True, if temple and image were planned at the
same time, as in the case of the Parthenos, they would naturally have been begun
together, but if this was not so there is nothing to fix one date for the image rather
than another. We cannot tell whether it was originally intended to set up a new
cultus image, or an image at all, in the temple, In the temple at Delphi, for instance,
which was built about the same time, no cultus image was erected so far as we know.
Loschcke further lays stress on the circumstance that Pliny, in the very passage
where he mentions a joint work of Panainos and Kolotes in Elis (xxxv. 54), assigns
Panainos to the same Olympiad as Pheidias (01. 83). This date Loschcke concludes
to be that of the completion of the Zeus. To my mind it is evident that Pliny, owing
to the paucity of chronological material at his disposal, is dating Panainos not
independently but merely from his brother Pheidias. He was only in possession of
the dates of a few leading artists, about whom he grouped lesser artists, according to
the traditions of their connexion with one another. The whole passage containing
the date of Panainos is one of Pliny's most characteristic and arbitrary pieces of
1 Paus. vi. 4, 5. The usual identification of the avaSov/xevos of Pheidias with the victor statue of
Pantarkes, of which we know neither the artist nor the motive, must be rejected. Pausanias says—
and we have no reason to doubt his word— that the statues stood in different parts of the Altis. The
dvaSov/xevos was a votive gift which only bore the name of the artist (cf. Gurlitt, Pausanias, p. 378 sqq. ) ;
the name of the person represented was unknown, as Pausanias expressly says (cf. Schubart's translation).
I shall attempt later to point to a copy of this Anadumenos. Cf. infra, p. 244. The three monuments—
viz. the victorious athlete, the Anadumenos, and the figure on the throne— are all quite correctly distinguished by
Dorpfeld, Olympia, Baudenkmdler, Textbd. ii. 21.
2 Olympia, Baudenim. Textbd. ii. 16, 20.

DATE OF THE OLYMPIAN ZEUS 41
chronology-making.1 It seems evident from a later passage2 that he found the
date of Pheidias ready to hand, and invented that of Panainos.
Loschcke finds confirmation for the early date to which he assigns Panainos
and the Zeus in Robert's hypothesis that the paintings of the Stoa Poikile were
finished about 460, and that the battle of Oinoe, there represented, happened about
the same time.3 But valid objections to this theory have been brought forward
by Judeich,4 and the earlier view, the evidence for which has been carefully
collected by Wachsmuth5 — viz. that the picture of the battle of Oinoe did not
belong to the original cycle of the Polygnotan paintings — is still the prevailing one.
This picture was not painted, as the others were, on the large central wall, the
name of its artist is not mentioned, nor did it enjoy the same celebrity as the other
three. The group by Hypatodoros and Aristogeiton (Paus. x. 4) in Delphi must
assuredly, as Robert assumes, have been an older work, and the battle of Oinoe, with
which Pausanias connects it, is certainly the same that formed the subject of the
picture in the Stoa Poikile. Whether, however, Pausanias is right in connecting
the battle with the Delphic group is another question. e Again, according to Robert's
hypothesis, the Stoa was erected about the time of Kimon's exile, and was intended
to celebrate the victory over the Lakedaimonians, who favoured the Kimonian
faction. This seems highly improbable, seeing that Peisianax, Kimon's brother-
in-law, dedicated the porch. It is more likely that the building was undertaken after
Kimon's return, i.e. after 457, and that in true Kimonian spirit it was designed to
celebrate the great exploits of Attic heroes against the barbarian Amazons, Trojans,
and Persians. We are not even sure that Panainos was employed on the Stoa at all.
It is true that Pliny and Pausanias call him the painter of the battle of Marathon,
but in other authorities the same picture is ascribed not only to Polygnotos
(which would mean little, considering the fame of Polygnotos), but also to Mikon.7
Even if Panainos was engaged upon the Stoa that is no reason why he should
not have worked about 456 — 450 in Athens, and about 438 — 432 in Olympia
and Elis.
Finally, Loschcke calls attention to some purely aesthetic reasons for dating
the Zeus in the older period. Of the ' archaic elements in the form and arrangement
of the beard ' which he sees in the copies on coins I can discover no trace.
The coins seem to me to show that the beard was similar to that of the seated
Poseidon on the Parthenon frieze. The line of the profile is decidedly against
Loschcke' s theory. In all the copies it is very straight, corresponding to the profiles
seen on the Parthenon frieze, so that the Zeus must be later than the Lemnia, whose
nose, as we have seen, forms an angle with the forehead.
1 Cf. my Essay in ix. Suppl. Bd. of Fleckeisen's Jahrb., ' Plinius und seine Quellen,' p. 16 sqq. ; and Robert,
Arch. Mdrchen, p. 25.
" Plin. 36, 15 . . picluram aut statuariam, qtiarum utraque cum Phidia coepit octogensima tertia
olympiade ; as regards painting, this refers to the passage on Panainos in 35, 54. Cf. Robert, loc. cit.
3 Robert, Hermes, xxv. 1890, 412.
4 Fleckeisen's Jahrb. 1890, 757.
B Stadt. Athen, ii. 502 sqq. Cf. also Benndorf, Jahrb. d. Samml. d. Kaiscrh. xi. 1, 22.
6 Probably the inscription, as usually in the fifth century, gave no more exact information. The words as
aural 'Apyeioi \iyovtrtv, which Robert would refer to an inscription in contradistinction to the following i/xo\ Sokiiv,
recall the exactly analogous ifiol SoKtiv and Meinrhvioi aino\ \4yovai in the passage which discusses the occasion
of the dedication of the Nike of Paionios, where the inscription on the statue says nothing about it. In the case
of the Argive votive gift we have also a mere hypothesis, and a very improbable one, on the part of Pausanias,
possibly suggested to him perhaps by the picture in the Stoa Poikile.
7 Probably as early as by Lykurgus in the speech irepl Trjs lepeios (cf. Brunn, K. G. ii. 19), then by Aelian,
Arrian, and Sopatros. G

42 PHEIDIAS
In a head of the Louvre, the so-called Zeus Talleyrand,1 Loschcke sees a copy
of a work belonging to the middle of the fifth century B.C., and evidently influenced
by the Pheidian Zeus. This is a daring assertion, depending on premises which will
not bear examination. It is true that many works which were formerly called
' eclectic ' have now been proved to be genuine copies of originals of the severe
transitional period of the fifth century, but it would be a strange perversion of logic
to proceed to deny in toto the existence of an archaic art containing mixed
elements of style, and to insist on calling every production which shows a touch
of the archaic manner a faithful copy of an earlier original. The mixed style
of the archaic period is an historical fact demonstrable in hundreds of monu
ments. The Zeus Talleyrand is a good instance. On the one hand the archaic
pointed beard which follows the convention of the sixth century, and on the
other the straight profile and the modelling of the eyes, form a mixture of hetero
geneous elements quite inconceivable at a period when art was developing in a
natural and healthy manner.2 The hair is drawn up above the ears into a kind of
loop, and then falls again in a broad band,3 according to a fashion much affected
by archaic artists.4
The work, which is imperfectly reproduced in the Talleyrand head, with the
admixture of later elements, certainly does not belong to the middle of the fifth
century, but is about half a century earlier. This is proved by a small statue in the
Museo Torlonia, a poor but faithful copy of an ancient work of art 5 representing
Dionysos wearing a short cloak, which falls in purely archaic folds, and a panther-
skin over it. The head is remarkably like the Talleyrand Zeus, except that the
affected modernizing element is wanting. The whole character of the statue and some
details, such as the roll of hair and the position of the body, which rests on the
left leg, while the right is only slightly bent,6 make it reasonable to suppose that this
figure is an early work from the cycle of Hagelaidas.7 So too the Zeus, as far as it
contains real archaic elements, must be referred to an Argive work of about 500 B.C.
It has nothing whatever to do with Pheidias.
The aesthetic reasons which Loschcke gives for his date of the Zeus are
thus demonstrably false. In addition to the straight profile, the mode of dressing the
hair also affords strong evidence for the later date. Numerous vase-paintings help us to
an exact knowledge of the development of the fifth-century Attic type of Zeus. In
the older style, up to about 450 B.C., Zeus has long hair, either hanging down over his
back or taken up behind. The Zeus of Olympia, according to the coins, had shorter
hair, falling in curls only as far as the shoulders. This mode of wearing the hair is
not found in the Zeus of vase-paintings till after 450 B.C., when the free style had
been perfected ; 8 in vases of the ornate style about 430 B.C.,9 the short curly hair,
1 The doubts which have recently been expressed concerning the genuineness of this head are completely
without foundation ; the head is assuredly ancient. [The head has lately been removed from the public galleries
to the Magasin. — E. S.]
2 Cf. Friederichs, Bausteine, 60 (ed. Wolters, 449) ; Kekule in Arch. Ztg. 1874, p. 98.
3 This is so stiffly worked that it might easily be mistaken for part of a diadem. This is also the case with
the hair in front of the diadem.
4 Numerous instances may be found on the so-called Campana terra-cotta reliefs (cf. the Gorgoneia and the
archaic Perseus, which exist in several variants, good examples in Bull. Nap. N. S. I, 5).
6 Mus. Tori. Taf. 124, No. 484. Cast in the Museum at Carlsruhe.
6 The right foot is wrongly restored as if drawn back.
7 Cf. 50th Berl. Winckelmannsprogr. p. 125 sqq. Studniczka refers to the statuette, and suggests Kalamis (Rom.
Mitth. 1888, 299, note 74).
8 One of the earliest examples is a kylix to be dated about 440 B.C. Mon. d. Inst. v. 48.
9 Style of Meidias and Aristophanes ; for the Io-vase see Overbeck, Atlas z. K. M. 7, 16, etc.

RELIEFS ON THE BASIS OF THE ZEUS 43
falling only to the nape, soon becomes typical of Zeus. These facts alone show the
Zeus to have been made after and not before 450 B.C.
The Pheidian Parthenos still wears the long curls, falling on the breast, of archaic
art : now, if the Zeus were earlier than the Parthenos, we should expect to find him
represented with the same long thick hair. It may be urged, on the other hand, that
the curly hair, lying close to the neck, as shown on the coin, begins to appear in works
of art just about the year 450.1 This is quite true, but it does not hold for Zeus,
in whose case the adoption of this fashion is of later date.
Except the coins there are no direct copies of the Zeus in existence. But it
would be very strange if no copies had ever been made at least of separate parts of a
work which comprised, besides the principal figure, so many reliefs and statuettes.
For instance, it is highly probable that the silver-gilt relief from Galaxidi, the
design of which corresponds exactly with the central group of the gold relief on the
basis of the Zeus, as described by Pausanias, is really a copy of this group.2 It has
recently been pointed out 3 that the main features in the design of the silver relief
(features which doubtless were to be seen in the work of Pheidias), such as the goddess
half emerging from the waves, the attendant divinity who raises her, the lifted head of the
goddess, her flowing hair, and the drapery held in front of her, are all borrowed from
an older work. But the composition has been remodelled in the spirit of the freest
style, with a tendency to vigorous, almost passionate movement and to picturesque
detail. It is the very manner — even to the rendering of the waves — which we admire
in the pediments of the Parthenon.
Another of the designs on the throne of Zeus — Apollo and Artemis slaying the,
Niobids — can, I think, be traced with tolerable certainty. Several reliefs of Roman
date have preserved to us in varying arrangement extracts from a once famous and
surpassingly beautiful representation of this subject,4 which must be based upon an Attic
original of the Pheidian epoch. That this is so is plain from the dress of the women,5
from the schemes of the drapery,6 from the style of head and body in one well-copied
figure,7 and finally from the character of the motives.8 It further seems probable that
the original was by Pheidias himself, for the bold and individual design of one of the
1 Cf. the two Apollo heads, Rom. Mitth. 1891, Taf. II, 12 ; Petersen, p. 379 ; cf. infra, p. 49 sqq.
2 Gazette Arch. 1879, PI. 19, 2 ; p. 171 sqq. Roscher's Lexikon, i. 1356.
3 By Petersen in Rom. Mitth. 1892, p. 49 sqq.
4 They have lately been collected by Hauser (Die Neuattischen Reliefs, p. 73 sqq. No. 104 — 107 b). Strangely
enough, even in this new discussion of the subject the most random and unfounded doubts are expressed about
the genuineness of the reliefs. I examined the St. Petersburg relief in the original not long ago ; it is proved to
be genuine by the remains of real incrustration on the marble. The relief is not complete ; it consists of two
pieces, while the connecting piece between the group of the two girls and the figure hastening to the right is a
modern restoration. The work is hard, and late in style. The disc in the British Museum is likewise
absolutely genuine, and is also partly covered with incrustation ; an ocular examination of it is sufficient to dis
prove the extraordinary arguments against its genuineness lately adduced by Overbeck, Sdchsische Berichte, 1893,
p. 58 seq. 5 The peplos girt low down with or without kolpos is characteristic.
6 The rendering of the folds of the drapery in the less agitated maidens is specially characteristic.
With the cloaks of the youths and of the Apollo (the best replica of the Apollo is the fragment Hauser 107 b)
cf. the mantles on the later Parthenon metopes, and those of the ' Kephissos ' and of the Kekrops in the
west pediment.
7 The Apollo on the above fragment, which though certainly not finely worked is stylistically faithful.
The nearest parallels to the head are found in the frieze of the Parthenon (e.g. north frieze, figs. 114, 115, Mich.)
Cf. also the Theseus head recently brought into this connexion by Helbig (Mon. Ant. dei Lincei, i. 4, 1892).
8 Hauser has already pointed out that the nearest analogies are afforded by the monuments in the style of
Polygnotos. The group of the dying boy, whom his elder sister has caught in her arms, is of peculiar beauty :
he clings round, her neck and greets her for the last time in death. The group belongs to a series of
Polygnotan creations (cf. 50th Berl. Winckelmannsprogr. p. 160 seq.), andi is one of the most marvellous
inventions of antiquity in existence. It was frequently copied in ancient times, especially on gems, and was

44

PHEIDIAS

figures corresponds to a figure on the shield of the Parthenos. It is that of a Niobid
who has fallen backwards over a rock. The head is down, the hands clasped behind
it, while the legs, bent at the knees, still seem clinging to the rock above (Fig. 7). Just
the same motive is introduced on the shield, but here the figure is that of an Amazon

Fig. 7.— Niobid from the disc in the British Museum. (From a drawing.)
dressed in a short chiton. Now we know that on the throne of Zeus, presumably
along the sides of the seat, were represented Artemis and Apollo slaying the Niobids ;
we may therefore venture to recognize in the reliefs, whose conception we have seen
to be Pheidian, echoes of those compositions. As we only possess fragmentary portions
of the original design, the arrangement of the separate parts remains uncertain. We
may be sure, however, that the composition was a frieze, as the divinities are aiming
straight in front of them.1 If our conclusions be correct, they afford a fresh proof that
the Zeus was not a work of the severe style, for the Niobid composition has a markedly
pictorial character, and its motives are full of consummate freedom and beauty.
The authentic copies of the reliefs on the shield of the Parthenos show the
same pictorial character, and the motives are equally free and daring. This relief of
the Amazonomachia is a priceless document for our knowledge of Pheidias ; it proves
beyond a doubt that it is not, as some have supposed,2 vase-paintings of the severe
style which will help us to recover the style of Pheidias. The designs of the shield
recall a distinct class of vases, distinguished by a free and animated style, and an
essentially pictorial quality, which Polygnotos was apparently the first to introduce.3
On these vases as on the shield the figures stand on undulating ground either above
one another or side by side. The same vigorous action characterizes the figures on
both — the long strides, the raising of the foot on an elevation, the fall, the headlong
plunge, the brandishing of weapons, the interchange of back and front view — all these
are to be found on the vases as on the shield. We have here a fixed standpoint
adapted for other figures in different designs. Cf. the Attic vase (Millin, v. p. ii. 49), which may be dated as early
as 430 B.C., and the famous Semele mirror (Friederichs, Kleine Kunst, 36).
1 Heydemann rightly emphasized this point. The arrangement on the disc cannot possibly be the right one,
for the divinities are shooting into the air. Note also that the female figure which was intended to be running to
the right has been placed as if lying on the ground. 2 Puchstein, Jahrb. d. Inst. 1890, 112 — 116.
3 Cf. Samm/ung Sabourojf, i. Vasen, Einleit. p. 5; also Robert, Die Nekyia des Polygnot, p. 42 sqq. Winter
(D. Jiingeren Attischen Vasen, p. 36) rightly saw in the Aryballos Fiorelli from Cuma (Not. dei Vasi Cumani, 8) the
direct influence of the Parthenos shield. To the same series belong the ' Giant ' vase of Melos ( Wiener
Vorlegebl. Ser. viii. 7), the Aristophanes kylix in Berlin ('EtpTifi. dpx- 1883, 7), and the Xenophantos vase (Compte
Rendu, 1866, 4) ; also Millin, Vases Feints, i. 56, 61 ; Naples, 3251 (Heydemann). All these are purely Attic vases
of about 430 B.C.

BASES OF THE PARTHENOS AND OF THE ZEUS 45
whence to meet all attempts to push back the art of Pheidias to an antiquated stage
which it had long left behind. True, Pheidias could represent calm and religious
repose in a masterly manner when it was fitting to do so. The assembled gods who
assist at the birth of Pandora on the basis of the Parthenos were naturally conceived
of in repose, as is shown by the copies.1 The reason lies partly in the nature of the
subject, partly in that of the place which the design was to occupy ; the ornamentation
on the basis of a religious image must have a certain solemnity. Even on the bases of
the Parthenos and of the Zeus, however, Pheidias did not limit himself to the represen
tation of divinities calmly looking on, but he bounded his composition on the one side
with the animated figure of the rising Helios, on the other with that of the sinking Selene.
The movement of the Helios on the Parthenos basis was, as the existing copy shows,2
of the fiercest, and the rearing horses were led by a figure whose impetuous stride
recalls the attitude of the warriors on the shield.
The cosmical frame thus formed by the sun and moon is a purely pictorial inven
tion,3 which, if it be due to Pheidias, is due to Pheidias the painter. Phidian ipsum
initio pictorem fuisse traditur clipeumque Athenis ab eo pictum, says Pliny. Pheidias
began life as a painter ; the shield to which Pliny alludes was presumably that of the
Parthenos, on the inner side of which the Gigantomachia was probably not in relief,
but painted.4 One vase-picture would seem to afford a conception of this painting,5
which in its main lines apparently resembled the Amazonomachia. On the vase,
however, the vault of heaven over-arches the whole, while the rising Helios and the set
ting Selene again frame the composition. There are a few more Attic vase-paintings
in which this same idea has been made use of.6 To judge from their style they all
belong to the period about 430, and are composed in the Polygnoto-Pheidian manner.
At this point it becomes possible to answer more definitely the question of what
was the share of Pheidias in the execution of the sculptures of the Parthenon.
The metopes of the temple have been admitted to differ greatly in style,7 and
1 Cf. Puchstein, Jahrb. d. Inst. 1890, 114; also Hub. Schmidt, Diss. Philol. Hal. xii. 131 sqq. In the
Lenormant statuette, the three figures in the centre are intended to give an indication of the peaceful assembly of
the gods, to which Helios and Selene form a kind of setting (cf. Schreiber, Parthenos, p. 57). This inter
pretation should never have been doubted, and is now confirmed by the Pergamene basis, which also only gives
an extract from the original.
2 On the basis of the Lenormant statuette. 3 Cf. Samml. Sabotiroff, Text to Plate 63, p. 2.
4 It is true that Pliny reckons (36, 18) the Gigantomachia among the other ornaments in relief of the
Parthenos, but this is a rhetorical passage (cf. Robert, Arch. Mdrchen, p. 24) ; on the inside of the shield ornament
in relief would be unsuitable, and finally the analogy of the Athena by Kolotes in Elis seems to point to painting.
The shield at Athens which Phidias painted can hardly be other than that of the Parthenos (as Urlichs, Chrestom.
Plin., already pointed out ; Robert, loc. cit., agrees with him). It is worthy of note that shields with painted
ornamentation inside only occur on those vase-paintings which are nearly analogous to the pictorial orna
mentation of the Parthenos shield and to its style ; cf. for example the ' Giant ' vase of Melos, also 'E<pi)fx.
apx- 1883, 7, and Overbeck, Gall. Her. Bildw. Taf. 11, I (where the Athena seems to be influenced by the
Fromachos), 5 Monum. d. Inst. ix. 6 ; Overbeck, Atlas z. K. Myth. Taf. 5, 8 ; cf. Robert, Arch. Ztg. 1884, 47 ; Kuhnerdt
in Roscher's Lexikon, i. 1659 ; M. Mayer, Giganten und Titanen, p. 268, 353 ; Petersen, Rom. Mitth. 1893,
p. 231. 6 The Judgment of Paris is framed in by Helios setting and Selene riding away and looking back : Wiener
Vorlegebl. Ser. E. II ; by Helios alone, on the hydria, Overbeck, Gall. 11, I. The setting Helios on the
Delphic pediment and on the vase (Compte Rendu, i860, 3) is probably influenced by Pheidias.
7 The supposition that the metopes of older style had been made for the ' Kimonian ' Parthenon (Kekule,
Weibliche Gewandstatue aus dem Werkstatt der Parthenongiebelfiguren, p. 22) is inadmissible, because the general
uniformity of the metopes, in spite of isolated variations, forbids referring any of them back to so remote a
period. Between the older Parthenon and the Parthenon of Perikles a long interval must have elapsed (see
below). The discovery of the treasury of the Athenians in Delphi can teach us what metopes were like at the
date of the first Parthenon. Besides, it is most improbable that there were ever any finished metopes belonging
to a temple which never rose beyond the lowest drums of the columns.

46 PHEIDIAS
clearly show the influence of various persons, while frieze and pediments are homo
geneous throughout, expressing the spirit and manifesting the methods of a single
artist. If we except a few instances of a transitional and less individual style, all the
metopes exhibit two conflicting tendencies. The one, represented only by a small
number of examples, is to harshness and angularity, both in bodily form and in move
ment, only a very feeble and clumsy use being made of drapery. This style resembles
that of the metopes on the so-called Theseion 1 and of the frieze on the temple at
Sunium.2 It is the direct successor of the style known to us from the ' Tyrant-
slayers ' by Kritios and Nesiotes,3 and shows little essential advance on those masters.
The other tendency is to greater softness, roundness, and flow in the forms of the body
(the abdomen more particularly is quite different) ; with it motion becomes incom
parably more vigorous and fiery, and the treatment of drapery richer, with a certain
leaning to the pictorial. The rendering of the woollen stuff is at the same stage as
in the Lemnia and Parthenos, i.e. it is heavier and more solid than in the frieze and
pediments ; where linen occurs it is sharply differentiated from wool, and is treated as
in the Promachos, not as in the frieze and pediments. Hence the metopes, as seems
natural from the history of the building, must have been made at the same time as
the Parthenos. Now the relief on the shield of the Parthenos proves that of the two
manners noted on the metopes, only the later and more pictorial (where the swing
and rhythm of the attitudes correspond to the reliefs of the shield) can belong to
Pheidias ; compare for example the figure rushing forward on the metope South VII.
or the fallen man South XXVIII. with the analogous figures on the shield. We may
conjecture that a small number of the earlier metopes were intrusted to pupils of
Kritios, and that afterwards Pheidias, who had at first been more exclusively occupied
with the Parthenos, turned his attention to this department of the work also, when the
other artists employed on the metopes tried to suit their style to his.
The frieze and the pediments, which, except for some trifling inequalities in
execution, may be considered uniform, must from stylistic reasons be later than the
metopes, while the pediments are somewhat later than the frieze. The frieze
must certainly have been finished before the last roofing was put on, and this
was probably done before the image was set up in 438 B.C.4 The figures of the
pediments, which were completed in the round below, and fixed in their places
afterwards, must have been made after 438 B.C.5 We have seen that Pheidias may
1 I cannot consider the Theseion to be later than the Parthenon : I hold it to be just a little earlier. The
arguments of Julius, Ann. d. Inst. 1878, 205 sqq., have not been refuted. The counter argument brought forward
by Dorpfeld, Athen. Mitth. 1884, 336, is not convincing. If the frieze on the cella of the Parthenon owes its
existence to an alteration in the plan of construction, as Dorpfeld plausibly argues, it cannot be placed at the
beginning of a new development. Dorpfeld could only urge this if the frieze with its regulae were part of the
original design (as was formerly held), for only in that case would the Doric regula be a sign of earlier date.
The frieze on the Theseion corresponds in style to the Parthenon metopes of the second manner. The fact that
the two styles occur together on the Parthenos shows that it is not necessary to suppose that the metopes of the
Theseion were taken from an older building.
2 Athen. ^ Mitth. 1881, 233 (Lange) ; 1882, 396(Furtw.) ; 1884, 338 (Fabricius) ; cf. 1884, 336 (Dorpfeld). It
is quite certain that the style corresponds as closely to the metopes of the Theseion as it differs essentially from
the frieze of the same building. If the architectural forms of the marble temple at Sunium do not admit of being
dated at about 450 B.C., the frieze must be assigned to the older structure of poros stone.
3 In addition to the head, abdomen and breast are specially characteristic. I can detect nothing peculiarly
Myronian in these works ; rather do they recall Kritios.
4 Dorpfeld has communicated to me his opinion that the frieze was finished about the same time as the cultus-
image, though he believes that the question cannot be decisively settled.
5 Two fragments most probably belonging to the building records of the Parthenon contain the accounts
which refer to the purchase of marble blocks for the pedimental figures (CIA. iv. 297 a.b.) Loschcke ( Tod des
Phidias, p. 46) pointed out that, as in one fragment mention is also made of a purchase of wood, the account is
probably to be dated after the making of the roof, i.e. after 438 B.C.

THE PARTHENON SCULPTURES 47
have retained the superintendence of the building after 438. There is not the smallest
reason for disputing the tradition which makes him overseer of all the works or
for denying his authorship of its principal sculptural decorations, i.e. of the frieze
and tlie pediments. The aesthetic quality of the work is all in favour of the tradition :
the fiery attitudes of some of the pedimental figures, such as the striding Poseidon
and the Hermes, can only be matched on the shield and basis of the Parthenos ;
the pictorial touches, which in the pediments find expression in the rocky seats
and the waves of Okeanos, are paralleled by the wavy ground lines on the shield
relief, while the rising Helios and the sinking Night of the east pediment have
been shown to be genuine Pheidian conceptions, which could be introduced in a
shortened form because the artist had expressed them fully elsewhere.1 In the
rendering of form, and especially of drapery, the frieze and the pediments are the
immediate outgrowth and development of the metopes of freer style. Compare
for example the mantle of the recumbent Greek on metope South XXVIII. with
that of the ' Theseus ' or the ' Kephissos ' of the pediments. The style is absolutely
the same, though equally characteristic in both, the only difference being that the
folds in front of the pedimental figures — not at the back, where the correspondence
with the metopes is absolute — are cut more deeply in order to make the drapery
look lighter. To ascribe the metopes to the artist and deny him the frieze and the
pediments means a refusal to acknowledge that he was capable of development
at all. Now the existence of the Parthenos and its relation to the Lemnia afford
irrefragable proof that Pheidias was by no means an artist who. remained wedded
to a fixed manner, but one who developed decidedly and rapidly. The Parthenos
differs from the Lemnia, as we saw, in the treatment of drapery, the artist showing
a tendency to abandon simple characterization of woollen stuff and to give more
attention to the effective arrangement of folds. Besides this, the stuff appears
lighter and thinner, and clings more closely to some parts of the body, which thus
become more prominent in the design. The difference in drapery between the frieze
and the pediments on the one hand, and the Parthenos and the metopes on the other,
only represents a further stage in the same perfectly normal progress. To make his
development close with the Parthenos and the metopes would be most unjustifiable.
On the contrary, as the frieze and pediments were made in the lifetime of Pheidias,
and as tradition tells us that he superintended the works on the Parthenon, we have
every reason for ascribing the growth of style — not to the school — but to the master
himself. Only in certain details of slight importance the assistants may quite
possibly have gone beyond the master's orders, but in all essentials the design is
emphatically his very own. That powerful and dominant personality which
undoubtedly governed the frieze and the pediments can, on the evidence of style
and of tradition,2 be none other than that of Pheidias himself. His name is and
remains closely associated with that flower of Attic art which continued unrivalled
for all time.
The decorative sculptures on the colossal images of Athena Parthenos and of
1 The pediment, owing to lack of space, gives a shortened form of the composition on the relief: it contains
no new elements, except that the rider Selene is replaced by Nyx in a chariot as filling more suitably the angle of
the pediment. 2 Kekule (Weibliche Gewandstatue) has recently attempted to assign the pediments to Alkamenes and
Agorakritos. As the basis on which his arguments rest is false, his whole theory falls to the ground. The
supposed differences between the pediments do not exist ; Kekule mistakes differences in the characterization of
stuff for differences of style (e.g. in the case of the 'Iris'); the supposed statues of Alkamenes which Winter
believed he could point to, and which Kekule uses as the basis of his contention, have no connexion whatsoever with
Alkamenes (see infra, p. 84, note 8), and, even if they had, the pediments would none the more be by Alkamenes.

48 PHEIDIAS
Zeus — the Amazonomachia, the boundary motive of Helios and Selene, the Niobids,
the Anadyomene, and the Gigantomachia — show how intimate was the artist's
connexion with painting, and thus confirm the testimony of Pliny, who says that
Pheidias began life as a painter, and only turned to sculpture later on. His brother
Panainos remained a painter. Pheidias himself carried out the painting on the inner
surface of the shield of the Parthenos, while the corresponding decoration of the
Zeus was handed over to his brother.
So far our conclusion is that the ascertained works of Pheidias are all comprised
within a period of about twenty years. Pheidias, it is true, may have worked before
this period, but the analysis of the Lemnia and its forerunners has shown, and will
show us further still, that it is improbable his artistic career began further back
than some ten years before the Lemnia, so that either he did not live to be old, or
he was no longer young when he turned from painting to sculpture. Of the two
hypotheses, the former seems to me the more likely ; in that case Pheidias would scarcely
have lived to the age of sixty. This theory seems, however, to be contradicted by the
tradition reported by Plutarch, according to which Pheidias introduced on the shield
of the Parthenos his own portrait as a bald-headed old man. On the copy of the
shield in the British Museum this figure is clearly to be seen. It represents an elderly
man with hollow cheeks, short full beard, and bald head, save for some scanty locks
at the back.1 The man is brandishing a weapon with full force against an Amazon.2
A man of about forty (the age of Pheidias at the time of the Parthenos, according to
our calculation) could very well, at least in our modern life, look like this figure, but
in the art of the fifth century such a figure would be meant to represent a more
advanced age. The short-cut beard and hollow cheeks especially are part of the
conventional type of the ' old man.'
To the question, then, whether Pheidias was an old man when he created the
Parthenos I should incline to answer distinctly in the negative, for this so-called
portrait is probably as apocryphal as are the numerous portraits of artists, their
friends and relations, which ciceroni identify in famous pictures of Christian art.
The legend given by Plutarch, that the portraits of the artist and of Perikles were
the occasion of the prosecution of Pheidias, is a foolish and meaningless invention,
as Scholl3 and A. S. Murray4 have already shown. The information given by
Plutarch about the portrait of Perikles, and confirmed by the copy, shows what sort
of likenesses we have to expect ; the arm of the figure was, it seems, raised and held
in front of the face, so as to cover it. This is said to have shown the cunning of the
artist, who wished to conceal the likeness. But it simply means that there was no
likeness at all. The whole story probably arose from the rather individual characteriza
tion of the old man : he was supposed to be Pheidias, hence the warrior beside him
must be Perikles, even if his face was almost covered. There is much evidence for be
lieving that the ' old man type ' was a general favourite in the cycle of Polygnotos. In
combination with youthful figures he served to vary the design, and in the representa
tion of the Amazonomachia this active greybeard was appropriate as showing how
all Athenians young and old united in patriotic resistance to the invading enemy.
1 For the whole shield see Collignon, Hist, de la Sculpture Grecque, p. 545.
2 A stone according to Plutarch ; the same in the Lenormant statuette (Michaelis, Taf. 15, I b.) ; a battle-
axe on the shield in the British Museum ; the figure of the Capitoline fragment on which Schreiber (Parthenos,
p. 600 (58) sqq.) recognizes the ' Pheidias ' is a copy of some other figure of the original.
3 Sits. Ber. d. Miinchener Akad. Phil. Hist. CI. 1881, 1.
4 Encycl. Brit., 'Phidias.' [Mr. Murray, however, supposes the Strangford shield to be wholly a later work,
¦produced subsequently to illustrate some current story on which that description (i.e. of the portraits) was
founded.' The same view, only very slightly modified, in Hist, of Gr. Sc. vol. ii. p. 121.— E. S.]

OTHER WORKS RELATED TO THE LEMNIA 49

VIII. Other Works related to the Lemnia. — The Master of Pheidias. — Early Pheidian
Works. — The Anakreon. — Pheidian Eros and Aphrodite.
Hitherto we have discussed only those works which are in close relation to the
Lemnia, either through the forms of the body or the treatment of the drapery. We
must now take the head as starting-point for a further inquiry.
There exist a whole series of heads really related, though in varying degree, to
that of the Lemnian Athena. A complete list of these is not given, as it would be
useless without illustrations. It must suffice to select a few of the more important.
These fall into two groups, according as they are earlier or later than the Lemnia.
Foremost in the first group must be placed the Apollo recently found in the
Tiber, and now in the Museo delle Terme in Rome (Fig. 8 ; head alone, Fig. 9).1 It is
only a mediocre copy, of poor workmanship, yet it preserves enough of the original to
produce a great effect ; magnificence combined with gentle beauty is its pervading
quality.2 This general characteristic, as well as the unusual pose of the head,
which is energetically turned to the side, recalls the Lemnian Athena. The head,
however, exhibits a somewhat older style, the hair coming from the crown of the
head and lying flatter than in the Lemnia. A still closer relation between the two
works makes itself felt in the spiritual expression of the refined face and in the peculiar
beauty of the closed mouth. By way of contrast we may compare the head of the
Capitoline Apollo, which has been grouped with the Apollo of the Terme? There is
a superficial similarity between the two, and they belong to the same period, but those
finer elements of form on which the spiritual expression depends differ totally.4 Very
different also is the Cassel Apollo {infra, Fig. 80), which reproduces another type of
the same period. The only points it has in common with the Terme Apollo are the
forehead broadening at the base, and the large open eyes ; but these are features
individual to the god represented. We shall have to return to this point later on.
Thus there is every probability that the Apollo is an earlier work by the
artist of the Lemnia.5 The attribution is of special weight, as the Apollo, in spite
of the individual character which distinguishes it from contemporary works, is still
clearly limited by certain traditions of the older school, while the Lemnia at once
takes its place as the creation of a finished and independent artist.
Now the school traditions alluded to are those of the Argive Hagelaidas.6 In the
first place the position is typical of that school, i.e. the body is supported on the left leg,
1 Rom. Mitth. 1891, p. 302; 377 sqq. (Petersen), Taf. x.— xii.
2 Petersen is right in speaking of the ' indescrivibile espressione di dolcezza ' in the head.
3 loc. cit. Taf. xi., xii., on the left ; Overbeck, Atlas Kunstmyth. Taf. 20, 22.
4 I cannot agree with Petersen in ascribing both statues to the same artist. The Capitoline head is much
more nearly related to the charioteer in the Pal. Conservators It would be nearer the truth to say that the
Capitoline Apollo must be by the same artist as the Cassel Apollo. In reality the original of the Capitoline
figure was probably by an artist of the second rank, who was influenced by the artist of the Cassel type in the
body of his statue and in the head by the artist of the Terme Apollo (infra, p. 197). It is impossible to
identify all the Apollo types of the period with the few which are handed down by tradition, for the simple
reason that the existing number is far larger than the number reported in literature.
6 The first suggestion is due to Petersen, loc. cit. After what has been said on p. 35 it is of course
impossible to identify this statue, as Petersen does, with the Apollo of the Delphic group ; nor is it likely that
the Terme statue reproduces the Parnopios. We must always remember that only a very small fraction of the
actual works, even by the greatest artists, is recorded by tradition.
6 See 50th Berl. Winckelmannsprogr. 1890, p. 134 sqq. The theories recently propounded by Kalkmann, 53rd
Berl. Winckelmannsprogr. (1893), according to which the figure by Stephanos goes back to an original by
Pythagoras, seem to me completely to miss the mark. H

Fig. 8.— Statue of Apollo in Museo delle Terme (Rome).

APOLLO OF THE TERME 5 1
the right leg is slightly bent at ease, but the foot is flat on the ground, the right arm hangs
somewhat quietly down, the left is in action, the head is turned to the left and inclined.
The figure has a broad firm pose, owing to the leg at ease being set well to the side.

Fig. 9. — Head of Apollo in Museo delle Terme.
This trait recalls the Apollo of Mantua,1 a work which I have already named as
one of the forerunners of the Lemnia (p. 25). The Terme Apollo forms the link
between them. Other marks of the school are the excessive breadth of the shoulders
in proportion to the hips, the hard contour of the right shoulder, which as usual is
1 loc. cit. p. 139 sqq.

52

PHEIDIAS

slightly raised, and the closeness of the right upper arm to the body. The motive
of&the left arm too (which was raised very high, and no doubt originally rested on a
laurel stem1) already occurred in works from the school of Hagelaidas. Such are an
Apollo, preserved in two small marble copies, unfortunately only torsos;2 a statue
turned into an Antinous by a copyist of the time of Hadrian ; 3 and finally an admir
able bronze statuette of Apollo in the Louvre.4 All these works, which are connected
more or less closely with the
school of Hagelaidas, show the
same motive as our Apollo.
The latter, however, differs
from them all in the more
decided and energetic bend
of the head to the left, by
which it gains in power and
majesty; 5 and in the markedly
individual modelling of the
spare but powerful body.
In spite of these differ
ences the connexion of this
work with the school of Hage
laidas is evident even in the
formation of the head. The
Mantua Apollo is clearly its
immediate forerunner, as may
be seen more especially from
the peculiar mouth.6 Thus
the surpassing beauty of the
mouth of the Lemnia is not an isolated phenomenon, but the artist derived it from
that older master to whom we owe the Mantua Apollo. The resemblance must
impress any one who compares the two statues.
In a former essay I tried to prove that the artist of the Mantua Apollo, while
belonging to the school of Hagelaidas, held his own definite position within that
school.7 His type of head is not the conventional type of the school, the mouth
especially being formed in a manner entirely his own. Now this artist, as the facts
we have just investigated reveal, must have been the teacher of the artist who made
the Terme Apollo — in a word, of the young Pheidias.

Fig. io.— Bronze Apollo of the Mantuan type from Pompeii, Naples.

1 A very favourite motive on vases of the fifth century ; similarly on coins and a gem, cf. Roscher's Lexikon,
i- 457- 2 (a) Louvre, Gal. Mollien (No. 2955) ; (b) torso in the Museo Chiaramonti, 199.
3 50th Winckelmannsprogr. p. 147; Bull. Commun. di Roma, 1886, 7.
4 Longperier, Notice des Bronzes Antiques, No. 439, ' Achille. ' This is not a Greek original bronze, probably
not even an exact copy, but a free modification of the Augustan period. The head recalls the Zeus Talleyrand,
the hair being dressed in the same affected archaistic manner ; the side curls are rendered as in the Munich
Artemis of Gabii (cf. infra, Appendix) ; the body however is an example of the type under discussion.
5 Another difference consists in the strong growth of curled hair on the pubes, u detail traditionally absent
from the other works of the Hagelaidas school. The front of the figure is much injured by the action of the water,
but the back is in good preservation.
c The different copies, the bronze of Pompeii, the marble statues of Mantua and Paris agree in their render
ing of this mouth, which means that they correspond to the original. On the relation of the copies to one another,
see my article Berl. Phil. Wochenschrift, 1894, p. 81 sqq. Fig. 10 gives the upper part of the bronze Pompeian
statue from a photograph of the original.
7 50th Berl. Wi mkelmannsprogr. p. 140 sqq.

THE MASTER OF PHEIDIAS 53
A bronze from the Akropolis x proves that this artist worked in Athens, or at
least for Athens. The head appears to be earlier than the Apollo ; it is rather harder,
and it should probably be dated in the period immediately preceding the year 480 B.C.2
By the help of tradition we can proceed a step further, and name this artist. It is
Hegias whom we know as the teacher of Pheidias. An inscription on the Akropolis
shows that he was working there shortly before the Tersian sack.3 And from the
monuments we have learned that, though a pupil of Hagelaidas, he followed out his
own artistic instincts, and gave to the head of his statues a new spiritual expression.
The beauty he strove to embody was softer, fuller, and more sensuous than the ideal
of the Peloponnesians, but it was left to his pupil to develop and idealize it.
Of the two ancient traditions concerning the masters of Pheidias, the one,
giving the name of Hegias, comes down to us on the excellent authority of the learned
Dion Chrysostom,4 who was specially interested in Pheidias. In the same passage
he shows himself to be well informed about the master of Polygnotos and his
brother. It is certain, then, in the opinion of the best ancient scholars, that Pheidias
was the pupil of Hegias. The other account, in which the better known and
more famous Hagelaidas was named,5 was however not entirely false, for the art of
Hegias was derived from that of Hagelaidas, and the youthful works of Pheidias
showed on the face of them their close connexion with the types of the school of
Hagelaidas. The monuments have taught us that the two traditions are really in
accordance. Hitherto we had not been able to form an exact notion of the style of Hegias,
because the ancients only characterize him in the most general way along with Kallon,
Kritios, and Nesiotes as a representative of the older and harder manner. We now
perceive that there are essential though subtle differences between the work of Hegias
and that of Kritios. How radically Kritios differs from the school of Hagelaidas I
have attempted to show elsewhere.6 I may here cite a work which on account of the
subject — it represents Apollo — is appropriate for comparison, and which is in my view
an excellent sample of the work of Kritios. The statue stands in the Pitti Palace,
and though scarcely known is of capital interest for the history of art.7 The
slender form of the god is supported on the right leg, the left being somewhat
advanced ; the small head (which is unbroken) is turned slightly to the left, but not
bent ; the eyes look straight out, as in the statue of a boy from the Akropolis, which I
consider to be likewise an early work by Kritios.8 The formation of head is the same
as in the boy, the hair too is similarly arranged in a roll subdivided into different
masses ; but the Apollo wears in addition long curls falling over the neck, after the
fashion of the Terme Apollo. The lifelike and slightly rounded modelling of the
body marks an advance on the Hagelaidas school, but seems severe and restrained in
1 Musies d'Athenes, PI. 16 ; for the discovery cf. Ath. Mitth. xii. 372.
a Cf. 50th Berl. Winckelmannsprogr. 140 seq., 148.
3 AeXriov dpx- 1889, p. 37 ; C. I. A, iv. 373, 259. The stone is blackened by fire, and therefore Lolling
dates it before the year 480.
4 De Horn, el Socr. i. O. Midler's emendation of the HnOT of the MSS. into HTIOT is scarcely an emenda
tion at all, whereas E. Gardner's suggested reading 'HyeXdSov (Classical Review, 1894, p. 70) strays far from the
traditional reading.
6 Only in the Scholia to Aristophanes (Frogs, 504), whence Tzetzes and Suidas derive their information. The
statement in the same passage about the date of Hagelaidas had of course no connexion originally with the
assertion that Pheidias was his pupil ; for whoever placed Hagelaidas in the period of the Peloponnesian War
could not possibly suppose him to be the teacher of Pheidias.
6 50th Berl. Winckelmannsprogr. p. 150 seq.
7 Diitschke, ii. 4; Overbeck, Apollon,^. 170, 5.
8 50th Berl. Winckelmannsprogr. p. 150; cf. pp. 7, 19.

a. — From the original.

Fig. ii.— Terminal bust in the Capitoline Museum.

b. — From the cast.

PHEIDIAS AND KRITIOS 55
comparison with the so-called Omphalos Apollo, and corresponds to what we know of
the manner of Kritios. The left arm is raised high,1 just as it is in that Argive work
which inspired the statue attributed to the young Pheidias, who must also have
known the Apollo assigned to Kritios, since their statues are connected by the arrange
ment of locks on the nape of the neck.
The Apollo of the Terme must be dated about 460 — 455 B.C., some five or ten
years earlier than the Lemnia. We possess a whole series of works belonging to the
same period, and showing in all essentials the same tendency, but only a very few of
these can be said to have a connexion with the Lemnia direct and personal enough
to warrant us in ascribing them to Pheidias himself.
The strongly plastic treatment of the hair in long lifelike curly waves, as we find
it in the Apollo of the Terme and, still more developed, in the Lemnia, is common to
several heads of the time between 460 and 450, whose type of face however makes it
impossible to attribute them to Pheidias. Such are a head in the Braccio Nuovo,2
probably representing a horned river god, and a terminal bust of similar style in the
Capitoline Museum (Fig. n).3 In these examples curls fall on the back of the neck
as they do in the Apollo of the Terme, but the whole expression, and especially the
harder rendering of the lower part of the face, shows more analogy to the Capitoline
Apollo 4 than to the Terme Apollo or to the Lemnian Athena, and lead us to suppose
that we have here two works by another artist. This was probably that contemporary
of the young Pheidias whom we shall shortly learn to recognize as the creator of the
Athena Albani : the two heads just mentioned certainly appear more closely related
to this Athena than to any other work. The forehead, the way in which the front
hair is parted over it, the eyes, the mouth, and the furrow between nose and mouth
correspond in a remarkable manner.
A bearded head in the Palazzo Barberini (Fig. 12) 5 seems to answer more nearly
to what we should expect of the young Pheidias. The hair, in short curly rings, is
treated in a thoroughly plastic and individual fashion. It is parted along the top of
the head as in the Lemnia. The line of profile and the shape of the skull also
correspond approximately to the Lemnia, but the head is somewhat earlier. In
many particulars it represents a preparatory stage to an Asklepios in Dresden,6
which belongs to the circle of Pheidias, and is some twenty years later, and, like it,
may have been intended for an Asklepios or some hero of kindred nature. The
expression of the face is calm and dignified, the forehead — still very low — is smooth
and unruffled. The head of a youth from the Jacobsen collection (No. 1095^) is of great
interest because of its relation to the Lemnia. It is a small head, turned ener
getically to the right, with short full locks, into which a fillet presses. Its connexion
with the old Argive type, as we know it from the athlete by Stephanos, is still
marked. 1 The cloak falling from the upper arm and the lyre below are additions of the copyist. In the original the
left hand probably grasped a laurel staff, and the lowered right hand held a bow.
2 Bonner Studien, Taf. 8, 9 ; Winter, loc. cit., interprets the head as Iacchos ; Rom. Mitth. 1891, p. 153 ;
(Sauer, according to whom the head represents Aktaion) ; Helbig, 9. The head is wrongly set ; it was intended
to be upright as in the cast in the Ecole des Beaux-Arts at Paris.
3 Mus. Capit. i. 83. In the room of the Philosophers, No. 85.
4 Rom. Mitth. 1891, Taf. 11, 12, to the left.
5 Matz-Duhn, Zerstr. Bildw. No. 1741. Cf. Graf, Aus der Anomia, p. 63, note; phot, in German Inst, at
Rome. s Treu has lately identified a replica of this work in Olympia (cf. Arch. Am. 1890, 107; 1892, 66). His
interpretation and restoration of it as a Zeus seem to me at present quite unproved.

TELEPHANES 57
Attempts have lately been made to refer the Hermes Ludovisi x to the earlier period
of Pheidias. It is certainly the work of a very great artist of the period in question,
but its relation to the Lemnia is not close enough to warrant us in assigning it
to Pheidias. The difference is plain, even in the fashion of the fillet, which lies
flat on the hair, without pressing into it as in the Lemnia. The formation of the
lower part of the face, the rather thin lips, and the picturesque boldness in the
rendering of the cloak which is slipping off,2 and even the bodily forms, are not in
harmony with the style of Pheidias. There is an admirable replica of this head
in the collection at Broadlands 3 (Fig. 13). The style is here rather freer than in
the Ludovisi example, the copyist having made plentiful use of the drill, and put
more modelling into the cheeks, but the winsome grace and the sweet roguish
expression of the head must be due to the original. Ionic art is the only possible
source of such a work.4 On a Phokaean electrum coin,5 an etcTV) QatcaLs (PL VI. 19),
occurs a head of Hermes very similar to that of the statue, wearing the same petasos
and short curls, with the same profile, and evidently belonging to the same period.
It seems admissible to bring this Hermes type into connexion with Telephanes of
Phokis, who apparently lived about this time.6 The further analogy between our
Hermes and the head of a youth on a Thessalian relief7 strengthens this theory,
since Telephanes is known to have worked for Thessaly.
There is a female type which may very well have belonged to Pheidias's early
period, and be regarded as a forerunner to the Lemnia. Its best replica is the
beautiful head in the Hermitage (profile in Fig. 15).8 The hair is dressed with the
simplicity so much affected by the wider circle of Hagelaidas, and which the Lemnia
retains, though in a richer form, i.e. it is waved back from the front and rolled up
behind. The parting has not yet appeared ; the lines of hair come straight from the
crown of the head. The way the hair frames the forehead, the line of the profile, the
mouth with its full lips — all seem to me so many analogies to the Lemnia, and indica
tions that Pheidias is the artist. There is a variant of this head with less severe
expression and more open mouth in the Coll. Barracco (Fig. 14).9 Like the one in St.
Petersburg, this head is distinguished by unusual dignity and beauty, and may also be
referred to Pheidias. A small and slightly worked head of the same type is to be
seen in the Braccio Nuovo, on an Artemis of more recent style, to which it does not
1 Cf. B. Graf in Aus der Anomia, p. 69. In the Museo Boncompagni (Helbig, Museums, 871).
2 In the replica by Kleomenes, the so-called ' Germanicus ' of the Louvre, the folds of the mantle are
modified in Praxitelean style.
3 Michaelis, Anc. Sculpt, in Gr. Brit. p. 219, Broadlands 9. The edge of the petasos is wrongly restored.
The nose is uninjured and antique. The head is placed incorrectly in an upright position on a modern term.
The length of the face m. o-i63 (the Ludovisi copy 0-165). To the list of replicas of the head should be added
Uffizi Diitschcke 13 = Arndt-Brunckmann, Einzelverkauf, 83, 84.
4 50th Berl. Winckelmannsprogr. p. 152, where I suggested Pythagoras of Samos as the artist.
5 The replica in the British Museum is illustrated in the Catal. Ionia, PI. IV. 23. The Berlin example
(PI. VI. 19) is more severe in style and more like the statue.
6 The author who makes him work for Xerxes and Darius (apud Plin. xxxiv. 68) is of course thinking of Darius
II. , but forgets that this monarch was separated from Xerxes by an interval of forty-two years ; he probably only
meant that Telephanes worked for the Persian kings of the fifth century B.C. The fact that Telephanes is grouped
with Polykleitos, Myron, and Pythagoras shows that he must have flourished about the middle of the century.
Cf. Julius Lange, Fremstilling af Menneskeskikkelsen, p. 133 seq.
7 Bull, de Corr. Hell. 1888, PI. VI.
8 Guedeonow, Sculp. Ant. No. 32, 'athlete.' Length of face CV156; length of head c. o'22. The propor
tions correspond to those of the Lemnia. Mouth to inner angle of the eye (64 mm.) = the length of the nose up
to the arch of the eyebrows = the lower part of the face = twice the length of one eye ; width of the mouth (42 mm.)
= height of the forehead.
9 Coll. Barracco, PI. 36, 'tete d'ephebe,' and in the text 'tete d'athlete'! The eyes are hollow. A
small replica of the Barracco head in terra-cotta from a high relief is now in Dresden (Coll. Dressel). I

Fig. 13. — Terminal bust in the collection at Broadlands. (From a photograph.)

Fig. 14.— Head in the Collection Barracco (Rome). (By permission of Messrs.
Bruckmann in Munich.)

Fig. 15. — Head in the Hermitage,

60 PHEIDIAS
belong.1 True, there is something of the nature of Artemis in the pure, straightforward,
maidenly, Amazon-like expression of the heads, so the mistake is not surprising.2
A beautiful female head in the Villa Albani (No. 63) is also possibly a replica of
an older Pheidian work.3 It may be placed about midway between the well-known
Madrid head,4 which is so closely allied to the Olympia metopes, and the Lemnia.
The hair is arranged as in the Madrid head. It is parted down the middle as in the
Lemnia, but is treated in a much simpler manner. The low forehead with its delicate
modelling, the mouth, and even the shape of the ears recall the Lemnia.
We can judge with greater certainty when we come to works contemporary
with or only slightly later than the Lemnia.
In connexion with the Albani head just mentioned, it is fitting to discuss an
Athena type which is certainly of Pheidian origin. It survives in three terms, two of
which are from Herculaneum — the first of these is given in Fig. 16 — the third is in the
Capitol.5 This beautiful head shows analogies to the Albani female head mentioned
above and to the Lemnia. The style is that of the Lemnia, but the expression, less
serious and solemn than kindly, as well as the carriage of the head, brings it nearer to
the Albani type. The goddess wears her helmet pressed low down on the forehead, as
does the Parthenos at a later date. A gorgoneion on an aegis-like background of scales
forms the symbolic ornament of the helmet. Curls fall down at the side as in the
Parthenos. Although this conception of the goddess differs not only in external
particulars but in expression from the Lemnia, yet the inner analogy between them
is in my view undeniable ; the mouth and the lower part of the face are peculiarly
characteristic. On the other hand, an Athena head in the Munich Glyptothek6
shows how style can differ in spite of striking external similarity. It is a replica of
a work belonging to the same period, but not by the same artist.
To the period of the Parthenos itself belongs a superb work of Pheidias which
survives, as I believe, in one complete copy and in various reproductions of the head.
I mean the Anakreon standing in an attitude of rapt enthusiasm and declaiming his
poems to his own accompaniment on the lyre (Fig. 17).7 This work is now universally
1 Helbig, Museums, No. 20 ; Clarac, 571, 1220. Arndt informs me that the head is put on with a straight
junction surface, that it does not belong to the statue, and that the whole back from the girdle upwards is new. The
type of this torso, a fourth-century modification of the Parthenos, was also used to represent Isis (cf. Beschr. d.
Antikcn Skulpt. in Berlin, p. 529, No. 60 a). The quiver band shows that the statue was originally intended to
represent Artemis. The quiver itself is omitted.
2 It is interesting to note that the St. Petersburg and Barracco heads, like that of the Lemnia, were formerly
supposed to be male.
3 Length of face 0-175. I fiud no mention of this admirable work in literature. (Cf. Berl. Winckelmanns
progr. p. 130.) ij. H. S. 1884, Taf. 45.
5 (1) Naples, Iim. 6322, Comparetti de Petra, Villa Ercol. Tav. xx. 1 ; Jahrb. d. Inst. 1893, Taf. 3, pp. 174,
177 ; (2) Naples, Inv. 62S2, Jahrb. d. Inst. 1893, p. 176; (3) in the Philosophers' Room of the Capitol, No. 54:
Mus. Capitol, i. 56, Jahrb., loc. cit. p. 176. In all three instances the term is antique. In (2) and (3) the breast
is covered with drapery which can hardly be copied from the original. The turn of the head varies, as is often
the case in term copies (see term in Index). Wolters (Jahrb., loc. cit.) attempts to trace back this type to
Kephisodotos, the artist of the Eirene ; but the forms of brow, hair, eyes, and mouth show by their style that the
head belongs to the fifth centuiy and to the Pheidian epoch.
6 Placed on a statue (No. 86) which does not belong to it.
7 The statue, formerly in Villa Borghese, and now in Copenhagen (Ny Carlsberg), is published Arch. Ztg.
1884, Taf. 11; cf. Friederichs- Wolters, Gipsabg. 1305. I know three replicas of the head: (a) in Rome,
Helbig, 599 ; (b) in the Palazzo Riccardi in Florence, to the left of the entrance upstairs (Diitschke, ii. 200)—
the head is wrongly set ; (c) in Berlin, Jahrb. d. Inst. 1892, Taf. 3. Kekule's notion, ibid. p. 119 seq., that
the Berlin head is a fifth-century original, is absolutely false : it is an ordinary Roman copy, in a bad state of
preservation, and much restored. The head in the Louvre, 1946, seems to be a fourth, very poor replica. A
fifth, badly preserved, in Glienecke, is mentioned by Kekule, p. 120.

Fig. 16.— Terminal bust of Athena fiom Herculaneum (Naples).

62 PHEIDIAS
admitted to be of fifth-century date.1 Any one who comes fresh from the study of
the Lemnia cannot fail to be struck by a wonderful likeness between the two heads
which can only be explained by identity of authorship. Although rather short, the
hair is parted as in the Lemnia, and the fillet presses deep into it, while the refined
forehead with its flat and restrained modelling, the eyebrows, and the thin eyelids are
so many points of definite resemblance ; even the ear has the shape characteristic of
the ear of the Lemnia. The head, although conceived in an attitude of vigorous move
ment, yet produces as a whole that wonderful effect of combined dignity, beauty, and
repose which was peculiar to Pheidias. Moreover, the only complete parallel to the
drapery occurs on the Parthenon metopes of free style, which are to be referred to
Pheidias. Nowhere else do we find the peculiar clinging of the mantle to the nude
forms, the broken folds, or the same true characterization of the soft yet heavy
material.2 This fixes the date of the statue at about 445-440. The frieze and the
pediments show a definite advance in style. The way the short cloak hangs round
the shoulders echoes archaic tradition and recalls the Oinomaos of Olympia. But
the influence of the circle of Hagelaidas has long ceased to dominate the artist.
The bodily forms are' nervous and fresh as in the metopes of free style. The feet
are placed on the ground in the old scheme, but the upper part of the body is
turned, giving to the whole composition novelty, vigour, and exquisite freedom of
rhythm.3 The Pheidian statue of Anakreon was doubtless identical with the one which
stood on the Akropolis, and whose attitude, according to Pausanias, was 'suggestive
of a man singing in his cups.' 4 In ascribing the animated and inspired turn of the
head to a liberal use of wine, Pausanias was making no unnatural supposition in the
case of a man who was the poet par excellence of wine and intoxication, and who
himself appears intoxicated in a work of art which, whether real or fictitious, is alluded
to in epigrams.5 Beside the statue of Anakreon, Pausanias saw that of Xanthippos,
the father of Perikles and friend of Anakreon. It has long been rightly supposed that
both statues were set up by Perikles.6 Of course they were not public monuments,
but purely private dedicatory gifts, meant to express the filial affection of Perikles
for his father and his reverence for his father's distinguished friend. If the name of
Perikles came first in the dedicatory inscription, it is easy to understand why
Pausanias begins to speak of Perikles just in the very passage where he names these
works, although the statue of Perikles himself, as he says, stood at some distance.
The artist was probably not named. If we remember Lucian's statement about
the inscription of the Lemnia, we can hardly suppose that the name of Pheidias was
to be seen on another statue of the Akropolis outside the Parthenon. Considering
the artist's close relation to Perikles, his authorship would at that time appear
so much a matter of course that his signature could be left out. The statue
1 Cf. Winter, Jahrb. d. Inst. 1890, 159 ; Helbig, loc. cit. ; and especially Kekule, loc. cit.
3 Cf. the metopes Michaelis, South ii., viii., xxviii. Cf. Kekule, p. 121.
3 The pubes with its thickly curled hair resembles that of the Terme Apollo.
4 [Tr. Harrison and Verrall, Myth, and Mon. of Anc. Athens, p. 471.]
5 Wolters (Arch. Ztg. 1884, 150) rightly doubts the existence of the work ; in any case there are no
grounds for identifying it with the statue of the Akropolis. Cf. also Kekule, p. 119, note 4.
6 Brunn in Ann. d. Inst. 1859, 183. Brunn is wrong, however, in connecting the portrait of Perikles,
which according to Pausanias stood in quite a different spot, with these two statues, and in ascribing them to
Kresilas. The Perikles was a private anathema of Kresilas. Kekule, p. 125, follows Brunn, and, arguing from a
supposed stylistic connexion between the Anakreon and the Perikles, suggests the Anakreon was also by Kresilas.
There is, however, no stylistic connexion, except in so far as both works belong to the same epoch. Sal. Reinach,
Chron. d' Orient, No. xxvi. p. 11, comes nearer the truth in suggesting that the Anakreon might be by Kolotes.

THE ANAKREON

63

was set up, as is proved by the style, during the period when Perikles was most
powerful, and could without opposition place upon the Akropolis, iv iirupaveo-TaTtp
to7t&), images of his father and of Anakreon, the friend of tyrants.
The influence exercised on more recent art by this work of Phedias is shown

Fig. 17. — The 'Anakreon Borghese' (Ny Carlsberg Glyptothek, Copenhagen).

in a beautiful head in the Vatican,1 a copy of a work of the fourth century B.C. : it
represents a poet with inspired upturned look, and is visibly a development, even to
the form of the beard, of the Anakreon of Pheidias.
A magnificent bearded head in the Museo Torlonia (Fig. 18) is closely analogous
to the Anakreon, and must likewise be by Pheidias. The strikingly noble and
elevated beauty of this head must dwell in the memory of all who have seen it.
1 Helbig, Museums, No. 287. The old interpretation of the head as Zeno, and Helbig's suggestion of Aratos,
are equally unproved and inappropriate.

64

PHEIDIAS

Fig. 18

For my own part, since I first beheld it in 1877, it has kept its place in my remem
brance as the gem of the collection.1 The delicate forehead swells almost imper
ceptibly in the middle, and is low, just like that of the Lemnia, to which the
eyebrows and thin eyelids also correspond closely. The hair is again bound by
a diadem, broader in front, narrower behind, with short
curls escaping from it. The general -similarity to the
Anakreon in hair and beard is striking, although many
of the details are quite different. In turning to identify
the head, we note that two small wings are attached —
not to the head itself — but to the diadem. Analogies
are found in contemporary vases of the period ranging
from 460 — 450, where daemonic attributes appear fast
ened to the diadem. Athena, for example, wears a
diadem on which Pegasoi are seated,2 and Thetis a
diadem adorned with small wings, to indicate her trans
formations.3 More important still is the painting on
a white lekythos of the British Museum ; the black
outline drawing shows a bearded man with wings on
his head rising from a wide fillet ; he rushes forward to
meet an advancing warrior, and strikes him on the arm.
Without attempting to interpret this vase-painting,
I wish to point out that the art of the time ap
parently had not the courage to effect an organic
connexion between the wings and the head. Further,
in the Torlonia head the wings are purposely kept small,
in order not to break the fine outline of the skull.
The winged diadem apparently has much the same signification as the wings
attached actually to the head in later times : one might imagine the head to be
that of a bearded Hermes, were it not that the type of features is absolutely
foreign to that god. The similarity to the Anakreon brings us seemingly nearer
the truth : may we not have here a head of Mousaios, the old god-inspired poet, who,
according to Attic legend, had received the gift of flight — and therefore, we may
suppose, the gift of wings — from Boreas ? 4
To the same group belongs a beautiful bearded head in the Museo Chiaramonti
(Fig. 19),5 which expresses marvellous elevation and distinction. It is nearly related
to the last-mentioned heads, and also to the Lemnia, and must be referred, if not to
Pheidias himself, as I believe, at least to a master of his school and of the period
about 450 — 440. The treatment of the wavy hair on the upper part of the head, the
roll of hair on the neck, and the small curls escaping from it recall the Lemnia.
Entirely Pheidian, too, are the finely modelled forehead, the thin eyelids, the spring
of the nose, the arrangement of the beard, and the shape of the ear.6 It is difficult
to form any conjecture as to whom the head represents. The tuft of short
hair rising in front above the twisted fillet is quite unique, and the tangled hair
1 No. 48 in the Catalogue of 1876 ; now No. 50. Benndorf, Rom. Mitth. i. 113, notices the analogy to the
Anakreon. 2 Elite dramogr. i. 29, and Millingen, Div. Coll. 49.
3 Mon. d. Inst. i. 37.
4 Onomakritos, apud Paus. i. 22, 7.
5 Mus. Chiaram. Tav. 33 ; Helbig, 71. The tip of the nose is restored.
6 The end of the nose has been distorted by the restorer, and too much curved.

Head with winged fillet, Museo
Torlonia, Rome. (From a drawing.)

i*

Fig. 19. — Head in the Museo Chiaramonti (Vatican).

66 PHEIDIAS
about the chin is also probably a characteristic mark. The current interpretation as
Dionysos is absolutely groundless. It very probably represents a hero, but, until we
know which hero, the finer intentions of the artist must remain a mystery.
Two female types still have to be mentioned in the present connexion : one is
the head of a young goddess,1 with a wide fillet wound three times round it and a
bunch of hair behind tied up in a <rdKKo<;. The type is of the same period as the
Lemnia, and probably owes its existence to Pheidias.
Closely allied to this head, although stylistically somewhat later, is the
• Sappho,' of which so many copies have survived (Figs. 20 b, 21 b).2 Here too we
have the wide fillet wound several times round the head, although arranged in a slightly
different manner. The forehead is freer and higher, the curls only escape freely
above the ears. This is quite in the manner of the Parthenos,3 of which the head
appears to be contemporary. The rich plastic waving locks on the upper part- of
the head are genuinely Pheidian. In front of and behind the ears (which again are
shaped like those of the Lemnia) a few wisps stray from the mass of hair. The
mouth is a little open, giving a glimpse of the teeth. 4 The head is slightly turned to
the side, the expression fresh and gracious.5 The long narrow form of the eyes is
characteristic. There can be no question of a portrait,6 but the rich fillet, the expres
sion of the face, and the form of the eyes are more appropriate to Aphrodite than to
any other goddess. This interpretation, already suggested by Helbig,7 is strongly
confirmed by a wonderful statuette from Syria, now in Paris in the Collection de
Clerq. This work represents a nude Aphrodite binding a wide fillet round her
head. Except for some few omissions and unimportant alterations, the head of the
Aphrodite is of the type under discussion. Although the body imitates forms of the
fifth century B.C., the whole conception is merely an invention of Hellenistic times,
uniting the head-type of Pheidias with the motive of the Diadumenos of Polykleitos.8
1 Four replicas of the head are known to me : (a) with inserted glass eyes in the Museum of the Capitol,
' Philosophers' Room,' No. 55, Mus. Capit. i. 57 ; (b and c) in the Museo Torlonia, Nos. 54 and 58 (formerly
52 and 56) ; (d) in the Ny Carlsberg Glypt. in Copenhagen.
2 I have examined and compared the following instances in the original : (a) in Berlin, Skulpt. 330 ; Arch. Ztg,
1871, Taf. 50 — poor ; (b) in Brunswick — mediocre copy, the face much polished over; (c) in the Brit. Mus. —
much damaged ; (d) in the Hermitage, No. 188 — face worked over, hair good ; (e) in the Palazzo Riccardi
(Diitschke, ii. 162) — a very good copy, clear and careful work, the nose antique ; (f) in Rome, Mus. Chiaramonti,
50 ; (g and li) in the Villa Borghese, Helbig, 926, 930 — the first specially good ; (i) in the Villa Albani, 109 ; (k) in
the Coll. Barracco ; (/ and m) two very good copies in the Museum at Corneto ; (»)a very good copy in Naples —
turreted crown restored, but draped term genuine ; (0) in the Louvre — a very good copy, but placed on a torso to
which it does not belong, No. 1 739 ; Frohner, No. 413; Bouillon, i. 50 ; (p) the double term in Madrid, Arch. Ztg.
1871, Taf. 50 ; Friederichs- Wolters, 1609. I only know this from the cast : it is one of the best (Figs. 11, 12).
All these instances are real copies, and correspond in size. There are besides : (q) a free adaptation in colossal
size from Smyrna in Constantinople. This is a Hellenistic, not a Roman work ; the severity has disappeared,
the manner is grandiose but somewhat coarse. The existence of this copy shows that the original must have been
famous in the Pergamene period ; (r) a free replica on a small scale, a head, which has been placed on the
Ludovisi group (Schreiber, 50). There is a beautiful antique variant in Oxford of the same style and with the same
narrow eyes (casts in the Ecole des Beaux- Arts, in the South Kensington Museum, at Bonn, etc. ) This head
must be referred to another work from the inner circle of the master.
3 K. Lange, Ath. Mitth. 1881, 93, has already classified this type with that of the Parthenos. Compare
especially the Minerve au collier in the Louvre, where the hair on the left side is evidently a faithful copy.
4 Especially in (e), (g), (0).
5 The attitude varies slightly in the different examples ; the heads (e), (/), (g) are turned to the left, while
(li), (i), (n), (0) are turned to the right. This is explained by the fact that all these heads belonged to terms ; the
copyists turned the head now one way, now another, for decorative reasons. The same was the case with the
Skopasian Herakles.
6 The colossal replica (p) is evidence to the contrary. 7 Helbig, Museums, 926.
8 This charming figure, about 30 cm. high, is intact, with the exception of the left arm. Beside the left leg
is a vase with the discarded drapery over it. The head is much inclined. The same motive, with reversed sides,

Fig. 20. — Double terminal bust in Madrid.

68

PHEIDIAS

If the head had not been originally meant to represent an Aphrodite, it could hardly
have been adopted for the goddess in this instance.
Unless the Pheidian original was itself a term,1 it must, from the indication
of a chiton on one of the busts, have represented the goddess draped.2
Thus, we may claim to have recovered an Aphrodite of Pheidias : it was perhaps
the one in Rome in Octavice operibus, distinguished by Pliny as eximiae pulchritu-
dinis. These words of praise, as well as the statement that the statue was in Rome,
fit in admirably with the existing copies. The eximia pulchritudo of the work must

Fig. 21.— Profiles of the double terminal bust.

at once be acknowledged. It has the majestic and elevated beauty which distinguishes
all that Pheidias did, combined with a winning sweetness of expression. This is the
goddess who, as the Attic vases show, dwells in fresh gardens where golden fruits
is to be seen on a carnelian, apparently a purely Hellenistic work (Caylus, Rec. d'Ant. vi. 38, 4). For the
signature Skopas cf. my article, Jahrb. d. Inst. (1893) viii. 185.
1 Terminal Aphrodite in Athens, Paus. i. 19, 2.
2 See the Naples copy (n). Some notion of the statue that belonged to this head-type may be formed
from the so-called ' Sappho Albani,' now generally interpreted as a Kore or Demeter (Helbig, 841) ; it is a
work of the Pheidian period, though manifestly not by Pheidias. Cf. R. v. Schneider, Jahrb. d. Oesterr. Hofmu-
seen, xii. 72. A good illustration in Brunn-Bruckmann, Denkm. No. 255. Besides the replicas mentioned
by Schneider there are : (1) a torso in the Louvre, No. 2001 ; Frohner, Not. No. 574 ; (2) a poor torso half life-
size, also in the Louvre, No. 2912 ; and (3) a half life-size figure with head restored in Naples. The type, as
Sauer (Festschrift fur Overbeck, p. 73) shows, already occurs on one of the metopes of the Parthenon.

APHRODITE AND EROS 69
grow, where nymphs and Erotes serve her, where reign harmony and bliss ; 1 but she
is a goddess still, not a mortal woman with human feelings and desires, such as
Praxiteles was to conceive her.
One of the copies of this Aphrodite head appears on a double term associated
with the head of a youth (Figs. 20 a, 21 a),2 which must also have been famous, since
it occurs separately in various replicas.3 The delicate
young face is framed in full curls which cover the ears ; a
circlet presses deeply into the mass of hair. In some
at least of the examples a fillet is wound round the circlet.
The artistic kinship of this head to the Aphrodite is
evident. They are both works of Pheidias, and nearly
contemporary. The stylistic treatment of the hair and
of the face (especially of the full, slightly open mouth)
represents the same stage of development as the Aphrodite.
The youth who appears united with Aphrodite on a
double term should, according to all analogy, be nearly
related to her in significance. He can be none other than
Eros, who in the time of Pheidias 4 was represented— not FlG- *2--Came° in Berii"-
as an effeminate youth, but — as an ephebe. A good
instance of the Pheidian Eros type, nearly akin to the head we are studying, is to
be seen on a cameo of the Berlin collection (Fig. 22), presumably of the Augustan
period. It represents just such a terminal bust of Eros as the one in question,
whose original may actually have been executed in this form. If our interpretation
is exact, it is plain that Pheidias did not give to his Love-god, any more than to his
Aphrodite, the expression of human longing which distinguishes him in the period of
Praxiteles.5 There is another ' Sappho ' type which should be noted here, as it belongs
originally to the circle of Pheidias. On a statue formerly in the Palazzo Cepparelli
in Florence (Fig. 23) 6 has been placed a head evidently of later Pheidian style, and
an immediate forerunner of the well-known Albani head which has lately been again
interpreted as Sappho.7 The close connexion of the two heads, evidenced by the
coif, by the twisted curls in front of the ear, and by the general type of face, is
unmistakable. Equally plain is the fact that the Florentine head, as shown by the
style, especially by the treatment of the regularly undulating hair, belongs to the
later Pheidian period. The body of the statue is of the same date as the head, but
1 Cf. the vase belonging to the period about 430, Stackelberg, Grdber, 29 ; Miiller- Wieseler, Denkm.
3rd ed. ii. 296 d.
2 The Madrid term p.
3 e.g. Museo Chiaramonti, 695 (nose and lips restored) ; in the Museo Torlonia, 44 (nose and lips restored) ; in
Dresden, 32 (much defaced). Published in Becker's Augusteum, ii. 85 ; the Madrid copy is the best ; mouth and
nose are antique ; the fillet hanging down is merely a decorative addition ; it is missing in the Chiaramonti head.
4 The witticism of Phryne, recorded in Athenaeus, xiii. p. 585, would seem to prove the existence of an
Eros by Pheidias; it at any rate implies a play on the name $ei$las (cf. Brunn, K.G. i. p. 187) ; and though
this in itself is insufficient to prove the existence of an Eros statue by this master, there is no doubt that the
ieu de mot would have been more pointed and appropriate had there been a well-known Eros by him.
& The Eros of the period, about 420 B.C., with which I identify a small head from Brauron (Arch. Studien
H. Brunn. dargebr. Taf. 3, p. 89) is less serious and more youthful.
6 Diitschke, Zerstr. Bildw. in Florenz, 413 ; now in the Museo Archeologico.
7 Jahrb. d. Inst. 1890, Taf. 3; p. 151 sqq. (Winter). Two poor replicas in the Bigliardo of the Villa,
Nos. 332, 333. The Pitti head published Ann. d. Inst. 1879, Tav. o, in which Gamurrini and Wolters (Gipsabg.
1609) recognize Sappho, is not a replica of the Albani head, as Winter (loc. cit. p. 152) affirms, but a different
though related type.

7° PHEIDIAS

it is not certain that they belong together.1 The narrow eyes are characteristic of
Aphrodite. In order to interpret the Albani head, which, to judge by the formation

Fie. 23.— Statue formerly in Pal. Cepparelli (Florence).

of the eyes, was derived from this older type, we must go to the group of figures
representing Aphrodite, or divinities akin to her. For there is no evidence whatever
1 According to a communication from Milani the head is of the same Parian marble as the torso. But as
the lower part of the neck is restored, and as the original head was probably inserted, the present head can
scarcely belong to the statue. > v

APHRODITE OURANIA 71
that the head is a portrait, and nothing in the work itself to warrant us in calling it
Sappho. The apparent coincidences found on coins are deceptive.1
Another Pheidian Aphrodite — the Ourania in Athens (Paus. i. 14, 7) I incline to

Fig, 24. — Statuette ot Aphrodite (Berlin).
think — has survived in a statuette from Corneto (Tarquinii), now in Berlin (Fig. 24),2
which is evidently an original free adaptation made in the actual school of Pheidias.
1 Pollux (Onom. ix. 84) says that the Mytileneans represented Sappho on their coins. This is the oldest
testimony on the subject, for to see an allusion to it in Aristot. Rhet. 2, 23 is purely arbitrary. None of the
extant coins of Mytilene with ascertained representations of Sappho (i.e. which give her name in the coin legend)
are earlier than Imperial date, any more than the coins of Chios with the portrait of Homer, which are
mentioned immediately after by Pollux. Now of these undoubted representations of Sappho, not one shows
the least resemblance to the Albani type ; they differ from it and from each other. They are as follows :
(a) the Paris coin with \f,air(ip)a, Jahn, Darst. Gr. Dichter, Taf. 8, I ; (*) Sappho seated, with and without the
name ; (c) the head inscribed iairipa on coins of Eresos. The coin inscribed Zaipovs, Jahn, Taf. 8, 2, is very
possibly a forgery. There is no reason to suppose that the hooded head of the autonomous bronze coins and
of an electrum coin of Mytilene (Sallet's Zeitschr. ix. Taf. 4, 4 — 6 ; Jahrb. d. Inst. 1890, Taf. 3 below)
represents Sappho. In the period before Alexander, to which some at least of these coins belong, there is no
instance of a portrait of a distinguished person on a Greek coin. Hence the head must be a divinity, perhaps
Aphrodite. The coif was a favourite headdress in Asia Minor. Thus Winter's hypothesis, that Seilanion was
the artist of the Albani head, is untenable. The Sappho of Seilanion is still to seek.
2 Berlin. Skulpt. 586, h. o. 83. Hands, feet, and some trifling details are restored together with the face
of the idol.

A. — From a lost original.

Fig. 25' — Two heads of Athena from casts in Dresden.

B — Head of Athena Farnese (Naples).

PUPILS OF PHEIDIAS 73
Aphrodite, leaning upon her own ancient image and drawing her veil aside with her
right hand, is represented in solemn repose — in a sort of subdued majesty. The left foot
rests on the ground with full sole, and is placed somewhat to the front. The drapery
— more especially the folded-over portion of the himation in front — is quite in the
style of the Parthenon frieze.1 The head is of the usual genuine Attic type of the
Pheidian epoch : its finer details, however, have been left out in this small
reproduction.2

IX. Pheidias and his Pupils — Alkamenes and Agorakritos.
We must now try to trace the career of Pheidias beyond the period of the
Parthenos. Among the many works which are more or less related to our subject
we select a significant statue, again representing Athena, which throws light on the
relations between the master and his pupils.
Fig. 25 reproduces side by side two heads of Athena, both from casts in
Dresden. The head on the right hand (B) is evidently taken from the well-known
Athena Farnese in Naples (Fig. 16)? The original of the other cast (Fig. 25, A ; profile,
Fig. 28) 4 has disappeared, but even from the cast it is absolutely clear that the bust,
with the ends of the curls and of the bunch of hair behind, is modern, while the
neck, with the portion of the curls adjoining it and the upper part of the bunch of
hair behind, is antique. The restorer copied the bust of the Cassel Athena, or of
one of the two Lemnia statues in Dresden, but he misunderstood the drapery on the
shoulder. The two heads go back to two different Greek originals, and yet they are
so similar in certain external particulars that we cannot suppose them to be entirely
independent of each other. The hair in front corresponds even in the form and
number of separate strands. The principal proportions (length of face and of eyes)
are the same in both. The helmets differ only in so far that while on A there are griffins
beside the Sphinx, on B (the upper part of the helmet has been left out in the
cast ; cf. Fig. 26) there are Pegasoi ; further, A has a volute at the side instead of the
turned-up cheek-piece of B. The curls on the neck are different. In B there are
two short artificially twisted spiral curls, while A shows a curl slightly waved and
falling naturally, apparently divided in the middle, and in the continuation, now
lost, probably parting into two separate strands. Great differences are noticeable in
the finer modelling of the face ; A is on the whole narrower and more delicate, the
forehead is lower than in B, the nose somewhat longer and narrower, and the mouth
narrower. Finally, the eyelids of A are thin, those of B very thick, while the
1 i.e. in short grooves rounded at the ends. Cf. Pallat, Jahrb. d. Inst. 1894, p. 17. For the chiton cf. the
' Peitho ' of the E. frieze.
2 A later echo of this Pheidian creation occurs in the statue recently published by Kekule, Weibliche
Gewandstatue aus d. Werkstatt d. Parthenon Giebel. 1 894.
3 Gerhard u. Panofka, Neapels Ant. Bildw. (1828), p. 41, No. 118 ; Clarac, PI. 458, 851 A.
4 Hettner says it is in the Hope collection (Verz. d. Abg. p. 109, No. 172) ; but in that collection there is
only the well-known statue (Michaelis, Anc. Sculpt, p. 290, No. 39), and the cast cannot possibly be of this statue,
because of the difference in the bust. Even if we leave the restored bust out of account the curls on the neck do
not exactly correspond ; in the Hope statue each is separated more distinctly into two parts, so that it could not
easily be mistaken for a single curl as the restorer has done. The Dresden cast is not from the Mengs collection ;
its original is unknown. There is a second example of it in Bonn ; Welcker (Bonner Ktmstmus. 155 = Kekule,
306) says vaguely that the original is in England. A third copy of the cast is in the Humboldt collection in Schloss
Tegel ; the Waagen catalogue says nothing of the original. In the New Museum of Berlin, built by Stiller, this
cast has been adapted for a decorative head. L

Fig. 26.— Athena Farnese (Naples).

Fig. 27.— Hope Athena in the collection at Deepdene, Surrey. (From Spec, of Anc. Sculpture.')

76 PHEIDIAS
modelling of the forehead, the curve of the eyebrows, and the formation between
eyebrow and upper lid are completely different in the two examples.
The head B is still unbroken on its statue, the Athena Farnese in Naples.
Head A, too, has a body belonging to it, for it is a replica of the famous Hope
Athena (Fig. 27) at Deepdene (Surrey), from whose priceless collection, unfortun
ately, students and public alike are now jealously excluded.1 This and the Athena
Farnese were formerly looked upon as copies of the same original ; 2 but more
exact comparison will show that this is not the case. Head A, in all those particulars
in which it differs from head B, corresponds exactly to the Hope Athena. In
this statue the curls disfigured by restoration on head A are seen in their proper
form. They split naturally into two wavy ringlets falling down over the aegis.
In spite of the remarkable external similarity between the two statues, they differ
not only in the type of head, helmet, and side-curls, but also in other respects.
To begin at the top, the opening in the middle of the aegis is wider in the
Hope statue, and shows more of the chiton ; the gorgoneion is of a distinct type
with hair not raised on end but smoothly waved close to the head, as on the aegis of
the Parthenos. The snakes on the aegis correspond on the whole, but they are
arranged in less conventional circles, and although not so effective are more various
and natural (note especially the second snake from the top on the left shoulder, which
is so much more simply rendered in the Hope statue, and thus affords a better
motive for the bend in the edge of the aegis). The folds of the chiton under the left
arm and of the edge of the mantle next it are more broken, less schematic and
more natural. The upper part of the sleeve on the left arm is shorter and simpler,
the folds of the mantle itself are heavier and better defined at the edges, the abdomen
is not so visible under the folds as in the Farnese Athena. In all these peculiarities of
the drapery the Hope statue more nearly resembles the Albani Athena, which we
shall study immediately. Further, the position of the feet is slightly different in the
two statues. In the Hope statue the left foot is not drawn back quite so far as in the
Farnese, which is really in the walking attitude.
The assumption that these variations are all due to different copyists, and that
there is only one original in the background, is not difficult to refute. For these
alterations are not arbitrary caprices intended to suit a later taste ; they are syste
matic, thoroughgoing distinctions, which manifest themselves most clearly just
where the finest artistic feeling comes to light — that is, in the forms of the face.3
We next have to consider which of the two types is the older and more original,
1 I have not seen the statue, but the engravings (Spec, of Anc. Sc. i. 25 and ii. 9) seem accurate and reliable.
Cf. Michaelis, Anc. Sculpt, p. 290. The head is inserted, not broken : only the nose is restored. On the head
A the nose seems to be ancient.
2 I did the same in Roscher's Lexikon, i. 700.
3 Of the Hope type I only know the one replica, now lost, of which we possess a cast. The Farnese type
exists in five replicas : (a) fragment of head in Dresden, formerly placed on one of the Lemnia statues, Becker,
Augusteum, i. 15 ; cf. supra, p. 5 ; (b) statue in the Louvre, wrong head, Bouillon, iii. stat. PI. 2, 4 ; Clarac,
320, 852 ; Frohner, 115 ; now No. 1650 ; (c) Demidoff statue, wrong head, Clarac, 470, 895 ; (d) Albani statue,
now lost, Cavaceppi, Raccolta, i. 1 ; Clarac, 458, 901 ; a sketch of it in Quatremere de Quincy, jup. Olymp. (1815),
PI. 9, 4. The statue is exactly like the Farnese one, except that the eyes are hollow. Winckelmann expresses
great admiration for this statue, and distinguishes it clearly from the Albani statue with the skin helmet.
When he speaks of the Pallas in hohem Stil of the Villa Albani, e.g. (Hist, of Anc. Art. Bk. viii. 2, § 4 ; tr.
Lodge, ii. 133), he invariably means the Cavaceppi statue, not the one with the skin-helmet, as Friederichs,
Bausteine, 86 ( = Wolters, 524), maintained. Cf. the Meyer-Schulze ed. vol. iv. p. 339, n. 331 =tr. Lodge, i.
476, and for the profile PI. XVIII. A of the latter work. According to the Specimens of Ancient Sculpture
(on i. 25), this Albani statue was brought to Paris : it is, however, not to be found in the Louvre.— A head in Palazzo
Torlonia is very similar to, but not quite identical with, the Farnese type.

THE HOPE ATHENA

77

and which is the derivative. After our enumeration of the separate distinctions the
answer can scarcely be doubtful : the Hope type is the earlier. The position of the
feet and the treatment of the drapery are evidences of an earlier stage of development ;
the long curls on the breast and the lower forehead are equally certain marks of the
same ; and in general the greater naturalness and simplicity which we have observed
in this work imply a greater degree of originality.
Pheidias himself must be the author of the Hope type. The head is another
example not only of marvellous and peculiar beauty, but of a style purely Pheidian.
In essential and personal characteristics its kinship to the Lemnia is so close as

Fig 28. — Profile of the head of Athena (A, Fig. 25).

to carry conviction. Quality of this sort is not easy to define in words ; all we can
do is to lay stress on certain details. The forehead, slightly raised in the middle, is
very delicately modelled ; the eyebrows are curved as in the Lemnia,1 the line being
more nearly horizontal than in the Athena Farnese. Thus the part between eyebrow
and upper lid is sparely modelled and slightly concave, a peculiarity which the Lemnia
and the Hope types have in common, and which distinguishes them from the
Farnese type. Again, the thin eyelids, the gentle transition between the under lid
1 It was when I -observed the analogy between the Dresden cast A and the Bologna head that it first occurred
to me that the latter was a copy of the Lemnia.

78 PHEIDIAS
and the cheek, the narrow, finely cut nose forming a decided angle with the forehead,
are as many points that resemble the Lemnia and differ from the Farnese head.
Lastly, the mysterious beauty of the mouth may be compared to the Lemnia. The
general contour of the face, too, with its slender and refined proportions, is similar.
The hair recalls the style of the Lemnia ; it springs out in soft full masses, on which
the separate lines are vigorously but not deeply cut. Although the hair of the
Farnese head is dressed in the same way, the rendering is harder and more schematic.
Compared with the Hope type the Farnese head seems rough, even coarse ; it is
also beautiful, but its beauty is as the beauty of a head by Giulio Romani beside one
by Raphael. The master who made it was on less intimate terms with nature than
the creator of the Hope type : he aims at more powerful effects, and, in doing so,
becomes more conventional and formal. The deeper mysteries of Pheidian beauty
were unrevealed to him ; his coarser sensibilities expressed themselves in ruder forms
which possibly seemed to him more appropriate to the powerful personality of Athena.
Wherever in modern art we have opportunity for comparison, we find that
between the works of the really great masters and those of their most imme
diate and favoured pupils a great gulf is fixed. However remarkable the external
similarity of their works, in the finer inner elements they are wide apart, for the
' inimitable never lies in externals.' x No wonder, then, if we recognize a decided
difference between the work of Pheidias and that of the pupil to whom we owe the
Farnese type.
The work of Pheidias should be assigned to the period shortly after the
Parthenos. The wreath of snakes round the aegis follows the new system of which
the Parthenos affords the earliest example.2 There is more animation and freedom
in the attitude, the left foot being turned more outwards and drawn farther back than
in the Parthenos. The goddess thus conceived manifests neither the serious severity
of the Lemnia nor the solemnity of the Parthenos in her hall of state. She is
gentler and softer. With an animated gesture of kindly greeting she bends her head
and steps forward to receive her worshippers. In her right hand she bore, not the
Nike of processions, but more probably only a kylix. Pier drapery is softer and
richer than in the other statues ; she wears not the peplos of the severe maiden, but
the old Ionic costume, which at that time, as we see from vases, was still in
great vogue in Athens.
It is extremely likely that the original was of bronze ; perhaps it was the very
statue mentioned by Pliny (xxxiv. 54) which Paulus Aemilius dedicated in Rome.
In the Villa Albani is a bronze head barely half life-size, broken off a statue, and
fixed on an alabaster figure to which it does not belong.3 It is a noble head of
admirable workmanship, a free rendering of the Hope Athena. The side-curls are
missing, and the hair leaves the ears freer. Otherwise not only in externals but in
expression the head corresponds with the Hope — not the Farnese type. This is a
very valuable monument, which deserves to be better known.4
In the same Villa is the well-known Athena wearing the skin helmet (Fig.
29).5 The great similarity of the body with that of the Hope statue compels us to
1 Justi, Velasquez, ii. 283. 2 Compare supra, p. 10.
3 Clarac, PI. 462 C, 902 ; Villa Albani, No. 945. Sphinx and griffins on the helmet are restored. The
alabaster figure is antique, but does not belong to the head.
4 Winckelmann, Hist, of Anc. Art. vii. I, § 21 (Vol. v. 89, note 2, ed. Eiselein), calls it 'a glorious
head of a lofty character' (tr. Lodge, ii. 385); later critics seem to have overlooked it. (Cf. Roscher's
Lex. i. 700, 23.)
5 Friederichs-Wolters, Gipsabg. 524; Helbig, Museums, 781 ; Brunn-Bruckmann, Denim. No. 226.

Fig. 29. —Athena in the Villa Albani, Rome.

So

PHEIDIAS

Fig. 30.— Head of Athena Albani. (From the cast.)

draw a close analogy between the
two. The drapery is essentially.
the same, except' that here the
diplois is longer, while the position
of the legs differs only in so far
that the left foot is rather less
drawn back in the Albani statue
than in the Hope. , Indeed, all the
earlier features which the Hope
statue shows in contradistinction
to the Farnese type appear in
rather more pronounced form in
the Albani statue. Another mark
of older origin is the thick stiff
edge of the aegis and the arrange
ment of the snakes.
This work must be connected
in some way with Pheidias. The
idea that it may have been one
of the prototypes from which he
worked may be dismissed, for how
•should Pheidias, at the highest
point of his creative power have
clung so closely to a type by
another artist ? The head-type is another proof to the contrary (profile, Fig. 30) ;
for it is remarkably severe in style, so •'that there is a certain incongruity between
head and body. This is a mark, not of an independent creative artist of the first
rank, but of an inferior worker who failed in unity of conception.1
1 The curious hide (clearly that of a wolf or dog, and not of a lion) drawn over the head deserves elucidation.
Winckelmann rightly observed that the muzzle was pointed, and called it a dog-skin, Hist, of Anc. Art. ii. 2, § 21
(=tr. Lodge, p. 184, and note, p. 428). A similar head-covering in ancient art occurs, so far as I know, only in
the following instances : (a) on a youthful male winged figure on an Attic statuette-vase belonging to the end of
the fifth century (Ath. Mitth. 1882, Taf. 12, p. 381 sqq. Mylonas) — the cap corresponds exactly to our statue ; (b)
on two images of Hades in Etruscan tombs of the fourth century (Conestabile, Pitt. xi. ; Mon. d. Inst. ix. 15) ; (c)
on a head on a coin of Amisos of the first century B.C. (Brit. Mus. Catal. Pontus, etc. , PI. 4, 3, p. xvi. , Head,
Hist. Num. 425) ; (d) it is worn by Athena on two Roman monuments from the neighbourhood of Treves
(Hettner, Rom. Steindenkmdler, No. 27, d. ; 55) : the pointed ears show quite clearly that the head is that of a
dog, and not of a lion. In (b) it is certainly the cap of Hades, the "Ai'Sos Kuviri of the Epos, which, by a play on
words, is represented as the skin of a dog (cf. Helbig, Annali, 1870, 27) ; the type must be of Greek invention.
The same interpretation is very likely correct for (a), and the daemon carrying off the girl may be Thanatos ;
in a replica in Berlin (Vases, 2906) the figure wears an ordinary Phrygian cap, and the artist probably had some
other intention. In the case of (c) also this interpretation seems the most likely, for the head, which; I think is
male, probably represents Perseus, so often celebrated on the coins of Amisos : he is wearing the cap of Hades.
The corresponding head-covering for Athena can hardly have a different meaning. The passage in the
Iliad where Athena in the conflict with Ares puts on the cap of Hades (v. 845) seems to prove that in
some cultus known to the poet Athena possessed the "Ai'Sos Kvvi-r\ as an attribute. Such, for instance, might
be the cultus of Athena Itonia, who was worshipped together with Hades in her old sanctuary near Koroneia
(Strabo, p. 411); Foucart's proposal (Bull. d. Corr. Hell. 1885, 427 sqq.) to read "ApTjs for "Ai5»js is made
impossible by Pausanias, ix. 34, 1, who calls the companion god Zeus ; Hades might be confused with Zeus,
but not so Ares. It is instructive to note that the same Homeric singer who thinks of Athena as wearing the cap
of Hades also gives her the surname 'AXaXKo/x^nis (II. iv. 8 ; v. 908) ; the ancient sanctuary at Alalkomenai was
the nearest neighbour of the Itonian sanctuary (for locality see Bull, de Corr. Hell. 1 894, 475 ) ; the group of Hades
and Athena by Agorakritos, in the Itonian temple, probably survives on a gem (Gori, Mus. Flor. ii. PI. 72,
i. ; Miiller- Wieseler, ii. 226 ; Overbeck, Zeus, 46 sqq. ; Gesch. d. Plastik, 3rd ed. i. 278) ; the two divinities
are here enthroned. Now Athena Itonia was also worshipped in Athens, and in the fifth century she

KALAMIS AND HIS SCHOOL

8 1

' The other hypothesis, that the work is by a pupil of Pheidias, is disproved by the
evidence of the head, which is of an earlier type, and has features which entirely
differ in style from the work of Pheidias. A third possibility remains, that the work
is by a contemporary of Pheictias, who belonged to a school more dependent on early-
tradition, but who made use of the creation of Pheidias as a prototype for the body
and drapery of his statue.
What this school was we shall be able to define more exactly if we look at
the head of the statue ; for the head is in style the immediate successor of a closely
connected group of monuments, of which the Hestia Giustiniani, the ' Aspasia,' x the
so-called Omphalos Apollo, and the 'Charioteer' of the Capitol are among the
most pronounced examples.2 The Albani Athena represents a later stage of the
same style. We have already assigned (p. 55) two heads of Pheidian period, the
river god of the Braccio Nuovo and the Capitol head (Fig. 11), to the artist who
made the original of the Albani statue. To these we may add, as nearly related,
the so-called Lysias of the Villa Albani (No. 62), and a beautiful head of a youth

Fig. 31. — Head in Munich.

in Munich (Fig. 31),3 of somewhat freer style. These works form a close group
sharply differentiated from- other contemporary productions by the arrangement of
the hair over the forehead, the formation of the lower part of the face, and the
peculiar expression of the mouth.
Now this school; whose earlier stage falls about 480 — 460, and whose later stage
coincided with the career of Pheidias, about 460—440, may in my view be identified
with the school of Kalamis, for this master rjf quadrigae and bigae was very probably
the author of the charioteer.4 A work like the Albani statue would certainly be
very appropriate to Praxias, the pupil of Kalamis and contemporary of Pheidias.
There are many other traces of this Kalamidian influence which kept its own
had her own treasury there, and therefore perhaps also her own image (C. I. A. i. 210), of which the Albani statue
may be a replica. On the vase-painting of the severe R.F. style (Mon. d. Inst. vi. 58, 2) there is a hint of Hades
being associated to Athena in Athens ; Athena is placed — not beside Zeus, Hera, and Apollo, but — beside Poseidon
and Hades, towards whom she turns her head. For a cult of Athena Itonia in Thessaly, cf. Paus. x. I, 10. (d)
would seem to prove that the cult of Athena Itonia made its way as far as the Roman provinces of Treves.
1 There are two good copies : (a) Berlin, Skulpt. 605 ; Arch. Ztg. 1877, Taf. 8. (b) Louvre, No. 558 (phot.
Giraudon, 1219). 2 Bull. d. Comiss. Comtnun. di Roma, 1888, Tav. 15, 16; Helbig, Museums, 597.
3 Glyptothek, 49. A very similar head in Palazzo Torlonia.
4 For the Omphalos Apollo as presumably a work of Kalamis cf 50th Berl. Winckelmannsprogr. p. 150.
M

8, PHEIDIAS
side by side with that of Pheidias. For example, in the interest! ng ' Demeter ' with the
veil1 of the Berlin gallery pose and drapery has been influenced by Pheidias, and
seems to be more recent than the Parthenos, while the head is a direct successor of the
' Aspasia ' type.
The Herakles head of severe style, also in Berlin2 (Fig. 32), while belonging
to this group, has a character and individuality of its own. I believe that the body
corresponding to this type of head survives in a Herakles torso of the Louvre » (Fig.
33) which mav be regarded as a direct development of the Omphalos Apollo.
The position and attitude are the same, the bodily forms also are similar, though
rounder and freer. Herakles has tied the lion-skin round his neck, the end of it falls
over his outstretched left forearm, the right hand rests on the club ; the hind at the
left has certainly been added by the copyist.
We must now return to Pheidias, and try to gain some more exact knowledge
about the pupil who created the Farnese Athena type. The great number of existing
copies justifies the conclusion that the artist was famous and distinguished.
Here too we must be guided by the type of head, always the part of a statue on
which personal style is most clearly impressed. The nearest parallels to the Farnese
type are three closely related heads, each of them preserved with its statue. These
are the Capitoline 'Hera' or ' Demeter,'4 the 'Venus Genetrix/5 and the Barberini
' Hera ' of the Vatican.6 The three are remarkably alike in arrangement and rendering
of hair, and in the essential characteristics of the facial forms— such as the outline of
the rather broad face, the form of the forehead, the attachment of the nose, the mouth,
and the somewhat heavy eyelids— except that the Aphrodite has been distinctly
modified through stress being laid on the qualities of grace and charm. To these
three sisters we may now add the Farnese Athena as a fourth.
Many have agreed with me in identifying the ' Genetrix ' with the famous
Aphrodite of Alkamenes ; this point may now, I think, be considered fairly settled.7 In
the Capitoline statue Petersen conjectures we have the Hera of Alkamenes.8 I cannot
feel quite sure that the work is a Hera,9 but it is certainly by the artist of the Aphrodite.
Lastly, the Barberini ' Hera,' usually but incorrectly supposed to be of much later
1 Skulpt. 83 ; Overbeck, Atlas d. Kuntsmyth. Taf. xv. 25. In a painting by Michelangelo Cerquozzi
(1600— 1660) in the Cassel gallery, No. 516, this statue is represented standing on a pedestal in a garden ; the
arms are still unrestored.
- Skulpt. 188. The bust seems to be identical with the one published in the Gall. Giustin. 2, 26. The
head is antique only as far as the neck, which is of Parian marble, while the restored bust is of coarse inferior
bluish marble. The head was originally turned to the right, as may be seen by the fact that the right ear is
carelessly worked, and the left ear well rendered.
3 No. 1404, Salle des Caryatides.
4 Overbeck, Atlas d. Kuntsmyth. xiv. 13, 20 ; Antike Denkm. i. 55, I ; Helbig, 507.
5 Antike Denkm. i. 55, 2 ; Gas. Arch. 1887, PI. 30.
6 Helbig, Museums, 301 ; Overbeck, Atlas, ix. 10, x. 33.
7 Cp. supra, p. 9, note 3. The doubts expressed by E. Reisch (Eranos Vindobonensis, p. 18 sqq.) are
easily refuted. He asks why, out of the great number of Aphrodites of the Pheidian epoch known from
literature, only the Aphrodite of Alkamenes should be taken into account in estimating the Genetrix figures.
But that ' great number ' consists solely of the two Aphrodites by Pheidias in Athens, and in Elis, and of
the Aphrodite iv ktjVois, by Alkamenes. R. argues further that this garden Aphrodite was an Aphrodite
Ourania, and as the impersonation of ' heavenly ' love must have looked more matronly than the Genetrix ;
but, first, we do not know that the Aphrodite iv k^ttois was surnamed Ourania ; and, secondly, even if she were,
that ethical meaning of the surname is a philosophic notion dating not earlier than Plato (cf. Plato, Symp. viii. D ;
Preller-Robert, Gr. Myth. i. 355), and can therefore have had no influence upon Alkamenes.
8 Rom. Mitth. 1889, 65 sqq.
9 The figure on the relief mentioned by Petersen cannot be a copy of the statue. It only shows that Hera
was represented in the same costume, and we know from the monuments that the same type was used for different
goddesses.

84 PHEIDIAS
date, shows clearly by the type of head 1 that it is a work of the same artist, nor is
there anything in the style of the body to contradict this view.2 The transparent
chiton is treated as in the Aphrodite, and the piece slipping down over the left breast
betrays the same taste. Whether the statue was originally intended for a Hera is a
question we must set aside for the present.3
The Farnese Athena, then, is probably also a work of Alkamenes, doubly
valuable to us because it shows in what relation the artist stood to his master
Pheidias. We have seen that in the head-type he followed out his own ideal— an
ideal inferior, it would seem, to that of Pheidias. The Aphrodite and the Athena
show that within its limits he could individualize strongly. He introduces the
complete walking motive — (in this respect the Athena and the ' Hera ' Barberini
are closely allied)— and everywhere strives after stronger effects than the master—
a tendency which removes him proportionately far from the simple observation
of nature.4 According to an anecdote which has only reached us in a late and evidently
exaggerated form,5 but which undoubtedly contains a kernel of truth, Pheidias and
Alkamenes made rival statues of Athena for a competition in which Pheidias
obtained the prize. The story cannot be a mere late invention ; it evidently underlies
the statement made by Pliny, and more in detail by Pausanias,6 to the effect that
Alkamenes was a contemporary and rival of Pheidias, but obtained in competition
with him only the second prize in respect of knowledge in the making of statues.
Now the story of such a competition might easily be invented by a later generation,
if two Athena statues very similar to each other were in existence, one by Pheidias
and one by Alkamenes. We have now sufficient evidence from the Hope and
Farnese statues for the assumption that this really was the case.
Alkamenes seems to have been a very productive artist,7 whose works were
much prized and extensively copied in antiquity. He must be largely represented
among existing antiques.8 A beautiful head over life-size, an admirable copy,
1 The head of the Vatican statue is not broken, but inserted with a piece of the chest, very well preserved.
A good replica of the head is in Museo Chiaramonti, No. 511 A ; Overbeck, Atlas, ix. 11. The head of the
replica formerly in the Villa Borghese, and now in Copenhagen, is the bust ; it has never been separated from the
statue ; no diadem.
- The Palatine torso, now in the Terme, is not an exact copy, but adds Hellenistic details.
3 The sarcophagus (Mon. d. Inst. iv. 9) quoted in support of the theory gives Hera an entirely different type.
4 On the ground of the statues just discussed some few more works may be ascribed to Alkamenes : e.g. the
type of the so-called Hera of Ephesos known unfortunately only in a torso (Friederichs- Wolters, 1273), which
exists in several replicas (cf. Lanckoronski, Pamphylien u. Pisidien, i. p. 94) ; it was also used in Imperial
times for portraits. — The much-restored Stockholm head, J. H. S. ix. 4 (p. 35, Farnell), belongs also to this
series. 5 Tzetzes, Chil. 8, 340 seq., 353 seq. ; Overbeck, 5. Q. 772, 810. Cf. supra, p. 21, where it was suggested
that the first part of the anecdote may have arisen from the existence of a statue by Pheidias set upon a column.
8 Plin. xxxiv. 49 : Pheidias flourished 01. 83, quo eodem tempore aemuli eius fuere Alcamenes (the following
names are arbitrarily added by Pliny ; cf. my essay, Plinius u. Seine Quellen, p. 22 seq. ) Paus. v. 10, 8 : ' ' AXicapivovs
dvSpbs yXiKiav te Kara. ^etSiav Kal Sevrtpua iveyxa/xivov ootplas is iroiriiriv ayaX/xdruv. This passage has been
much misunderstood (also by me, Preuss. Jahrb. vol. li. 380 ; the later literature is quoted Jahrb. d. Inst. 1890,
p. 97, note 37), owing to the connexion with that anecdote not being recognized. — Plin. xxxvi. 16, where
Pliny expressly names Alkamenes as pupil of Pheidias, and only as such, goes back to another source.
7 Opera complura in the temples of Athens, Pliny, xxxvi. 16.
8 Lately Winter thought he had discovered two originals by Alkamenes (Jahrb. d. Inst. Anz. 1894, p. 44,
46) ; but the statue of the Akropolis cannot represent Prokne and Itys. It is an insignificant coarse piece of
work ; it is quite impossible that Alkamenes should have dedicated such a monstrosity to the goddess (Paus. i.
24, 3), or exposed it to the ridicule of his fellow- artists. The other statue from Pergamon is obviously not
a fifth-century original, but only the copy of one ; its workmanship is quite similar to that of the Athena
of Pergamon (supra, p. 27). The treatment of the drapery differs totally from that of the originals of the
Pheidian school.

AGORAKRITOS 85
certainly after Alkamenes, is reproduced in Fig. 34.1 The body is unknown. The
hair is gathered into a coif like that of the Aphrodite and of the ' Hera ' statues ;
the stylistic treatment of the waves of hair, the thick eyelids, and the shape of the
ear are the same. It is difficult to decide whom the head represents. The form of
cheeks and chin and the full mouth resemble the Farnese Athena, so that the head
very probably represents a bareheaded Athena.2
The works of Agorakritos, Pheidias's second pupil of distinction, must have been

Fig. 33. — Torso of Herakles, Louvre. (From a photograph by A. Giraudon.)
closely allied to those of Alkamenes. This is clear from the remaining figures of the
basis of the Nemesis of Rhamnus : their type of face and the rendering of the thin
transparent chiton 8 recall the works we assigned to Alkamenes. Among the statues
1 Berlin, Skulpt. 608. Length of face 32 cm. The end of the nose is restored. Compare L. Mitchell,
Hist, of Sc. p. 321, and Roscher's Lexikon, i. 413, Z. 1. A head on a statue in the cortile of the Pal. Borghese
(Matz-Duhn, 1374) appears to be a replica.
2 In Roscher's Lexikon, loc. cit. , I was misled by the stylistic resemblance of the head to the Genetrix into
thinking it an Aphrodite, but the forms are too massive and powerful, the carriage too vigorous and erect, for the
ipiXofifieiSiis. 3 'Eipri/x. dpx- 1891, Taf. 8, 9; Jahrb. d. Inst. 1894, PI. 1—7. Cf. especially 'E</>. 9, i=fahrb. i. 6, with the
Hera Barberini and the Ephesian torso. In the Jahrb., loc. cit. p. 16, Pallat makes some correct observation

Fig. 34.— Head of a goddess (Berlin).

Fig. 36. — Apollo Barberini (Munich).

Fig. 35. — 'Ceres' in the Rotonda of the Vatican.

Fig. 37. — Athena in Capitoline Museum.

PHEIDIAS

represented by existing copies there are three which, although they resemble the work
of Alkamenes and are certainly to be referred to a pupil of Pheidias, yet point to
a different personality. They form a closely
compact group. The heads have been pre-
/_.;.-_, -.-.., \ served with the bodies. I mean the colossal
' Ceres ' in the Vatican (Fig. 35),1 the Barberini
Apollo in Munich (Fig. 36J,2 and an Athena
in the Capitol (Fig. 37).s Even external
details, such as the peculiar wide girdle, the
folds in the hem of the diplois, besides the
rendering of other parts of the drapery and the
attitude, prove that the three works are all by
one man. I am inclined to name the artist
Agorakritos, and to see in the ' Ceres ' his
famous Nemesis.4 This artist conceives divini
ties otherwise than Alkamenes : his gods walk
with solemn processional gait, looking straight
out, with head erect. The artist aims at giving
a general effect of majestic dignity rather than
at individualizing like his fellow-pupil.
Statues of male divinities by Alkamenes
must certainly exist. I believe that the He-
phaistos admired by Cicero survives in a beauti
ful statue at Cassel. Unfortunately it is only
a torso, but the workmanship is unusually
good (Fig. 38).5 That it is a Hephaistos and
a work of the Pheidian school seems to me
self-evident. The torso wears the exomis of
the craftsman, girt with a leather strap. On
the left shoulder lies a small cloak which fell
down over the outstretched forearm.6 The
chiton is treated in the style of the Parthenon
frieze.7 The torso stands in a specially close
relation to the Orpheus relief, another monu
ment of this epoch. Here the clinging drapery,
the fall of the folds between the legs,8 and
the delicate folded hem on the thighs and

Fig. 38. — Conjectural restoration, showing the torso in
Cassel combined wiih the head of the Hephaistos
Chiaramonti.

on the difference between the treatment of drapery still observed on the Parthenon frieze and that of the
Agorakritos basis and other later works.
1 Helbig, Museums, 297; Brunn-Bruckmann, Denkm. No. 172.
2 Munich, Glypt. 90. Replica, Bull. Commun. 1887, Tav. 20, 21.
3 In the lower hall, No. 3 ; Clarac, 461, 858. An important work with a genuine head spoiled by the
modern curls on the shoulders ; behind the ears are remains of the original curls. The head is very like the
Barberini Apollo.
* The head seems to correspond with the original fragment (Brit. Mus. Cat. 460 ; cf. Rossbach, Ath. Mitth.
xv. 46) ; a large diadem formerly rested on it. The hair is simply taken up behind, the left arm is wrongly
restored as raised.
6 Cassel, ii. 9. The statue was a short time in Paris, and is engraved by Bouillon, ii. stat. Thisie. The right
interpretation as Hephaistos is hinted at in his text. The beautiful youthful head does not belong to the statue.
Our sketch, Fig. 38, gives the antique parts without the restorations ; marble Pentelic, workmanship fresh and good.
6 The beginning of the elbow is still in existence. 7 Cf. north frieze, 133 ; west frieze, 23.
8 Best example Villa Albani, 1031 = Helbig, Museums, 790 ; other replicas in Louvre, Naples, etc.

HEPHAISTOS BY ALKAMENES «9
upper arm 1 are so surprisingly like the torso that one is inclined to assign both
works to one artist. The peculiar formation of the body proves that the torso
represents Hephaistos. The chest is broad and robust, and what is left of the
arms (especially the right) shows a powerful muscular development. On the other hand,
it is noteworthy that the remains of the legs — i.e. the knees and parts immediately
around them — are unusually small and slender. Now the combination of a strong
body with weak legs is a characteristic of Hephaistos already established in the
Homeric poems.2
An Hephaistos from the school of Pheidias, as we have seen the Cassel torso to
be, could hardly be any other than a copy of the famous statue of Alkamenes formerly
in Athens. We know from Cicero 3 that the figure stood with both feet planted on
the ground. So did the torso ; the weight of the body is on the left leg, and the
right foot is placed at ease a little to one side, but with the sole full on the ground.
Cicero says the figure was draped, and in this point also the torso corresponds. The
attitude of repose notwithstanding, Cicero professes to see the lameness of the god
plainly indicated in the statue ; but he remarks that this is done without detracting
from the beauty of the figure. It seems clear from this that the figure must
have resembled our torso. Like other authors who wish to say something witty
and epigrammatic about a work of art, Cicero perhaps read rather more into the
statue than the unbiassed observer will find there. Strictly speaking, a limp in the
gait could not be expressed in a figure standing on both feet ; we may be sure that
Alkamenes did not wish, any more than the Attic vase-painters, to make the
god limp, but he suggested the characteristic peculiarity in a sufficiently true and
natural manner by the contrast between the slender legs and the powerful body. The
head, now lost, no doubt gave more effect to this contrast by its broad, strong forms.
I believe that the head-type survives in the well-known term of the Vatican, so
eloquently analyzed by Brunn.4 In order to show how well this head would suit
the torso, I have combined them in the drawing.5 The term is only a poor, shallow,
and sketchy copy, worked with a plentiful use of the drill. Yet in the rendering
of the curls of hair, in the massy arrangement of the beard, and in the large eyes it is
easy to recognize the style of the later Pheidian epoch. The form of the hair rising
on the forehead is very like the Asklepios head of the same school.6 We need hardly
lay stress on the masterly way in which Alkamenes has individualized the head.7
The artist shows himself no less great in the creation of his Ares. The statue
by Alkamenes in the temple at Athens is preserved to us, as I think we may now
assume, in the famous Ares Borghese and its replicas.8 The head is stylistically
1 Cf. the striking agreement between the right upper arm of the Hermes on the Albani replica and the left
upper arm of the torso. 2 Cf. Roscher's Lexikon, i. 2039.
3 Cicero's words are: Volcanum . . . in quo stante in ulroqtte vestigia atquevestito leviterapparctclaudicatic,
non deformis. Reisch, loc. cit. p. 21, suggests that this Hephaistos was identical with the temple-image mentioned
Paus. i. 14, 6, as standing together with Athena ; he further identifies these two statues with the two images
named in C. I. A. i. 318, 319, which must have been completed in42i — 420. His further attempt to recognize the
Athena in the distinctly Praxitelean Borghese statue (Helbig, Museums, 935) is not so happy.
4 Mon. d. Inst. vii. 81 ; Brunn, Gbtterideale, Taf. ii. ; Profile, p. 23 ; Helbig, Museums, 89. Loschcke refers
the head to Alkamenes.
s The head of the torso was evidently turned as it is in the term. It is true that the neck of the term is
thicker than the neck of the torso, to judge from what remains. But the copyist may easily have increased the
size of the neck in order to adapt it to the terminal bust. The head is not broken, and the greater part of the
term is antique.
6 Cf. especially the Palatine head, now in the Terme ; and the head of the Naples statue, Roscher's
Lexikon, i. 634.
7 The inequality of the two sides of the face (Brunn. loc. cit. p. 25) is due to careless execution.
8 Conze (Beitrdge, p. 9, note 2) already inclined to ascribe this work to Alkamenes. The ring above the
N

90 PHEIDIAS
analogous to the Hephaistos, the body and the attitude are both purely Attic.
The Diskobolos ' taking up position,' 1 long ago ascribed to Alkamenes, is its nearest
parallel. It is also instructive to compare this Ares with a helmeted and bearded
head,2 of which the best replica is the so-called ' Miltiades ' in the Louvre, a work-
belonging to the period of the Parthenos, and so pronouncedly Pheidian in character
that it may with some reason be ascribed to the master himself (Plate IV.)
Alkamenes thus becomes a personality we can grasp.3 We have seen what his
relations to Pheidias were, and how his creative instinct asserted itself independently.
None of the works referred to him are earlier than the late period of Pheidias — i.e.
than the epoch of the Parthenon frieze. What we otherwise know about his life
shows that his activity must have extended to the close of the century, when he
executed the commission for Thrasybulos (Paus. ix. n, 61), about 403 B.C., and made
the Dionysos for the new temple beside the theatre at Athens. According to
Dorpfeld's observations of the material used in the foundations, this temple must
be later than the Periklean buildings.4
To return to the master himself — for we have not yet enumerated all the works
which may be brought into direct relation with him.
Even among the Athena heads there is one more which is not, like so many others,
a derivative from the Pheidian types, but apparently a new creation of the master's,
belonging to the period when he was working on the Parthenos. The magnificent
head, intended to be inserted into a life-size statue, is at present in the museum
at Brescia (Fig. 3C5).5 A helmet, presumably of metal, was placed on the head,
the back of which was only roughly hewn out ; only the edge of the helmet is
indicated just on the forehead. The curls escaping from under the edge of the
helmet, and spreading as far back as the ears, are remarkably like those of the
Parthenos, and are purely Pheidian in style.6 The wavy hair which falls closely round
the neck, although a new feature for Athena, is most appropriate to the maiden
goddess, and occurs besides in two works of Pheidias — the Terme Apollo, and the
Zeus of Olympia. The profile of the forehead which slightly swells out below, the
right ankle, which occurs only in the Borghese, not in the other replicas, must have been added by the copyist to
illustrate the favourite story of Ares bound by Hephaistos. Immediately above the ring on the inner side of the
leg is a hole (antique) filled with lead. This probably served to fasten some part of the fetter. The copyist
wished to represent Ares in love, a favourite conception of later times. He did not understand that the
melancholy earnestness in the face was appropriate to the god of war, and had nothing to do with love-
stories- l Helbig, Museums, 331.
2 Cf. p. 36, note 1. Replicas of less importance in the Capitoline Museum (Helbig, 488) and in
Pal. Colonna. The helmet resembles that of the Parthenos ; the beard, that of the Poseidon on the Parthenon
frieze. 3 The belief in the existence of an older Alkamenes is based on an old suggestion of mine (Ath. Mitth.
1878, p. 194 ; cf. Loschcke, Westl. Giebelgr., Dorpater Progr. 1887, p. 7). But the only testimony which could
make this assumption necessary— that of Pausanias concerning the artist of the west pediment at Olympia— seems,
like the corresponding information about the east pediment, to be a random statement. What Pausanias remarks
in the same passage on the relations of Alkamenes to Pheidias is, as we saw above (p.- 84), nothing but a
jumbled reproduction of the anecdote respecting the rivalry of the two artists. Further, Kopp (Jahrb. d.
Inst. 1890, p. 277) has shown that the story of the destruction of the temple of Hera (Paus. i. 1, 5) by the
Persians is a mere invention : the Hera therefore is of no use as a date, any more than the Hekate, as
we shall see later on (Appendix).
4 Kindly communicated to me by Dorpfeld ; cf. Jahrb. d. Inst. 1890, 276, note. A full exposition will be
given in Dorpfeld's forthcoming work on the theatre. Cf. E. Reisch in Eranos Vindob. 1893, p. I. He dates
the Dionysos on historic grounds, B.C. 420—413.
6 Dtitschke, Oberitalien, vol. iv. No. 335 ; Labus, Museo Bresciano, i. 44, 2. Length of face, 0-17. A
good copy. s
6 Dtitschke, loc. cit., compares this to the Bologna head (i.e. the Lemnia).

Fig. 39. — Head in Brescia. (By permission of Messrs. Bruckmann, Munich.)

92 PHEIDIAS
obtuse angle formed by nose and forehead, the slightly receding lower part of
the face, the formation and expression of the full mouth and even the elongated ear-
are, as is evident from a comparison with facts so far established, purely Pheidian in
character. The head is decidedly later than the Lemnia, and perhaps even than the
Hope Athena ; the characteristic part between upper lid and eyebrow is, especially
towards the temple, much softer and more naturally modelled than the same part in
the Hope Athena : if the copies are trustworthy in this point, it follows that the
Brescia head is the more recent of the two. The whole expression bears out this
view. The quiet repose and restraint to be observed in the Lemnia and the Hope
Athena are lacking here. The carriage of the head is freer and bolder, the face shows
the power and full proportions which apparently were characteristics of the Parthenos.
In this series may be further placed a beautiful head of a youth with a helmet,
known from two copies in the Louvre (Fig. 40) 1 and the Museo Torlonia respectively.2
In the latter replica the frontlet is adorned with a design in relief of a little Eros
surrounded by sprays. This, although of course an addition of the copyist, enables
us to determine the original meaning of the head. It must represent Ares, for Ares
in love was a favourite theme of later artists ; the copyist's addition here is to be
explained much in the same way as the fetters on the Ares Borghese (p. 89, note 8).
The ears, swollen as from the boxing contest (especially clear in the Torlonia copy),
are quite appropriate to the war god. I know no other copies of the head, and no
torso which could belong to it. A helmeted head in the Lateran,3 placed on a statue
that does not belong to it,4 is closely analogous to the Louvre head ; but it is only a
variant, and not a replica.5
A colossal Ares statue in the second cortile of the Palazzo Borghese is another
interesting work (Fig. 41).6 The head, which has never been broken, shows many
points of similarity with our Paris type, though it is not a replica. It is turned
more decidedly to the left, the edge of the helmet on the forehead is different, and
the hair, notwithstanding a general resemblance, varies in detail ; finally, " slight
whiskers have been added similar to those worn by the Ares Borghese of the Louvre.
The motive of the statue is much the same as that of the Munich Diomede, a work to
be assigned, as we shall try to show later, to Kresilas. The right hand is lowered, and
doubtless held the sword, while the empty scabbard is suspended on the left side ; the
left hand probably held a spear. The chlamys falls from the shoulder in perfectly
simple folds.
1 Louvre, No. 2986, in the same room as the Ares Borghese. Thasian marble. First discussed by Treu,
Arch. Anz. 1889, 57. He referred it to Polykleitos. I subsequently thought of Kresilas (Arch. Anz. 1891, 36).
Restored in plaster : nose, left upper lip, most of left eyebrow, and left upper eyelid.
2 Museo Torlonia, Tav. 26, No. 104. Galleria Giustiniani, ii. 45 (here the Eros in front is wrongly drawn
as a sphinx ; on the helmet appear the antique remains of an animal couchant to serve as plume-bearer ; in the
Museo Torlonia the figure has been restored). The nose and chin are restored. The points of hair over the
forehead are broken away.
3 Helbig, Museums, 638.
4 Also observed by Overbeck, Ber. d. Sachs. Ges. 1861, p. 80. A replica of this torso in Lansdowne House
(Michaelis, p. 453, n. 63) preserves the original head, which is that of Marcus Aurelius. From the workmanship,
which belongs approximately to the period of this Emperor, it is probable that the Lateran statue also represented
Marcus Aurelius.
5 As Treu had pronounced it to be, loc. cit.
" Matz-Duhn, Zerslr. Bildw. No. 1338. The figure is much defaced, owing to exposure ; the head was never
broken. Restored : nose and chin (lips formerly), right arm, the left hand, and the lower part of the body. The
latter was probably made to replace a still existing but battered original. This seems evident from the complete
correspondence in style, and from the fact that a piece of unbroken drapery belonging to the antique upper part
of the torso hangs down below the join, which would hardly be possible if the whole of the lower part of the
body had not been originally preserved.

Fig. 40.— Head ot Ares, Louvre. (From a photograph by A Giraudon.)

94 PHEIDIAS

We possess, then, three variants of the same type : (a) the variant represented by
the Paris and Torlonia heads, body unknown ; (b) the Lateran variant, body unknown ;
(c) the variant in the cortile Borghese, of which both head and torso are preserved.1
I do not think that these variants are merely due to the copyists. The Borghese
statue and the Paris head give the impression of being essentially faithful copies
from fifth-century originals. Of these originals we may suppose that one was a new
creation by a distinguished artist, and the others more or less dependent conceptions

Fig. 41. — Ares in the Pal. Borghese (Rome).
by inferior artists of the same period. Now there can be no doubt that of the three
variants the Paris head has the best claim to represent the original creation. It is
analogous to the Parthenos, and must be a product of the circle of Pheidias. The curls
are bunched out in front of the ears and partly cover the edge of the helmet, in a
manner marvellously similar to that shown in the Parthenos and in the Brescia head
1 The head of the so-called ' Theseus' at Ince Blundell Hall (Michaelis, p. 351, No. 43 ; Arch. Ztg. 1874,
Taf. 1) may be a fourth variant. I have not seen the original, but Michaelis doubts whether the head belongs to
the body ; or the latter, cf. infra, p. 359, n. 3.

THE DIOSCURI OF MONTE CAVALLO 95
discussed above. The ' Aphrodite ' (p. 68) and the Anakreon should also be compared.
The treatment of detail is somewhat freer than in the other works ; the Ares must
at any rate be later than the Parthenos. The helmet is another point of likeness :
like that of the Parthenos, it has three plumes, the middle one supported by a
sphinx,1 and turned up cheek-pieces decorated with griffins rampant.2 I can see
nothing to prevent our attributing the original to Pheidias himself. Individuality
manifests itself specially in the lower part of the face ; the distance between nose and
mouth is rather longer than usual, and the lower lip recedes ; the mouth is close,
the chin firm, the forehead broad and powerful. The head is held upright with an
easy turn to the right. It conveys an impression of youthful defiance and brave
resolution, of passion latent in sensuous beauty. And these are traits eminently
characteristic of Ares.
The Ares Borghese of the Louvre, which we have attributed to Alkamenes and
compared to a bearded and helmeted Pheidian head (Plate IV.), should now be
compared with this Pheidian Ares : it is clear that the system of form common to
the Parthenos and to the Pheidian Ares is very different from that affected by the
younger artist. Starting from the same principles, the later work has developed
and emphasized the older conception, and brought in new elements. The Ares of
Alkamenes seems restless and excited beside the other, and more stress is laid
on the gloomy wildness of the god's personality : instead of the carefully dressed
curls which belong to the style of the Parthenos, the hair hangs straitly, giving an
effect of neglect and carelessness intended perhaps to indicate the Thracian origin of
the god. X. The Dioscuri of Monte Cavallo, and the Elder Praxiteles.
We have kept for the end a magnificent creation of Pheidias belonging to the
last period of his activity. Not only is this work easily recognizable from the style,
but it has the additional advantage of being authenticated by an inscription. I refer
to one of the two colossal figures on the Monte Cavallo in Rome — to the one holding
the horse with the right hand, whose basis bore till 1589 the antique inscription3
opus Fidiae.
Strangely enough, this work has been so misunderstood that even at the present
day people are astonished if it is ascribed to the artist whose name is attested by the
inscription. This inscription is of a kind of which there are many examples in Rome. They
are preserved on the bases of statues, but the statues themselves are lost. These
works, designated as opus Polycliti, opus Bryaxidis, opus Praxitelis, opus Tisicratis,
opus Timarchi? were, as no one disputes, either originals by the artists named, set
up again in Rome, or copies from these originals. The similar inscription Kvaitrirov
epyov on a copy is definite evidence for the last case,5 sometimes the formula 6 Seiva
1 Cf. p. 92, note 2.
2 The legs of the griffin are still to be seen on the Louvre head. The cheek-pieces of the Borghese head are
ornamented in the same way.
3 It is well known that when the Colossi were set up again in the year 1589 the two artists' inscriptions
were not only renewed but reversed. The antique inscriptions designated the figure holding the horse with
his right hand as the work of Pheidias, the one holding the horse with his left hand as the work of Praxiteles.
We keep exclusively to the antique designations.
4 Cf. Lbwy, /. G. B. 489 seq.
' Ldwy, 506 ; for the genuineness cf. Hermes, vol. xxii. 153.

96 PHEIDIAS
eiroiei is used on the copy to designate the artist of the original.1 The fact that the
Colossus of Monte Cavallo can as a matter of course not be an original by Pheidias
is no evidence against the correctness of the inscription, since copies were marked in
the same way. Nor should the date of the inscription (late Imperial epoch) 2 arouse
any suspicion as to its exactness. The two Colossi originally formed the decoration
at the entrance of a large building : 3 to judge by their workmanship, they cannot
be later than the earliest period of the Empire. In more recent times they were
placed as nearly as possible in their original positions on a massive pedestal of
masonry into which were built older architectural fragments,4 probably belonging
to the building at whose entrance the Colossi at first stood. The pedestal was
faced with marble slabs, and on these stood the inscriptions in fine large letters.6
It is of course possible, though very unlikely, that the inscriptions were invented
at that time. The fact that so much trouble was taken to re-erect the statues
on large and handsome pedestals proves that they were works of some note, and
that the inscriptions were merely renewals of older ones. Such renewals must
have been frequently necessary in ancient Rome, where works of art were constantly
changing locality. The ascription of these two statues to their respective artists
probably took place in the same early Imperial period to which the precisely similar
inscriptions quoted above belong. There is no reason to suppose that any one of
these was arbitrarily invented : on the contrary, all the facts seem to point to the
conclusion that such inscriptions formed a storehouse of authentic material on which
were based many of the literary notices of works of art in Imperial Rome.
Only in one case could we be justified in rejecting the testimony of the inscription
— i.e. if the style of the statue were absolutely at variance with the style of the artist
whose name it bears. Now it will not be difficult to show that exactly the reverse is
the case. It has, it is true, been maintained that the style has nothing to do with Pheidias.
Martin Wagner the sculptor, the first who made a close study of the style,6 confines
himself to combating the opinion that the statue was an original by Pheidias : he
misses the ' geistreiche Auffassung der Natur ' to be seen in originals, and finds here
only a ' mehr systematischen, auf Regeln gegriindeten Vortrag ' ; according to him the
works are Roman. Otfried Miiller7 was more definite, and maintained the style to
be not Pheidian but Lysippian — a theory which has found many adherents. It is
strange that the eye can be so deceived ! For in the whole range of ancient art of
the free style no stronger contrast exists than that between these statues and the
works of Lysippos. Any one who will take the trouble to examine the matter for
himself, even only through the medium of photographs, will I am sure be convinced
that this is so.
The two Colossi of Monte Cavallo (quite irrespective of the inscriptions) can be
accounted for historically and artistically in no other way than by referring them to
the circle of Pheidias, within which they more especially belong, to the epoch of the
Parthenon friezes and pediments. Placed side by side with these originals, the
Colossi at first naturally suffer by the comparison. They lack the spontaneity and
refined execution which form the peculiar charm of first-hand work. While in the
Parthenon figures the muscles seem to show naturally under the skin, in the Colossi
1 Cf. e.g. Ldwy, 488 ; Rom. Mitth. 1891, 323, etc. 2 Cf. Lowy, p. 326.
3 Fogelberg, Ann. d. Inst. 1842, 202. * From the testimony of Flaminio Vacca.
5 Cf. the engravings of Lafrerie. « Kunstblatt, 1824, p. 373 sqq., 381 sqq.
7 De Phidiae Vita et Op. i. §8, note (Kunstarch. Werke, ii. 13).

THE DIOSCURI OF MONTE CAVALLO 97
they arc more drily rendered, ' mehr schulgerecht,' as Wagner expressed it. But
the same laws of style arc obeyed. The different parts of the body are marked
off in large distinct planes — in direct contrast to the rounded modelling and fine
transitions of fourth-century technique, which reached their highest development
under the influence of Lysippos. The structure of the mighty chest is specially
characteristic. In the metopes of the Parthenon, even in those of freer style, the
chest is less developed than in the figures of the frieze and pediments, to which
the Colossi correspond. Further, the manner in which the edge of the ribs projects,
and the modulation of the abdominal muscles (cf. especially the deep depression
between the straight and oblique muscles, and the inscriptiones of the former),
are precisely the same on the Colossi and the torsos M and H of the west and
H of the east pediment of the Parthenon.1 The evidence of the heads is equally
plain (Fig. 42). In this respect also the youths of the Parthenon frieze afford the
nearest analogy — the heads turned to the front and wearing a slightly excited
expression, such as No. 2 of the west frieze (PL V.), are surprisingly like the Colossi.
For the profile, No. 118 of the north frieze should be compared. The shape and
modulation of the forehead, the sharp edge of the eyebrows, the wide-open eyes, the
complete absence of all those natural forms which art after Praxiteles and Lysippos gave
to the parts around the lower eyelid both in the region of the nose and of the temples,2
the formation of the mouth and chin, the characteristic wavy lines of the hair which
frames the face as with a crown of rays, even the way the fillet cuts in and the
hair rises on either side of it — all this is as genuinely Pheidian, and of the Parthenon
period, as it is diametrically opposed to the style of Lysippos. Each one of the
forms analyzed would be strong proof by itself ; taken all together, they remove all
possibility of doubt. Only the difference in size, and the fact that the Parthenon
frieze is original work while the Colossi are copies from bronze, make some variations
inevitable ; in artistic essence and individual style they are the same.
Again, it is striking how nearly the horses of the Colossi correspond with those of
the Parthenon frieze, and how widely they differ from those of later art — compare,
for instance, the sculptures of the Mausoleum and the basis of Bryaxis,3 or, as a stronger
contrast still, the horses on the Pergamene reliefs. In the proportions of the bodies,
in the relation in size to the human figures, in the details of form — especially in the
head — in the shape of the eyes, the jaw, the nostrils, and particularly of the lower half
of the mouth — these horses of the Colossi resemble those of the Parthenon frieze and
pediments, except that they lack the fire and spirit of the originals. I say nothing
of the armour and drapery. It has long been acknowledged that the coat of mail is
an addition of the copyist, who wanted a marble support for the outstretched leg
of the figure. The piece of drapery hanging from the arms, which is evidently meant
to serve the same purpose, is most likely another addition of the copyist.4 It is not
worked as clearly and definitely as the other parts of the figure.
1 Even the pubes is characteristic, and quite different from the Lysippian.
2 Precisely those peculiarities which Hugo Magnus (Die Darstellimg des Auges in der Antiken Plastik, 1892)
dwells on are quite decisive in favour of assigning the works to the Pheidian epoch. They consist in the
extremely sharp arch of the eyebrow and of the upper rim of the eye-socket, in the modelling of the space
between that upper rim and the upper eyelid, and in the absence of any indication of the lower rim of the eye-
socket. It is strange after this accurate analysis that Magnus (p. 76) should separate the figures of Monte
Cavallo from the Pheidian epoch : he was misled by the deep hollowing out of the iris and the pupil ; but this is a
most unessential detail ; the copyist added it because in such a colossal work he could not rely on the effect of a
merely painted eye-pupil (cf. Conze, Sitzungsber. Berl. Akademie, 1892, Febr., p. 49 seq.) ; the bronze originals
had of course inserted eyes.
3 Bull, de Corr. Hell. 1892, PI. 3, 7. 4 Cf. p. 56, note 1.
O

THE ELDER PRAXITELES

99

Finally, the motive of the youths with their horses and the rhythm of their
movement are special to the Parthenon and its artistic circle. It has long been
noticed that one group, the youth leading the horse with his right hand, reappears
in very similar form on the west frieze of the Parthenon (No. 27).1 And the
same figure — the one designated by the old inscription as the work of Pheidias —
occurs again in all its essential features on another authenticated work of Pheidias,
the relief on the basis of the Athena Parthenos.2 The only distinction is that the
youth on the basis leads with his right hand not one horse but a team of four. The
horses are rearing in both instances. If then it be admitted that Pheidias used
the 'horse-taming' motive, the question next arises whether it was his own inven
tion or borrowed from another artist. All the evidence seems to be in favour.
of the former view. The motive in its principal features — a broad swinging stride to
the right, the head thrown well to the left, one arm (usually) raised, the other bent
at an angle — enters Greek art as a complete idea, and from the time of its entrance
becomes a common property to be employed over and over again. It is quite unknown
before the time of the Parthenos. It first appears in that free and animated style
which we have learned to recognize as the style of Pheidias in his Parthenos epoch
(cf. p. 44 seq) Among the metopes of the Parthenon this ' horse-taming ' motive is met
with only in a few of the freest examples ; 3 on the ' Theseion ' it appears only in the
frieze 4 (which is similar in style to the more advanced metopes of the Parthenon), and
not in the metopes at all ; it is represented on the shield of the Parthenos as well as on
the basis, it is quite familiar on Parthenon frieze and pediments, and in all works nearly
related to the style of these sculptures, such as the reliefs of Phigaleia, a certain series
of Attic vases, etc.5 We may, then, assume that Pheidias, if not actually the inventor
of the motive, is principally responsible for its adoption into Attic art.
Formal analysis has taught us that the Colossi of Monte Cavallo, irrespective of
their inscriptions, must be referred to originals from the epoch of the Parthenon frieze
and from the circle of Pheidias. This being so, we have no grounds for doubting the
correctness of the traditional inscriptions, which give us the artists' names : Pheidias
and Praxiteles. This Praxiteles is no longer unknown. He is of course identical with the artist
of the Promachos (p. 32). This statue, we saw, was a work closely connected with
Pheidias, perhaps partly invented by him, and we drew thence the conclusion that
Praxiteles was an intimate colleague of Pheidias. This accords admirably with the
facts gathered from a study of the Colossi of Monte Cavallo and the inscriptions.
The two figures are conceived in the same spirit, and there must have been a strong
bond between the two artists. Such collaboration and division of labour were common
enough in antiquity in the case of larger commissions involving several figures.6 It
was a great mistake to imagine that the two different artists' names inscribed on the
Colossi could not possibly both be correct. Yet if a work of this kind is to have any
general uniformity of character, the root-idea must be the invention of one mind. In
the present case we surely cannot be wrong in ascribing this root -idea to Pheidias,
though in the working out of it and in the execution of the details each of the
artists concerned would become alone responsible.
1 Cf. also north frieze No. 58, and the Hermes on the Echelos relief, 'Eiprif- dpx. 1893, Tliv. 9.
2 In the portion given by the Lenormant statuette. Cf. supra, p. 45.
3 South xiv. xvi., Eastix.- 4 Cf. Overbeck, Plastik, 3rd ed. i. 348, Fig. 5.
5 Those named on pp. 44, 45.
" For instances corroborated by inscriptions see Lowy, /. G. B. p. xv. In addition to these we have the
examples handed down by literature.

IOO PHEIDIAS
In spite of the great similarity between the two figures and the disappearance in
the copies of many an individual and distinctive trait which no doubt existed
in the originals, we can still detect differences in the artists. The work of
Praxiteles— the youth holding the horse with his right hand and turning his head
to the left — shows more stress and fire in the attitude. This difference comes
out most clearly in the heads (Fig. 42, 1, Praxiteles ; 2 and 2 a, Pheidias). The
head by Praxiteles has a more passionate, free, and enthusiastic expression, and on
this account has been generally preferred to the other.1 This is just the distinction
which we found to subsist between the Promachos and the works of Pheidias : in the
Promachos, too, we found a mood of greater excitement expressed by a more restless
and animated exterior. We could however assert this only of the body, for the head
of the Medici torso is lost.
Yet the head of the Promachos, the work of Praxiteles, seems to have been pre
served in one copy. This is a colossal head of the Jacobsen collection in Copenhagen
(Fig. 43). We saw that the Promachos held her head turned to the right, and showed
in her whole attitude defiant warlike excitement. The Copenhagen head corresponds
admirably to such a conception, and I think every one must be struck by the similarity
in style to the heads of the Dioscuri, especially to the one by Praxiteles. The vigor
ous movement, the expression, the forms of the individual features, more particularly
of the open mouth and the chin, are wonderfully like. The angle formed by forehead
and nose is rather more marked in the Athena, thus pointing to a slightly earlier date.
The hair springs out in full masses in front of the ears, and recalls the Lemnia rather
than the Parthenos.2
If we imagine the Medici torso completed by this head 3 or one of its replicas,
and the arms restored with lifted shield and lance held ready for the fray, we have a
majestic and moving image of the war goddess gazing in joy of conflict and assurance
of victory from her citadel away across city and country towards the quarter whence
her worst enemies threaten — whence a few decades later was to come the power which
crushed and overthrew her proud supremacy.
The originals of the Monte Cavallo Colossi can hardly have stood in Athens. They
represented, there is little doubt, the Dioscuri.4 Hence it seems probable that the
Colossi were set up in some city specially devoted to the cultus of the Dioscuri, where
1 Cf. Friederichs, Bausleinc, p. 128 (= Wolters, p. 456), where the names have changed places according to
the modern mistake.
- The copyist has used the borer freely, and apparently not understood his original. The work recalls the
Parthenos copy in Berlin, Ant. Denkm. i. 3. On lips and nostrils and on the frontlet of the helmet are traces of
red colour. The back and the upper part of the head are made of separate pieces. Holes in the ears for
earrings. The teeth are indicated, the lower edge of the frontlet is broken, it originally ran downwards in a
point, as in the Parthenos.
3 The face of the Jacobsen head is 31 cm. long. This would agree approximately with the dimensions
required by the proportions of the Medici torso. There is a curious similarity between the shape and size of the
hollow intended to receive the head on the torso Medici and the portion of bust which ib prepared for insertion
in the Jacobsen head. Both form an ellipse narrowing into a point in front ; on the torso the length is 40 — 41
cm., its breadth 32 cm. ; on the head length and breadth are 43 cm. and 24 cm. respectively. It is certain,
however, that the head does not belong to the torso Medici, but to a replica of it with a similar hollow for
the head. 4 The coin of Maxentius quoted by Donati shows, it is true, the Dioscuri standing quietly and the horses
rearing. But the Dioscuri are represented in the motive of the Monte Cavallo group — on a sarcophagus now
in the Museo delle Terme (in the centre a married couple, on each side a Dioscurus with his horse in the
Monte Cavallo motive, but with the addition of pileus, chlamys, and sword ; workmanship of the third
or fourth century A.D.) The fact that the Colossi do not wear the pointed cap does not tell against the
interpretation, as in the fifth and fourth century B. c. this attribute was not yet given to them. (Cf. Roscher's
Lex. i. 1 172. )

Fig. 43.— Head in the Jacohsen collection at Copenhagen. (By permission of Messrs. Bruckmann, Munich.)

102 PHEIDIAS
a huge votive gift of the kind would be appropriate. Such a city was Tarentum,1
where bronze Colossi were in great favour, and whence the originals could easily
have been removed to Rome. The copies are adapted as decoration for a gate,
but it does not follow that the originals served the same purpose. Nor is the sym
metry of the composition a proof of decorative intention. In ancient art, from the
earliest to the latest times, symmetry was a necessary factor in the typology of the
Divine Twins. The one is always the exact counterpart of the other. Originally
the horses would be placed not at a right but at an obtuse angle to the figures,
according to the more natural arrangement. Since the publication of Wagner's
penetrating remarks on the subject it is generally and rightly assumed that the
originals were of bronze, and if so they were probably not decorative, but intended
to stand free.
The Dioscuri belong to the time of the Parthenon frieze. The Promachos,
which we assigned to the years between 445 and 440, is therefore a somewhat older
work of Praxiteles than the Dioscuri — a conclusion which we reached independently
from a study of the style.
It remains to find out whether anything more is known about this Praxiteles.
He is of course the elder Praxiteles, long acknowledged as a person distinct from
his younger namesake.2 However, since all conjectures concerning this artist
have hitherto been of extreme vagueness, it will be best to sum up what is really
known, rejecting what is false or untenable. First of all we have the notice in
Pausanias (i. 2, 4) about the ypafip-aTa aTTi/cd on the group of Demeter, Kore,
and lakchos, but this only gives us a general date before the time of Eukleides.
More definite information is supplied by the statement that the temple image of
Hera at Plataia was by Praxiteles, for the splendid large temple to which this
dyaXfia p,iya belonged was built in 427 — 426 (Thuc. iii. 68). This would roughly give
us the period from B.C. 445 to 425 for the activity of the elder Praxiteles, who accord
ingly would be a somewhat younger contemporary of Pheidias, and would be still at
work in the time of the Peloponnesian War.3 His works themselves have shown us
that his connexion with Pheidias was a close one ; Pheidias must have held him in
high esteem, and probably obtained for him the commission to make the Promachos.
Very soon afterwards the two artists in collaboration made the group of the Dioscuri.
No doubt the commission was intrusted to Pheidias, and he is responsible for the
original design, but Praxiteles brings his individuality strongly to light in the figure
which bears his name. Although standing in the shadow of one greater than himself,
his own significance and importance cannot be concealed.
The idea we have formed of the work produced by the elder Praxiteles is well
illustrated by an interesting head (Fig. 44), of which there are two replicas,4 representing
1 The cultus of the Dioscuri at Tarentum is proved by numerous coins and terra-cottas ; it was derived
from Laconia.
2 Cf. Klein, Oesterr. Mitth. 1879, 8 ; Brunn, Bayr. Sitzber. 1880, 443 ; Kroker, Gleichnamige Kiinstler, 44 ;
U. Kohler, Ath. Mitth. ix. 78 ; Robert, Arch. Mdrchen, 62, 156.
3 The Artemis on a kylix of about 480 cannot, as Robert, Arch. March, p. 156, assumes, be referred to the
Brauronia. I consider Studniczka's hypothesis about the latter very probable (cf. infra on Praxiteles, p. 323).
The head called 'lakchos' by Winter (Bonner Studien; cf. supra, p. 55) is too severe for our Praxiteles. If
Praxiteles worked with Kalamis (Pliny, xxxiv. 71), it could only have been at the end of the period of Kalamis,
and this does not justify the supposition that the style of Praxiteles was severe.
4 (a) Fig. 44 ; in the Louvre, No. 2547, from the Coll. Campana ; d'Escamps, Marbres Campana, PI. 63. The
whole neck with the term, the lower part of the curls, the nose, the upper lip, and the back of the head are
restored, (b) Vatican, Mus. Chiar. 145 ; Visconti, Mus. Chiar. i. 10; Miiller-Wieseler, Denkm. ii. 119 ; Overbeck,
Apollo, p. 118; Helbig, Museums, No. 72. Poor late work. Brows and pupils plastic. Fillet in the hair (not in a).

THE ELDER PRAXITELES

103

a delicate youth with rich curls falling over his forehead and down over his neck.
The Pheidian manner is evident in the stylistic treatment of hair and face, particularly
in the eyes and mouth. But a slight hint of inward excitement, the mouth open as if
to ask an impatient question, and the intentional asymmetria of the hair over the
forehead, are signs of the artistic tendency we have already traced in Praxiteles.
Moveover, the type of head is remarkably like the Eubouleus of the younger Praxi
teles (Plate XVI.)1 — ¦ so like that there must be a bond of some kind between the two
productions. The essential features are the same, but they are worked out in the
one instance according to Pheidian tradition, in the other after the manner of later
Praxitelean art. It is very instructive to notice how the same idea is expressed in the

Fig. 44. — Head in the Louvre.
forms of two widely different epochs. The hair lies in the same general scheme of
arrangement, but the style of rendering is not the same. Forehead and eyes differ ;
only in one detail the older head— if we may trust the Louvre copy — oversteps the
Pheidian circle and touches on the newer formation — i.e. the lower eyelid is more
definitely set off from the cheek, through the indication of the lower rim of the eye-
socket, a nicety of modelling not customary in the fifth century.
As the ' Eubouleus ' of the great Praxiteles certainly represents an Eleusinian
divinity, the interpretation of the earlier head must be sought for within the same
The head in the Louvre is therefore also called Virgil, like the replicas of the Eubouleus in Rome and
Mantua. Both the Louvre head and a head in the Villa Albani (No. 48, 'Alessandro'), which I also attribute
to the elder Praxiteles, have already been touched upon in connexion with the ' Eubouleus ' by Benndorf in the
Anzeigerd. Wiener Akad. 1887, 16th Nov.

104

PHEIDIAS

mythological circle. It is probably a copy of the lakchos of the elder Praxiteles.
The delicate face framed in curls would answer admirably to the wpalof 0eo?
(Aristoph. Frogs, 395) who stood as torch-
bearer beside Demeter and Kore.1 The ele
vated beauty of this head accounts for its
great popularity in Athens (Cic. in Verr. iv.
60, 135). First among other surviving works be
longing to the same series of Pheidian works
as the Monte Cavallo horsemen may be placed
the fragment of a colossal statue, the ' Jupiter
de Versailles ' in the Louvre (Fig. 45).2 The
head is near akin to the Dioscuri, the strong
undulating hair, the form of the forehead, the
large eyes, and the open mouth being unmis
takable evidences ; even the hair on the upper
part of the head and on the neck behind
shows correspondence. The short curled beard,
on the other hand, recalls the earlier Pheidian
manner as exemplified in the Barberini head
(Fig. 12). The majesty and energy of the
whole conception have no parallel except in
the Dioscuri. Unfortunately only the upper
part of the body survives. A mantle passes
behind from the right hip to the left shoulder,
the right arm was stretched forward. The short hair and the excited expression
seem to me less appropriate to Zeus than to Poseidon. It is possible that this
statue also may have its source in a work of the elder Praxiteles.3 At any rate
it must have proceded from the school of Pheidias. I may mention in conclusion
a female type with waving, fluttering hair and open mouth, known in different
variants, and representing a goddess of animated nature.4

Fig. 45. — 'Jupiter de Versailles ' (Louvre).

XI. Pheidian fnfluences in Sicily and Magna Graecia. — Coins and Vases.
It is not, however, my intention to write an exhaustive treatise on the works of
the Pheidian school ; I only aim at tracing a ground-plan for future investigation. I
believe we have found in the Lemnia on the one hand, and in the Dioscuri on the
other, two fixed points which represent the opposite poles of Pheidian art : here calm
repose, there impetuous motion ; here manifold threads of connexion with old forms,
there the full freedom of the new style.

1 This lakchos was certainly not represented with horns. The Povitipuis *Ia«x« (Sophokles, frag. 871,
Nauck) is not the lacchos of Attic cultus, but only a poetic name for Dionysos : it is expressly stated that he
dwells on Nysa, which is true of the bull-Dionysos, but not of the Attic lacchos. The poetic blending of lakchos
with Dionysos, which occurs from Sophokles downwards (cf. especially ^72^-1115 seq.), does not prove that
the mingling had taken place in Attic cultus and art of the fifth century.
2 Bouillon, i. 1 ; Miiller- Wieseler, Denkm. ii. 4; Overbeck, Kunstmyth. Atlas, Taf. 2, 15, 16 ; Zeus, p. 83,
No. 14; Frohner, Not. No. 31.— Our illustration omits the restorations.
3 There was a Neptune by Praxiteles, possibly the elder, in Pollionis Asinii Monumentis, Plin. xxxvi. 23.
4 Head on a statue in St. Petersburg (Compte Rendu, 1881, PI. VI. 1, 2). Similar, but not identical, is
the head of the Vatican Artemis (Helbig, Musetims, 37).

COINS OF THURII 105
But in order to make our notion of these two consecutive developments of
Pheidian art still more complete and definite, we must see how they were reflected
and reproduced in coins. And first we must emancipate ourselves from the current
dating adopted for the coins in question. When coins can only be dated from their
style, this should only be done by reference to the fixed points gained by a study of
the aggregate of other monuments, and especially of sculpture. It is true that in this
way only the date when any given type arose can be discovered, and not the period
of its duration, which, as Athenian coins show, is sometimes disproportionately long.
From the point of view of art, however, it is just the period of the rise of the coin-
type that is the most significant.
In B.C. 445 ten ships sailed westwards from Athens to found Thurii, which in
B.C. 443 was strengthened by immigrations from all parts of Greece. The new colony
naturally had the same patron goddess as Athens, and on its coins was stamped a
head (PI. VI. 1, 3) whose typical features — form of helmet, wreath of olive, arrange
ment of hair — were taken straight over from the Athenian coins of the period.1 The
severe style however, which had been retained in Athens from external reasons, was
abandoned on the coins of Thurii ; they follow the style of the time. Only in the
rendering of the hair on the forehead with regularly waved contour and parallel inner
lines is there an echo of the older prototype. The establishment of this type with
the wavy front hair may certainly be assigned to the earliest period of the colony.2
From the existence of numerous dies with slight variants it seems clear that the type
held its own for a considerable time, but its place in the history of art is fixed by the
fact that it is an Attic creation of the period when Pheidias was working at the
Parthenos. Its close relation to the art of Pheidias is quite evident. This earlier
type of Thurii most definitely recalls the Lemnia. The undulating front hair of the
Lemnia (cf. p. 19), the line of her profile, with the low forehead and the nose at a
slight angle to it, the restful beauty of the full closed mouth, and the finely wrought
transitions in modelling — all this has no such close parallel on coins as the earlier type
of Thurii. Somewhat later there appears in the same city a type corresponding to
the later style of Pheidias and his circle ; the hair above the forehead flows back
over the edge of the helmet nearly as it does in the Parthenos, and the expression is
more animated and restless than in the other coins (Plate VI. 6).3
Among the dies of the older type those signed <J> are specially fine. Among these
are found both staters (the helmet with simple olive-wreath)4 and distaters with Skylla
on the helmet (PL VI. 5).5 These dies are most probably the work of the artist who
signs with a <E> in Terina, and whose personality makes itself so distinctly felt on its
1 Cf. Head, Guide, PI. 13, 21, 23, 30, and the imitations of this type in Lykia, ibid. PI. 11, 38, 20, 39.
On the Phokaian electrum hektai the type is very similar (Berlin Coll.), but without olive-wreath.
2 This is most clearly proved by the small coins of New Sybaris, which was founded immediately after
Thurii ; the Athena head is exactly the same as on the older Thurii type (Garrucci, Le Moncte, Tav. 108, 23, 25 ;
Gardner, Types, PI. I. 31, 16, 18 = our PI. VI. 2). Further imitations of that Thurii type, which are to be
traced in Kyme, must be dated before 420 B.C., in which year Kyme was destroyed. Head (Hist. Num. p. 71)
follows the majority of numismaticians in dating the coins of Thurii after 420 B.C. ; and the type with Skylla on
the helmet he places after 390, although, as the style of face and hair proves, it must have arisen very soon after
the type with the olive-wreath. Gardner ( Types, p. 103) estimates the dates more correctly. [A beautiful early
coin of Thurii (Brit. Mus. Cat. Thur. i.), belonging to the series referred to by Prof. Furtwangler, but still
earlier than any of the examples he cites, has been pointed out to me by Mr; G. F. Hill. It is now published
for the first time, PI. VI. 1.— E. S.]
3 E.g. Gardner, Types, PI. V. 18 ; Garrucci, Le Mouete, Tav. 106, 9, 10, 14 ; Friedlander u. Sallet, D. Kgl.
Miinzcabinet, Nos. 739 (pub. ibid, on PI. VIII. but marked by mistake 736), 743.
4 Head, Guide, PI. 15, 7.
5 Friedlander u. Sallet, Das Munzcab. No. 741 ; the head as in Head, Guide, 25, 17. P

106 PHEIDIAS
coinage.1 His style, in the design of the seated Nike of the reverse (PI. VI. 7), for
instance, bears so surprising and unmistakable a resemblance to that of the Parthenon
frieze 2 that he must have stood in the closest relations to the artist of the frieze— in
other words, to Pheidias. His activity in Terina is evidently somewhat later than his
Thurii period, for the Thurii work recalls the style of the Lemnia, while the head of
the nymph on the coins of Terina is, like the reverse, in the manner of the Parthenon
frieze. These coins of Terina, following, as they do, closely on coins of the period
in which the trammels of archaic convention are still felt, must from evidence of style
be dated in the same time as the Parthenon frieze—/.*, between 440 and 430 B.C.
It is highly probable that the same artist's hand is to be traced in the reverse of
a coin of Pandosia,3 also signed 3>. The young huntsman seated is quite in the
manner of the Parthenon frieze (PI. VI. 10). The obverse {ibid.) is even more
striking, for it bears a head of Hera facing, with wavy hair fluttering out to the
sides — another instance of the more recent Pheidian tendency, which we saw brought
to expression in certain figures of the Parthenon frieze (PI. V.) and in the Dioscuri
of Monte Cavallo. The whole scale of gradation from the Lemnia to the Dioscuri,
which we traced in the large monuments, is exactly reproduced, step by step — and
this is a confirmation of the results attained so far — in the works of a die-cutter.
The remarkable fact that the Pheidian style in its different phases was carried
straight from Athens to Lower Italy comes to light nowhere more clearly than on the
coins just discussed. But the influence which transplanted art exercised on native
work, principally through the agency of the colonists of Thurii, is to be traced in a
number of other circumstances.
The types of the Athena heads of Thurii, especially the one with the regularly
waved hair, were rapidly and widely adopted in Lower Italy. More particularly in
the Greek towns of the Campanian coast, Kyme,4 and Neapolis (PL VI. 4),5 we
find exact copies of the oldest type of the staters of Thurii. About the time when
Thurii was founded, Athenians seem to have settled in Neapolis : 6 at all events it
became a second centre from which Attic art spread to other places. The neigh
bouring Italians — not only the Campanians, but the inhabitants of Hyria, Nola, and
Phistelia 7 — stamped on their coins the head of the Attic Athena. Other Greek
1 Compare Stuart Poole, Num. Chr. 1883, 269 sqq. PI. II, 12. Poole was the first to emphasize the purely
Attic style of these coins. In Terina the artist * distinguishes himself very clearly from his much less important
contemporary and imitator, the artist who signs P. — Poole's attempt (p. 273) to identify the artist # with the
artist who signs #<, $iAu, $iA.io-ti, QiXurriuv has been refuted by Evans, Tarentine Horsemen, p. 1 10, note. The
Terina coin with iiXis is quite different in style from those with $, and considerably later : it is by the same
Philistion who worked for Velia, Tarentum, probably also for Metapontum and Herakleia, at the end of the fourth
century (cf. Evans, loc. cit. 106 sqq.) Gardner's hypothesis (Types, p. 121), that the artist $ might be Phrygillos, who
made the bull for the reverse of the coin of Thurii, cannot be entertained, as this artist's style is quite different.
2 The correspondence between Num. Chron. 1883, PI. 11, 4, and the Athena of the Parthenon frieze is
specially striking.
3 Cf. Poole, loc. cit. p. 276, PI. 11, 14; Head, Guide, PI. 25, 22.— In the series of coins of Pandosia the type
follows immediately on a head of severe style ; Head, Guide, PI. 15, II ; Hist. Num. p. 92, fig. 60. The coin
of Velia with *, which Poole, PI. 11, 13, ascribes to the same artist, must, I think, be later.
4 Poor example, Brit. Mus. Cat. Italy, p. 86, No. 6 ; a better one Berlin, exact imitation of the oldest
staters of Thurii. Garrucci, Le Monete, Tav. 83, 27 — 29. Cf. supra, p. 105, note 2.
5 Head, Guide, PI. 15, 3; Garrucci, Tav. 84, 19, 85, 14. 6 Cf. Busolt, Gr. Gesch. ii. 591.
7 Imhoof-Blumer ( Wiener Numism. Ztschr. 1886, p. 226) has shown that these coins of Campania were
struck in Neapolis : the Athena head is identical on both. Among the coins of Hyria (cf. Imhoof-Blumer, loc.
cit. p. 206 sqq. ) there occur exact imitations and freer modifications of the older Thurii type. The same Thurii
die appears in Hyria and in Nola. As the nymph's head of free style on the coinage of Neapolis and Nola
appears to be from the same die, it is probable that the Athena heads of Hyria and Nola were also stamped in
Neapolis.— Small silver pieces of Phistelia with the older Thurii type in Berlin (not mentioned by Head, Hist.
Num. )— Allifa : Garrucci, Tav. 88, 20.

VI.
Greek Coins. i
1,3,5,6. Thurii.— 2. Sybaris.— 4. Neapolis.— 7. Terina.— 8, 10. Pandosia.— 9. Hyria.— 11. Ne polk— 12-15.
Syracuse.— 16, 17. Naxos.— 18. Herakleia.— 19. Phokaia.— 20. Lokri Epizephyrii.— 21. Ronian denarius.—
22, 23. Elis.— 24, 25. Amphipolis.— 26. Miletos.— 27-29. Klazomenai.— 30, 31. Athens.— 32!, 33. Argos.—
34, 36. Alexandria Troas. — 35. Heraia. — 37. Coin of Antoninus Pius. — 38. Corinth.

COINS OF SYRACUSE 107
cities, like Poseidonia 1 and Velia,2 imitated the Thurii type ; even proud Syracuse used
it for small gold coins,3 in the period about 430 B.C. Herakleia, too, which was founded
in 403 — 402, stamped coins first with the earlier, then with the later, types of Thurii,4
and the bareheaded Athena peculiar to Herakleia (PL VI. 18) is, as we saw above,
a thoroughly Attic conception (p. 15), its nearest parallel in style being the older type
of Thurii. Tarentum borrowed from her daughter-city for her small silver coin, the
diobol, not only this bareheaded Athena (cf. p. 15), but notably also the helmeted
goddess of the Thurii type.5
Next to the Athena type of Thurii, which recalls the Lemnia, that other and
later Pheidian type, showing a head facing, with fluttering hair, exercised a powerful
influence in Lower Italy. We found it in Pandosia adopted for Hera ; it appears in
Kroton and Poseidonia,6 and again in Campania — i.e. in Hyria-Vesernia (PL VI. g),7
in a fairly exact copy of the Attic original belonging to Pandosia. Essentially the same
type, except that the high headdress is replaced by a broad fillet, is seen on a series
of beautiful coins from Neapolis (Plate VI. 11) and Phistelia.8 This is the same
head which we know in sculpture from the two variants in St. Petersburg and
the Vatican (p. 104). On coins its noblest and most beautiful form is due to
Kimon, the die-cutter of Syracuse, who made the famous head of Arethusa facing
(Plate VI. 12). Arthur Evans has lately shown in a convincing manner that Kimon
originally entered the circle of Syracusan die-cutters as a foreigner, and introduced into
Syracusan coin-types certain elements belonging to Neapolis, such as the form of the
earrings, and above all the type of head with the fluttering hair.9 Now this type,
like the Attic Athena, had come to Neapolis from Pheidias and his circle. Another
design, to which we may ascribe Attic-Pheidian origin, is the beautiful Athena head
with curly hair, wearing a triple helmet, which appears on the Syracusan tetradrachms
by Eukleidas (Plate VI. 13),10 and which seems, like the head on the St. Petersburg
gold medallion,11 to have been suggested by the Parthenos.
If we now pass from these minute details to the whole development of
style in the coinage of Lower Italy and Sicily, we shall see that a sudden
and almost universal revolution took place about 440 B.C. At the same time the
coin-engravers begin to sign their works. Eumenes in Syracuse is the first to do so
(Plate VI. is).12 The magnificent but hard features of the transitional Arethusa head
1 Garrucci, Tav. I2r, 36—38.
2 The older Thurii type is rare (Garrucci, Tav. 119, 17) ; the later is more common (ibid. 119, 11).
3 Head, Num. Chron. N. S. XIV. PI. 3, 10. A better copy in Berlin; Head's date, 412, is certainly too
late ; the inscription also is older.
4 Exact imitation of the old type on a. coin in Berlin stamped over an incuse of Metapontum or Kroton ;
hippocamp on the helmet. Cf. Garrucci, Tav. 101, 19, 20, 36. Those with the griffin are late poor imitations of
the type (ibid. 101, 31) ; those with Skylla (ibid. 101, 32) follow the later Thurii type.
5 The earlier Thurii type.with a hippocamp on the helmet, as in Herakleia, Garrucci, Tav. 100, 5 ; the latei in
many variants, ib. Tav. 99, 51, 52 ; 100,1 — 4. The reverse, Herakles with the lion, also corresponds with Herakleia.
6 Kroton : Garrucci, Tav. 109, 37, 38. A later modification ib. 36, 39. Head, Guide, PI. 25, 20. Posei
donia : Head, Hist. Ntim. p. 68 ; Garrucci, Tav. 121, 4.
7 Imhoof-Blumer in Wiener Numism. Ztschr. 1886, Taf. v. 6 — 13.
8 Fine illustrations in A. J. Evans, Syracusan Medallions, PI. III. I — 3.
9 A. J. Evans, Syracusan Medallions, p. 77 sqq. ; PI. III. (Num. Chron. 1891), where the imitations of
Kimon's head of Arethusa facing are discussed ; it was copied with special exactness at Larisa in Thessaly.
10 Weil, Kiinstlerinschr. Taf. 3, 7 ; Head, Guide, PI. 26, 31. Cf. Evans, loc. cit. p. 70 seq.
11 Ath. Mitth. 1883, Taf. 15.
12 For the date of this artist see the remarkable researches of A. J. Evans, Syracusan Medallions, London
1892. This book makes it unnecessary for me to controvert the usual date given to the Syracusan coins, and
which I have always held to be false. Evans has proved finally that the tetradrachms all belong to the fifth
century, and that even the dekadrachm types of Euainetos and Kimon were created towards the close of this
century (after 413 B.C.)

108 PHEIDIAS
(Plate VI. 14) now give place to softer and fuller forms ; the hair is disposed in thick
undulating locks ; the sharp lines are given up, and the whole treatment becomes
more pictorial. Just as the older style was presumably developed under the influ
ence of Pythagoras, so this later style may be referred to the Attic influence of
Pheidias. So too in Naxos, the superb Dionysos head of the severe style, probably
suggested by some famous plastic work of Pythagoras (Plate VI. 16),1 is imme
diately succeeded by the dignified head with diadem and short hair, which looks as if
it had been cut out of the Parthenon frieze (Plate VI. 17).2
The great streams of artistic tendency due to the separate influence of men of
genius and their schools may everywhere be traced on coins. The theory which
refuses to see in the smaller art of the die-cutter anything but local style and auto
chthonous development is entirely mistaken.3
In the time of Perikles, Athens, as is well known, was in constant com
munication with Lower Italy and Sicily. Traces of peace negotiations with Egesta
about 454 — 453 have been accidentally preserved on the fragment of a record.4 Coin-
types prove that Attic art of the Periklean-Pheidian epoch spread from Thurii and
Neapolis to other places. This stream was too strong to be checked either by the
political vicissitudes or even by the final defeat of the Athenians before Syracuse.
The splendid dekadrachms of Kimon and Euanietos, struck immediately after the
victory over the' Athenians, are a product of that Attic style5 which was brought to
Syracuse chiefly by Kimon's agency. In the fourth century, in Syracuse as in the
whole of Magna Graecia, artists drew at first only from the great storehouse of types
provided by Attic-Pheidian art, and new conceptions grew but slowly.
There is still a second class of monuments which shows clearly how strong was
the influx of Attic art into Magna Graecia in the Pheidian period, and how powerful
was the stimulus it gave to fresh production. I refer to the vases. The style and
composition which we have proved to belong to the reliefs on the shield of the Par
thenos, and to correspond with one series of Attic vases, reappear all at once in a class
of vases made in Southern Italy. These vases must be considered as the first products
of new potteries, since there is no early red-figured painting proper to this district.
They are not the superficial half-misunderstood imitations of imported vases which the
non-Hellenic craftsmen of South Etruria and Campania were in the habit of making
about the same time or earlier ; they are a purely Greek product. Their relation
first to the Attic vases of about 440 B.C. and then to those of the following decades
is a very close one. At first they correspond exactly to the Attic vases in form and
decoration 6 as well as in the composition and general style of the paintings. And
yet there is an individual touch in the style and typology of the figures from the veiy
beginning ; "' although they cannot compete with their Attic prototypes in beauty and
1 Cf. the analogous type of the bronze head from Herculaneum (the so-called ' Plato '), which on account
of its similarity to the Dionysos of the Naxian coins I am inclined to refer to Pythagoras (cf. 50th Berl.
Winckelmannsprogr. p. 130, note 18).
2 Good illustrations of both types in Head, Guide, PL 17, 29, 30 ; Gardner, Types, ii. 22, vi. 14.
3 In this respect P. Gardner estimates more correctly ; he feels the necessity of connecting the coins with great
artists ; he is however mistaken in ascribing (Types, p. 118) to Polykleitos a greater .influence on coins than to
Pheidias. i Cf. Lolling in the AsXriov, 1891, p. 105. 5 Cf. Arthur Evans, loc. cit.
6 The technical marks are very different. The clay is much paler and duller than that of Athens, and is
usually covered with a thin coat of red colour intended to imitate the colour of the Athenian vases, but which is
rarely well preserved. Superimposed colours, red and white, are only sparingly used, and many details, which in
Attic vases are wont to be painted, are constantly merely blocked out.
7 E-g- the type of the bearded Seilenos differs from the Athenian type.

VASES FROM MAGNA GRAECIA IOO,
refinement of drawing and technique, they have a rough incisive force all their
own. To make clear which are the vases I mean, I will cite a few of the principal types.
We possess a series of bell-shaped kraters which only the most exact and careful
observation of technical and stylistic points can distinguish from the corresponding
Attic kraters of the period between 440 and 430.1 An excellent example, though not
one of the earliest, is the krater with the Expiation of Orestes in the Louvre ; 2 it may
be dated about 420 B.C. Among the hydriae, the one with the frieze of Argonauts in
the Cabinet des Medailles 3 should be named first. It is exactly akin to the Attic
hydriae of about 440 B.C. The Berlin hydria (No. 3164) with Zeus and lo is some
what later, and may be grouped with Attic hydriae cf 430 B.C. and after, such as the
one with Paris and Kadmos (Berlin, 2633, 2634). The hydria with the death of
Kanake, a subject derived from the Aiolos of Euripides,4 also belongs to the
fifth century. Further, there are a few amphoras with volute-handles to be named in
this connexion ; more especially one in the Jatta Coll. in Ruvo,6 which displays on one
side an Amazonomachia closely analogous in composition and motives to the shield
relief of the Parthenos, and on the other the Rape of the Leukippidai conceived just
as on Attic vases of about 440 derived from Polygnotan prototypes. Equally
Polygnotan in design is the episode of Phineus and the Argonaut on another volute-
amphora of the same collection.0 A specially conspicuous instance from the period
about or soon after 430 is a krater in the Cabinet des Medailles : on one side is
Odysseus conjuring the shade of Teiresias, on the other the Judgment of Paris.7 To
the same class belong a krater of somewhat more recent style in the collection of
Sir Francis Cook, in Richmond, representing the Blinding of Polyphemos 8 (from the
Kyklops of Euripides), and the charming krater with the humorous and forcible
rendering of the Pursuit of Dolon.9 The very shape of the 'Vasi a colonnette ' was
taken over by Lower Italy straight from Athens, in the middle of the fifth century B C.
Vases of the kinds just mentioned are found in Lucania {e.g. Pomarico, Anzi, etc.)
and also in Apulia (especially Ruvo). Technically and stylistically they form a pre
paratory stage to the later Lucanian vases, which rapidly assume a strongly local
stamp and are hardly ever found in Apulia, which has a special vase-manufacture of
its own {i.e. the Tarentine). It cannot be doubted that these earlier Lower Italy
vases belong to the fifth century, for their existence can only be understood on the
supposition that they are nearly contemporary with the parallel productions of
Attic art. The real ' Apulian,' i.e. Tarentine, class develops rather later, but it still
belongs to the fifth century ; its nearest analogies are the Attic vases of the period
1 Many in Naples and in the Museo Jatta at Ruvo ; good examples in Berlin, Nos. 3179, 3180, 3182.
2 Mon. d. Inst. iv. 48.
3 Gerhard, Auserl. Vas. 153, 154.
4 Now at Bari. Shoulder decoration pub. Arch. Ztg, 1883, PI. 7, I (p. 51, Kalkmann). Another hydria
of the series, Millingen, Vases de Div. Coll. PI. 4, 5.
5 Jatta, 1096; Bull. Nap. n. s. ii. (1854), Tav. 4; Mon. d. Inst. xii. 16. — Another volute-amphora with a
Pheidian Amazonomachia is in Geneva (Giraudon, Vases Ant. des Coll. de la Ville de Geneve, PI. 31, 32); even
the details of the drapery exactly imitate Attic vases of 430 B.C.
6 Mon. d. Inst. iii. 49. Ornamentation precisely like the Attic. The line indicating the ground is incised
A third volute-amphora with Herakles and Kyknos in Jatta (Bull. Nap. n. ». i. 6 ; cf. Roscher's Lexikon,
i. 2231) stands very near Attic prototypes of 430 B.C.
7 Mon. d. Inst. iv. 1 8, 19.
8 Jahrb. d. Inst. 1891, Taf. 6 ; the original (which I examined in 1888) shows burning fire on the logs 01
wood ; the youth clearly wears a pilos. Winter's suggestion that the vase is Attic is untenable, alike from
technique and style; with the head of Polyphemos cf. for example the Phineus, Mon. iii. 49.
9 Bull. Nap. i. (1843), Tav. 7 ; Overbeck, Gall. Taf. 17, 4.

110 PHEIDIAS
after 430. A conspicuous instance occurs on the volute-amphora (St. Petersburg, No.
523), where the Gigantomachia1 is treated as on the shield of the Parthenos, and the
drawing of the wavy hair, the ornaments on the drapery, etc., are plain echoes of
those Attic prototypes. In this connexion it is instructive to study the fragments of an
unusually large and splendid Attic krater (unpublished) from Kertsch in St. Petersburg,2
which show in a striking manner how closely Tarentine work clings to Attic tradition
of the end of the fifth century. Both groups of vases, the early Lower Italian and the
Tarentine (Apulian), are merely offshoots of Attic art transplanted into new soil. The
subjects of the earlier group, such as the various Argonaut pictures and the Rape of
the Leukippidai,3 are at first derived from the circle of Polygnotan painting, but
subjects from the Attic drama soon make their appearance and afford the strongest
confirmation of the essentially Attic character of these vases ; this dramatic material
is brought straight over from Athens fresh and undiluted. In the Tarentine group,
which was still developing in the last decades of the fifth century and continued to
flourish through the first half of the fourth, the tragedies of Euripides, as is well
known, hold the chief place. They dominate vase-painting, not as ' classical ' literature,
but as the new, sensational, and fashionable poetry of the day. Even the Sicilians,
we know from reliable tradition,4 were enthusiastic admirers of Euripides as early as
412 B.C., and eagerly learned by heart fragments of his dramas brought to them by
Athenian deserters.
The facts touched upon so far seem to me to be capable of the following
explanation. With the founding of Thurii, Attic ceramic art was transplanted
into the district. Its exotic development began about 440 B.C. The products of
the new workshops were distributed not only in the inland district of Thurii and
the Siritis, but also in a north-easterly direction towards Apulia. Then came
the contest with Tarentum, and the treaty which found outward expression in the
establishment of a common colony, Herakleia. I think it probable that the manu
facture of vases was continued in the latter city and spread thence to Tarentum,
where it would flourish greatly owing to the wealth and power of the city, and thus
gradually come to supplant the import trade from Herakleia, at least so far as Apulia
was concerned. The Herakleian potteries thus lost their importance, their wares
(the later Lucanian vases) became poorer, and assumed a more and more exclusively
local character.5
The export trade of Athens itself was naturally materially injured by the
existence of this new centre of vase-painting in Southern Italy. After 400 B.C.
no Attic vases of later style seem to have come to Apulia, although the export to
Campania continued to flourish for some time. Thus the attempt of Athens to
found an independent colony, Thurii, was prejudicial, in this particular as in others,
to the Mother City. Athens gave of her strength, and received nothing in return.
1 Bull. Nap. ii. (1844), Tav. 6.
2 On one side, above, a great assembly of the gods with Zeus enthroned, below the Dioskouroi riding (?)
and Herakles as victor : presumably the picture represented the institution of the Olympic games and the first
victors. White and yellow are laid on freely, many of the heads are drawn facing, all this as in the Tarentine
vases. 3 Since the appearance of KSpp's admirable essay (Arch. Anz. 1892, p. 124 sqq.) the paintings on vases of
Tarentine manufacture relating to the Persian war may be reckoned among those representations inspired by the
Polygnotan circle.
4 Plut. Nikias, 29.
6 During this whole epoch Sicily had no potteries of her own. Her market was supplied, after the fall of
the Attic power, by different potteries of Lower Italy. This is the result of my observations in Sicilian
collections.

VASES FROM MAGNA GRAECIA III
Nevertheless, for a right appreciation of the power and significance of Attic
art as it was when Pheidias had impressed upon it the mark of his genius, the
history of its triumphal entrance into Magna Graecia will always be of the highest
importance. We have only touched upon some of the chief features of this
history ; much remains unsaid. I will close with one more example. The famous
Ficoroni Cista in Rome by a master of Praeneste is the genuine offspring of
Polygnoto-Pheidian art. Its types are derived from the same source as those of
the Argonaut pictures on early Lower Italy vases — viz. from the paintings of the
Polygnotan circle. This is evident from the separate motives and their details,
although there are many concessions to Latin taste. The Cista must be of nearly
the same date as the vases.1
We have travelled far from our original starting-point, the Lemnia of Pheidias.
The great creations of the brilliant Attic period are like suns, each the centre
of a multitude of smaller stars on which they pour light and life. The art of Athens
in the fifth century was as far-reaching and widespread as her empire ; and the
style which, as we have tried to show, was individual to Pheidias accomplished
the vast success and gave the strong stimulus which we have studied in some of
their manifestations.
1 The inscription, as is well known, only gives a terminus ante quern. The ornamentation and the picture
on the lid point to an older period ; no less the feet of the cista, which, like the relief of Theseus, Peirithoos, and
Herakles (Helbig, Museums, 826), are derived from a relief of Pheidian style. It is as certain that the cista, the
feet, and the handle were all made at the same time, as it is that the whole was made in Rome.

KRESILAS AND MYRON

KRESILAS AND MYRON

I. Literary and Epigraphical Evidence for the Life of Kresilas.
SIDE by side with Pheidias and his immediate
circle there were other artists who followed
an independent line. Among these Kresilas
has long been recognized as one of the most important.
Pliny, in his history of bronze statuary, mentions him
no les.1 than three times : first, in the larger alphabetical
list of the insignes (xxxiv. 74), where he names him as
the artist of a statue representing a wounded man
' in which might be seen how much of life remained,'
and of a portrait of Perikles, worthy of the epithet
' Olympian ' — adding that it is marvellous how the
art of portraiture can make a celebrated man still
more celebrated. The second mention occurs as an
appendix to the list of artists given under C. It states
that Cresilas also made a 'doryphoros' and a 'wounded
amazon ' (xxxiv. 76). The MS. reading of the name is,
it is true, corrupt, but there is little doubt that Brunn's
reading, Cresilas,1 is correct. Finally, Pliny names him
among the laudatissimi who made statues of Amazons for a competition ; Kresilas
being placed third in the order of merit, after Polykleitos and Pheidias (xxxiv. 53).
Neither Pausanias nor any other writer mentions the artist by name, if we except
a corrupt passage in the Auctor ad Herennium where Brunn2 restores with strong
probability .... caput Myronium, brachia Praxitelia, pectus Polyclitium, ventrem et
crura Cresilaea. We can however scarcely conclude from these words that Kresilas
really excelled in the rendering of the abdomen and the legs, for the rhetor is
obviously connecting at random particular parts of the body with the names of
celebrated artists ; 8 so that, even were Brunn's reading absolutely certain, we could only
gather from it that the rhetors reckoned Kresilas among the most famous artists.
Fortunately, several inscriptions from works of the artist himself have been pre
served, no less than three of which are from the Akropolis. The first of these belongs
to the portrait of Perikles ; 4 from the character of the epigraphy, Lolling has dated it
circa 440— 430. The second inscription, belonging to the base of a votive statue put up
by a certain Hermolykos,6 can be dated as far back as the middle of the fifth century.
The third 6 seems to be rather later ; it is metrical, and found its way into the
Anthology. In it the artist mentions Kydonia in Krete as his birthplace. The work

1 Gesch. d. Kiinstler, i. p. 261. " Sitzungsber.
3 See Kekule, Kopf des Praxit. Hermes, p. 23 seq.
5 Ldwy, /. G. B. No. 46.

Bayr. Akad., Philos. -philol. Classe, 1880, p.
4 AtXrlov apxt"oA- J889, p. 36.
6 Ldwy No. 47.

116 KRESILAS
was the gift of a certain Pyres ; and the wording of the inscription, Tov&e Hvprj?
dveOrj/ce Ho\vp,v>]crTov cf>i\o<; vtb<;, ev^dfievo<; Be/caTrjv UdXXaBt, rpiToyevel, shows that
it was probably the portrait of the donor.1 All these inscriptions are in pure Attic
characters, thus proving that the artist must have resided at least some time in Athens.
A fourth inscription shows him, on the other hand, at work in the Peloponnesos,
in the province of Argolis ; it comes from the basis of an offering to Demeter
Chthonia at Hermione,2 dedicated by a certain Alexias. In it Kresilas again calls
himself a Kydoniate, but this time he uses the characters of the place where he is at
work, and in so doing is moreover guilty of a blunder.3 An inscription from what is
evidently the companion work to this gift of Alexias,4 an offering on the part of his
son, gives the name of an artist from Argos. Kresilas is thus shown engaged on
work for a family of Hermione in company with an Argive artist ; and the probability
is that he was living at Argos at the time, and received the commission conjointly with
the native artist. As these inscriptions have nothing archaic about them, it seems
probable that the artist came to Argos after — and not before — his sojourn in Athens.
An inscribed basis of blackish limestone which has lately been discovered at Delphi
during the French excavations 5 must be of still later date, as on it the name of the
artist and his nationality appear inscribed in pure Ionic characters.
Even from these few dates we can gather a good deal of information concerning
the artist's life. Born at Kydonia in Krete, he must have left home early, in order to
cultivate his talents — for we know of no school of artists in Krete at that time. To
judge from coins, the dominating note of Kretan art in the fifth and fourth centuries
was purely barbaric. Yet side by side with an astonishing lack of artistic skill we can
detect here and there proofs of a special aptitude for pictorial and realistic representa
tion. The celebrated coins of Gortyna with Europa or with Zeus Velchanos seated
in a tree, the pieces representing oxen, much foreshortened, licking themselves, the
coins of Phaistos with the seated figure of Herakles resting — all seek to reproduce
nature in a bold and direct manner.6 These coins are later, it is true, than the youth
of Kresilas, but they show in what lay the special artistic quality of the islanders, and
we shall not be surprised if we meet in him with the like characteristics.
It was, of course, only natural that the young artist should turn his steps to the
brilliant and artistic Athens. Here he seems to have succeeded in working his way
up among the first artists, and in obtaining a great reputation, especially as a
portraitist. Otherwise he would certainly not have been intrusted with the bust
of Perikles, the most distinguished and powerful man in the city. The commission
was, it is true, only a private one, as we learn from the inscription ; but this makes no
difference, for Perikles and his friends could certainly command the best artistic skill
of the day just as well as the state. And supposing the portrait to have been,
as has been suggested, a gift on the part of the artist (see infra), this would at
once prove him to have stood in close personal relations to Perikles. As to the
migration of Kresilas to Argos, it was doubtless occasioned by the outbreak of the
Peloponnesian War, which must have rendered residence in Athens unpleasant for an
1 Cf. Ath. Mitth. 1880, p. 28, note 1, and more especially the inscription of a votive portrait-statue from
about the close of the fifth century, 'EipmepU dpX. 1891, p. 56, Plate 6 : AvcriKX^ns . . . 'avidr,™ to,5<f
9e? 75e • • • 2 Ldwy, No. 45.
Writing in the second line X instead of T. See Kirchhoff, Studien, 4th ed. p. 160 seq.
4 Unfortunately only known from a transcript taken by Fourmont, Ldwy, No. 51. Cf. Kirchhoff,
loc. cit. 5 Shortly to be published by the French.
6 Cf. St. Poole, Num. Chron. 1864, 240 ; P. Gardner, Types, p. 161.

PORTRAIT OF PERIKLES 1 17
artist, especially if he were a foreigner. The cessation of artistic activity, the uncertain
turn of the war, the breaking out of the plague, and above all the death of Perikles, whom
we may look upon as his patron, would be sufficient reasons for leaving Athens.
Argos on the other hand had, at this time, undoubted attractions for an artist : being a
neutral state, it enjoyed absolute peace, and its famous school was gradually becoming
the leading artistic centre in Greece, now that numerous calamities were forcing
Athens into the background.
A study of the marble copies executed in Roman times after the works of
Kresilas will help materially to fill up this outline of the artist's career.

II. The Portrait of Perikles.
The existence of copies of the portrait of Perikles by Kresilas may be taken as
proved.1 Two terminal busts, one in the British Museum and one in the Vatican, bear
the ancient inscriptions of the name of Perikles ; they go back to the same original, and
from the style this original must have belonged to the time of Perikles. Their
identification as copies after the celebrated work by Kresilas has lately been further
confirmed by the discovery on the Akropolis of the inscribed basis that supported
the original. It is only a fragment, but it bears the inscription, restored indubitably
correctly, by Lolling : — Ti.ep\ikXeo<i
KjOe<r]tA,a? eirole
If, as is probable, the inscription occupied the middle of the block, the basis
cannot have been more than some 40 cm. broad, a size to which the smallness of the
lettering2 would be well adapted. Now so narrow a basis could not carry a statue,
but only a term, so that it seems proved that the terms with the inscription
' Perikles ' 3 are reproductions of the actual work by Kresilas. The original
inscription shows that the work was only a private offering ; as the donor is not
mentioned, Lolling supposes it to have been put up by the artist himself. Pausanias
also mentions this portrait of Perikles on the Akropolis, without however naming
the artist.4 Of the two replicas, the one in London undoubtedly gives the best idea of the
original (Plate VII. ; profile in Fig. 46).5 The head has never been separated from the
bust and the neck is unbroken, so that the exact carriage and turn are preserved. In
the Vatican replica,6 the head had been broken off and a piece of the neck is restored ;
the neck has become shortened in the operation, and the characteristic pose of the
head — and with it one of the chief charms of the work — is thus lost. Both these
copies are well and carefully executed ; their substantial agreement testifies to their
1 First asserted by Bergk and Brunn (K. G. i. 262).
2 Lolling in the AeXrlov apxaioX. 1889, p. 36 seq. I have examined the basis myself. Its original breadth
was probably 0-44. The depth is now only 8 — 9 cm. In the upper surface on the right a mortise hole is
distinctly visible (also in Lolling's illustration), but it need not be original. The fragment affords no decisive
evidence for or against the assumption of a terminal bust.
3 The breadth of the shaft of the London term is 28 cm., which would suit o. basis of about 40 cm.
very well.
4 Paus. i. 28, 2 ; 25, 1 ; dvipi&s need not necessarily mean a whole statue. For the site cf. supra, p.
10 seq. It has been remarked (p. 62, note 6) that the Xanthippos and the Anakreon stood elsewhere, and had
nothing to do with the Perikles portrait or with Kresilas.
5 The older publications are cited in Friederichs- Wolters, Gipsabg. No. 481.
6 Brunn-Bruckmann, Denkmaler ; Helbig, Museums, 281.

Il8 KRESILAS
fidelity. The Vatican replica, however, reproduces the separate curls of the beard
with greater precision and severity, and shows the hair on the top of the head
immediately below the eyeholes of the helmet, a detail which must have been in the
original, since it emphasizes the high shape of head peculiar to Perikles. There is a
third very mediocre replica (without inscription) in the Barracco collection at Rome.1
This portrait of Perikles is a vivid commentary on what Thucydides wrote con
cerning the temper and manner of the great statesman. Artist and historian alike
show us a man raised far above his fellow-citizens by the superiority of his mind
and the distinction of his character. Although we possess numerous contemporary
portraits, it would be difficult to match the tranquil dignity of this bust of Perikles.
The pose of the head, which is inclined a little to one side and slightly thrown back,
is so personal that it must have been studied from the life. In this pose lies the
secret of that gentleness which is a distinguishing characteristic of the head — a
gentleness arising not from weakness, but from an innate nobility of soul. The artist,
though confined to a bust, has yet contrived to suggest the whole personage. We
can almost fancy that we see Perikles before us raising his head in dignified uncon
cern, however loud the voices of his accusers and slanderers might rise about him.
This outer security well expressed the inner purity of that incorruptible nature to
which Kresilas and Thucydides bear equal witness.
The mouth, which almost borders on ugliness, is specially distinctive.2 Broader
than is customary, with voluptuous lips — the under lip is especially full — it is not the
mouth of a man accustomed to impose his will on others with an arbitrary and iron
determination, but its mobile and sensuous curves accord admirably with the eloquent
manner and persuasive grace of Perikles, and are not without a suggestion of the
tender and faithful lover of Aspasia and of the patron and protector of artists.
The external mien of Perikles was conspicuous not only for mildness and tran
quillity, but for the seriousness of its expression : he had ' a gravity of countenance
which relaxed not into laughter ' (Plutarch, Per. 5). Should any one interpret this
to mean starched and formal dignity — a quality for the rest quite foreign to the
Greek genius — the portrait would promptly undeceive him. A profound seriousness
pervades the features, but there is no trace of self-importance. Rather do we seem
to be looking upon the ideal portrait of the ruler of a democracy, on whom it is
incumbent to be the first in intrinsic merit as in mental distinction.
The other peculiarities are more superficial : as, for example, the mode of wearing
the hair and beard, which are cut somewhat short, the beard allowing the full round
chin to be clearly seen underneath. The short curly hair is very thick, and
the back of the helmet seems to rest on it, as on a pillow. The close, well-ordered
hair and beard are not without significance in the portrait of the man who sur
passed his fellow-citizens even in evKoa/j.ia and evayjipcoavvr). We must also
remember that in the days of Perikles artists left off representing the gods with
long abundant locks, and gave short close hair even to the most august of the
Olympian gods.
The helmet, as noted long ago, marks the office of strategos, which Perikles
filled without a break from about 445 B.C. It would be absurd to suppose— with
later classical authors— that the artist represented Perikles wearing the helmet in
1 Coll. Barracco, PI. 39, 39 a. The head in Munich, called Perikles (Brunn, Glypt. 5th ed. 157),
evidently represents quite a different person, as Friederichs (Baust. p. 125) had already observed.
2 In the Vatican head the characteristic form is somewhat weakened, but the London head reproduces it
with evident fidelity.

VII.
Portrait of Perikles.
IiIUTISI-1 MUSEUM.

Printed by GlesecXeft Devrient, Leipzig {Germany).

PORTRAIT OF PERIKLES

119

order to hide the long shape of his head, for, as we sec from the Vatican replica,
he evidently tried to make this characteristic obvious even through the open
eyeholes. The inscription from the Akropolis, as noted above, is approximately dated at
440 — 430 B.C. Within this period there is no event so likely to have occasioned the
dedication of the portrait in the sanctuary of Athena Polias as the return of the

Fie. 46.— Teiminal bust of Perikles in the British Museum. (From the original.)

victorious general from the most difficult and glorious of his campaigns — the one
undertaken against Samos in B.C. 439. Gratitude for his successful return would
afford a specially suitable occasion for his friends to dedicate his portrait to the
goddess who had lent him her aid. The work would thus have been executed at
a time when Perikles had roused immense popular enthusiasm by his funeral oration
over those who had fallen in the war — an occasion on which the women adorned him

120 KRESILAS
with crowns and chaplets (Plut. Per. 28). Maybe that the image of Perikles as he
stood there before the people on that day inspired the artist at his work.
In turning to consider the formal peculiarities of the work, we note first that the
eye is very long and narrow, and, although the glance is inclined upwards, it yet seems
scarcely opened, while the lid has a heavy droop. The lids are very strong and full,
and the lachrymal glands are prominent. The distance between the inner corners of
the eyes is not great, and the root of the nose is narrow.1
The face is, in the manner of the time, represented with the smooth skin tightly
drawn over and without any lines, except that on either side of the root of the nose,
at the birth of the eyebrows, the artist has introduced a little vertical wrinkle, and
below it, to mark off the nose, two little horizontal cuts. These lines are of course
appropriate to the serious thoughtful man, no longer in his first youth, but the way in
which they are introduced and executed must certainly be peculiar to the artist.
The distinguishing quality of the hair is its plastic fulness, indicated in spite of its
shortness. It is not arranged symmetrically, but is composed of a maze of little curling
locks. Similar locks occur here and there in the beard,2 though the hair of the beard
is on the whole still shorter, and instead of curling merely turns up at the ends.
It is scarcely likely that this portrait of Perikles was the only work of the kind
undertaken by Kresilas, — and, among the many similar helmeted heads preserved,
it is not unreasonable to hope for some trace of his hand. The Berlin collection,
for instance, possesses a marble head,3 which — allowance being made for the differ
ences of copyists — resembles in the highest degree the peculiar style of the Perikles
(Fig. 47). It has been placed on a term that does not belong to it ; the neck below
the beard and the hair at the back are modern. Nothing else is restored except the
lower half of the nose, the top of the helmet, and the point of the visor. The head
is turned a little to the right. The copyist has made great use of the drill both in
hair and beard.
In the general conception of the individual, a quality more easily felt than
described, there is singular coincidence between the two portraits, as may be seen
by comparing them to other contemporary works. The separate features, however,
are in some respects completely different. The Berlin head has long hair, parted
in front and smoothed away behind the ears, at the back falling loose upon the
neck, thus affording no opportunity for the little close curls of the Perikles. But for
all that the treatment is closely allied ; as in the one the little curls, so in the other
the .larger strands, cross and intertwine — perhaps just a trifle mechanically. The
beard, too, in the Berlin head is longer than in the Perikles, but has the same round
outline, and is composed of similar little meshes of hair with curling ends. As the
beard is thicker, these meshes do not lie close to the chin and cheeks in the same way,
yet the resemblance would be materially greater had not the copyist, by using the
drill freely, introduced an element quite foreign to the original. The likeness to
the Perikles is enhanced by the manner in which the beard grows from the cheek and
by the cut of the moustache, which leaves the line of the lip quite free. The hair
is kept well away from the under lip, another point in which the head resembles
1 The proportions, measured on the London head, are as follows : clear length of eye = 36 mm. ; without
lachrymal gland = 32 mm.; height = 12 mm. ; distance between inner corners only 28—29 mm- instead of a
whole eye-length as elsewhere. The length of mouth, according to a very usual proportion = \\ times the
length of the eyes (54 mm.) ; it seems that the length of the mouth (a feature peculiar to the man pourtrayed)
brought the unusually long eye in its train.
2 At any rate in the Vatican head, which in this respect is the more faithful ; cf. p. 117.
s Skulpt. No. 311.

I

122 KRESILAS
the Perikles. Then in the eyes we have again the heavy eyelids and pronounced
lachrymal glands — just a little altered by the hand of the copyist — together
with the relatively narrow nose.1 It is especially instructive to find that the
Berlin head displays, upon an otherwise smooth skin, precisely those same
stylistic little lines at the root of the nose which in the Perikles were seen to
be treated in a manner quite personal to the artist. The helmet differs slightly from
that of the Perikles in having just above the ears what seems to be the turned-up
end of a broad leather strap.
The Berlin head, then, may be accepted as a copy of another portrait of a
Greek general from the hand of Kresilas, rather later than the Perikles, for in
spite of substantial agreement the treatment of hair and beard gives an impression
of greater freedom and of technical advance.2 We may think of it, therefore, as
executed somewhere about the beginning of the Peloponnesian War.
III. The Diitrephes.
Another celebrated work of Kresilas mentioned by Pliny (xxxvi. 74) represented
a wounded man on the point of death. An ingenious theory connects this work with
the bronze statue of a certain Diitrephes pierced with arrows, noted by Pausanias 3 in
his description of the Akropolis as standing close to the Athena Hygieia the basis of
which may still be seen on the eastern side of the Propylaia. In the year 1839
a square basis inscribed ' Epp,6\vico<; Aieirpecpos dirapykv Kpecr/Xas eiroeaev 4 was found
built between the Propylaia and the Parthenon ; and, since this basis might easily have
got shifted farther eastwards from its original position for later building purposes,
L. Ross at once suggested that it had belonged to the statue of Diitrephes mentioned
1 The head is life-size, like the Perikles ; the nose and lower part of the face have the same proportions;
the eyes are shorter (33 mm. ), but the mouth is also of normal width. The distance between the inner corners
of the eyes is a full eye-length.
• There also exist copies of two beautiful heads of strategoi by a master closely related to Kresilas, though
he is quite distinct from him both in general conception and in the treatment of details. The first is the so-called
' Themistokles ' in the Vatican (Helbig, 482); the second is the Pastoret head (Fr. -Wolters, 484), of
which the Munich head in the Glyptothek, 157 (F.-W. 483), is only a mutilated replica, though valuable on
account of the careful execution of what remains ; a third replica of the same portrait is in the Villa Albani, 40.
It is especially in the peculiar treatment of the beard that the same hand becomes apparent for both heads. The
' Themistokles ' is probably the earlier work ; the Pastoret strategos has very disordered hair, and sundry
realistic details, especially in the neck. An immediate development of this tendency in portraiture is the Archi-
damos (Rom. Mitth. iii. PI. 4), in whom I recognize the Archidamos of the Peloponnesian War (cf. infra, p. 321).
It is not impossible that the personality at the back of these works (with which many others may be grouped)
is that of Demetrios, the famous av8panroiroi6s, who worked at the time of the Peloponnesian War and at
the beginning of the fourth century. In later antiquity it was a commonplace of the rhetors (as appears from
Quinctilian and from Lucian, Philops. 18) to find fault with Demetrios as a realist. Probably this merely goes
back to some exaggerated statement in an Attic comedy (it is only necessary to remember the sentence passed by
the older comedy on the realist Euripides) ; for no artist can go beyond the limitations of his time, and actually
realistic portraits, as understood in later days, would be out of the question at that epoch. The caricatured
description in Lucian is moreover put into the mouth of the lying philosopher Eukrates, and cannot in the least
imply what the moderns have taken it to mean.
3 From the wording it might be only the statue that was struck by arrows, but both internal and external
evidence are against that supposition. Moreover, arrows shot at the statue could not stick in the bronze but
would glance off, and the marks that they would leave would not be of a kind to show what they came from.
4 Lowy, Inschr. Gr. Bildh.No. 46, where see all the older literature. For the latest discussion of the question,
Jahrb. d. Inst. 1892, p. 185 sqq. (J. Six), cf. infra, p. 123, n. 2. My earlier opinion (Ath. Mitth. v. p. 28), that— as
Bcrgk first supposed— the inscription referred lo the statue of Hermolykos (Paus. i. 23, 10), and that Pausanias
was mistaken in identifying him with the Pankratiast of Herodotos, I now withdraw as improbable ; for
Pausanias could scarcely have overlooked the difference of the father's name. Cf. still more recently
Weizsacker in Fleckeisen's Jahrb. 1886, p. 16 seq.

DIITREPHES 123
by Pausanias. Certainly the son of Diitrephes would be the most likely person to
have set up the statue, since Diitrephes, who was represented as dying, could not well
have done so himself. Further, Ross also identified this Diitrephes pierced with arrows,
which the inscribed basis showed to have been a work of Kresilas, with the wounded
man by Kresilas mentioned in Pliny. The one serious objection to the theory is that
the information given by Pausanias about Diitrephes is impossible to reconcile with
the date of the inscription. According to Pausanias, he was a general in 414 B.C.,
while the characters of the inscription belong to a considerably earlier date. Kirchhoff
accordingly {C.f.A. i. 402) rejected the whole identification. There is, however, no
definite proof that the information of Pausanias really applies to the Diitrephes of the
Akropolis statue. Pausanias has evidently no personal knowledge of the matter, and
is merely drawing from Thucydides (vii. 29, 30), where it is narrated how Diitrephes
led back the Thracian troops in the year 414 and took Mykalessos by the way. This
same general is again mentioned by Thucydides (viii. 64) in connexion with the events
of the year 411: we know nothing about the manner of his death. Clearly this man
cannot be the Diitrephes of the Kresilas basis. Thucydides, however, also mentions
another and earlier Diitrephes as being the father of a certain Nikostratos (iii. 75 ;
iv. 1 19, 129), who was an Athenian general at the beginning of the Peloponnesian War ;
and it is not impossible that this father of Nikostratos was also father of Hermolykos,
and therefore identical with the Diitrephes pourtrayed by Kresilas. Like his son
Nikostratos, and like his later namesake, he may have been an Attic strategos, and
have been slain by arrows in some engagement of which the record is lost to us. Con
sidering the scantiness of tradition concerning the Pentekontaetia, it is not surprising
that so little should be known of him ; but we at any rate do know that just about
the time when this elder Diitrephes would be active there took place the last bloody
campaigns against the Persian hosts and those great expeditions to Cyprus (459 —
449 B.C.), in all of which the Attic troops would be brought face to face with Oriental
bowmen. It is certainly not difficult to believe that Pausanias only remembered the
Diitrephes whose exploits were recounted at greater length by Thucydides, and knew
nothing of his earlier namesake.1 The death of Diitrephes, whose statue was
dedicated to Athena by his son, probably in fulfilment of a vow, must have been
an episode well known at the time. The general, holding out to the very last, may
have secured privileges to his family by his heroic death, and would thus well deserve
the honour of a statue at their hands. Pliny's description shows that the man was
actually represented as dying — besides, any less serious situation would have cheated
the sympathies of the spectators.2
Pausanias could not gather from the extant inscription that Diitrephes was the
name of the wounded man : he either learnt it from older literary sources, or, as seems
more probable, the name of the person pourtrayed was inscribed on the actual statue.3
The statue of the older Hermolykos, the hero of Mykale (Herod, ix. 105), was on a
different spot of the Akropolis, nearer to the Parthenon and next to the hoplitodrome
Epicharinos (Paus. i. 23, 10). He was probably represented as a victorious athlete,
and his prowess in the pankration was probably known to Herodotos from his statue
on the Akropolis.
1 Ldwy, loc. cit. p. 37, has already hinted at this possibility.
2 The purely arbitrary theories propounded by J. Six, Jahrb. d. Inst. 1892, p. 185 seq., have been successfully
refuted by A. Gercke, Jahrb. d. Inst. 1893, p. 113 seq.
3 So Myron inscribed his name on the thigh of his Apollo (Overbeck, S. Q. 537). Since the Diitrephes dates
about the middle of the fifth century, it would still come within the limits of the period in which it was usual
to set inscriptions upon the statues themselves. Cf. R. v. Schneider, Erzstattie vom Helenenberge, p. 20.

124

KRESILAS

From the actual basis of the Diitrephes wc
learn1 that the figuie was not lying but standing,2
and further that the main weight rested probably
upon the ball of the right foot, which was drawn
back and fastened with a strong peg, while the left,
with the sole on the ground, must have been ad
vanced. A vase-painting from a white-faced Attic
lekythos (Fig. 48) 3 shows that the attitude was not
an impossible one for a wounded man. This single
figure can well be conceived of as a statue. The
warrior, pierced by two arrows, is falling backwards ;
he is naked, as beseems the idealizing treatment of
sculpture ; he only wears a helmet, and carries
shield and lance. The latter he holds still raised,
although it is evident that he can never drive
home another shaft.
As it is precisely among the class of vases to which
this lekythos belongs that reproductions of statues
have been found,4 it is reasonable to suspect one
here. The lekythos, to judge from its shape, is
almost contemporary 5 with the red-figured vases
of the fine period, and cannot therefore be much
earlier than about 450 B.C. Therefore
it perhaps really gives us a free render
ing of the actual statue of Kresilas
directly after it was set up.6 The sculp
turesque nudity of the figure and the
uniqueness of the representation 7 both
favour this supposition. I know of
one similar figure only (Fig. 49) ;8 it
occurs on a gem. Here the warrior is
pierced in the breast by an arrow, the knees give
way, the left hand holds the shield, the right not a
lance but a sword. The gem belongs to a special kind of later Italian product,
closely connected with the severe scarab style (about the fourth to the third

Fig. 49. — Gem in
Berlin (slightly
enlarged).

Fig. 48. — White-faced Lekythos.
(Bibl. Nat., Paris.)

1 Michaelis, Ath. Mitth. i. 289 ; Six, loc. cit. 187. 2 Weizsacker, loc. cit.
3 Original in Paris, Cabinet d. Med. ; Luynes, Vases Peints, PI. 16 ; photograph in Vases Feints du Cab. d.
Medailles, PI. in A. Poor illustration (with the lightly incised inner lines left out), Jahrb. d. Inst. 1892,
p. 185. Gercke's objections (Jahrb. d. Inst. 1893, 113 sqq.) against using the vase-painting for the recovery of
the statue are not valid, still less so is Kekule's remark (Arch. Ztg. 1893, 76) that the vase was too archaic
to be compared.
4 Cf. Loschcke in the Ath. Mitth. v. p. 381 ; PI. 13 ibid, gives the statue of a hoplitodrome on its basis,
which Loschcke supposes may be that of Epicharinos on the Akropolis.
6 Shape and decoration are those of the red-figured Iekythoi of the style mentioned. The eye is approximately
oval. Helmet and shield are only indicated in outline. Just at this time purely outline designs first make their
appearance on Iekythoi. The incision of the muscles is thin and light. Cf. Ath. Mitth. 1891, Taf. ix.
(Mayer), p. 307 seq.
6 Benndorf made this suggestion some time ago, but without publishing it. In the meantime Six expressed
the same hypothesis.
7 The wounded warrior falling backwards, a favourite figure of Duris and his school (cf. Robert, Scenen d.
Ilias u. Aethiopis, p. 6 sqq. ), is similar and yet very different : it shows what Kresilas had to work upon.
8 Tolken, Verz. der Geschn. Steine, iv. 257 ; Winckelmann, Descr. iii. 230.

DIITREPHES 125
century B.C.), and it also may give an indirect echo of the statue by Kresilas. The
position of the legs on gem and vase corresponds roughly with the requirements
of the basis of the Diitrephes.
We may further expect to find traces of the statue among the marble copies of

Fig. 50. — 'Gladiatore Farnese' in Naples. (The restorations are omitted.)
later times, since it is mentioned by Pliny among celebrated works, and since it stood
in so accessible a spot. It has been shown (p. 115) that the statue must be the
earliest of the works of Kresilas known to us, and must be dated about 450, so that
it doubtless retained a certain severity of treatment. All the required conditions are
met with in a torso at Naples, there restored as the so-called Farnese Gladiator

126 KRESILAS
(Fig. 50).1 It represents a man wounded to death, still standing upright on both
feet ; the left is advanced and flat on the ground, the right is drawn back, and rested
probably only on the ball. In order to preserve his equilibrium, the man places his
legs as far apart as possible, yet it is apparent from the bending knees and the
backward inclination of the trunk that he is already losing his strength.: he will soon
grow faint, and in another moment he must fall, for the principal wound is near the
heart. Thus the breath is already leaving him ; the heaving breast and contracted
abdomen show that he pants for air. In effect, here is a volneratus deficiens, in quo
possit intellegi quantum restet animae, and the muscular man, almost exaggerated in
his strength — there is a fine heightening of effect in so representing him — is the sure
prey of death.
The small but fast-bleeding wounds on two corresponding places in the right
and left breast favour the supposition that they are inflicted by arrows. In the
bronze original actual arrows were doubtless inserted. The equipment of the
warrior may be pictured approximately from the vase and gem : he would wear the
shield on his left arm, and carry either the lance or, more probably, the sword, as on
the gem, in his right hand ; 2 on his head he would wear a helmet. The supporting
tree-trunk was, of course, absent in the original, and the position of the legs would in
consequence produce a much more direct effect.
A certain hardness and severity of style at once strikes the eye.3 The pro
minence of the lower edge of the ribs and the straining of the whole trunk are
true and correspond to the situation, though the divisions effected by the straight
abdominal muscles are rather harshly indicated ; the transition from the deltoid to
that portion of the large pectoral muscle that adjoins the clavicle is also character
istic, and not less hard are the nipples with their sharp rim (in the original they were
doubtless inserted). Other noticeable features are the stream of blood, which is
represented plastically and truthfully, the swollen veins at the birth of the arm, and
the rendering of the navel with the little skin round it. In all these things, as also in
the working of the flesh, we recognize an artist striving, within certain limitations
of style, regardlessly after realism, and expressing himself with force and energy ;
the feeling for measure and repose is still quite foreign to him.
The statue gives a hint of the master to whom the youthful artist must have
attached himself in Athens. Not only does it recall the descriptions of the Ladas
who lost his breath,4 but also the Marsyas of Myron, which, in the instantaneous
attitude with legs wide apart and trunk thrown back, offers a close parallel to this
wounded man. We shall enlarge upon these resemblances in a wider connexion.
A work of decidedly later date presents so much affinity to the one just
described that it must be mentioned here. It is a statue in the Vatican (Fig. 51)3
1 Museo Borbon. vii. Tav. 25; Clarac, 870, 2210 ; 872, 2210. Cf. Gerhard and Panofka, Neapels Antike
Bildw. i. 14, No. 30.
2 The gem is also more faithful than the vase in indicating the situation of the wound ; on the vase the arrows
are sticking only in the leg, because they could not well introduce them into the crowded upper half of the picture
without coming into collision with arm, shield, or lance.
3 It is especially marked in the working of the little tight curls of the pubes ; they recall the hair of the
Perikles, but have a still more archaic quality.
4 Brunn, K. G. i. 264, relying on Pliny's description, laid stress on the resemblance between the wounded man
and the Ladas. Cf. J. Six, loc. cit. 188.
5 Sala della Biga, from the Villa Mattei ; Helbig, 329 ; first published in Venuti, Mon. Matth. i. 101
(here with an older restoration ; right forearm stretched forward, right foot placed on a flat rocky elevation, the
support draped; the left arm unrestored) ; then in Visconti, Mus. Pio-Clem. 2, 42; Pistolesi, Vatic. 6, 10, I ;
Clarac, 837, 2099 (the present restorations indicated) ; Brunn-Bruckmann, Denkmdler, No. 129. Cf. Brunn,
Bayr. Sitzungsberichte, 1892, p. 660.

STATUE OF A RUNNER

127

which passes as Alkibiades. It resembles the wounded man in a number of
characteristic points, such as the contracted abdomen, the powerful chest that
expands in the act of fetching a deep breath, the position of the upper arm, and the
sharp marking off of the deltoid from the large pectoral muscle. Also, so far as the
indifferent copy permits us to judge, the indication of the veins on the upper arm
and the treatment of the flesh are similar, though more harmonious and less hard.1
The head is fortunately un
broken, but the greater part of
the face (almost the whole
right side, the nose, mouth,
chin, and the front part of
the left jaw) is restored. What
remains shows unmistakable
indications of being a portrait.
The beard on the left cheek is
very similar to that of the
Perikles, the stylistic treat
ment of the little locks being
the same. The hair has a
very individual character; it
is not crisp and curling as in
the Perikles, but smooth and
straight, and in life was ap
parently worn cut short, not
parted but combed down over
the forehead. These peculiari
ties the artist has rendered
clearly, without however sacri
ficing any plastic quality : on
the forehead, for instance, the
hair lies full and massive, while
all over the head the separate
locks tangle in and out in life
like confusion.2 The forehead,
which is prominent above the
eyes (with a flat depression
in the middle), the eyes (the
lids unfortunately are much
damaged),3 and the rather thin cheeks are somewhat less stylistic than in the
Perikles, and bear accordingly a more individual stamp, though the difference is
not vital. The short strong neck, too, must be a feature characteristic of the person
pourtrayed. The restorer has certainly not been happy in placing a helmet under the right
foot. A warrior actually fighting may be expected to have his helmet on his head.

Fig. 51. — Alkibiades in the Vatican. (Attempt at a reconstruction ; old
restorations omitted.)

1 The realistic detail of the navel is likewise modified, and the pubes is less strong : there are, however,
the same little curls, but lying somewhat smoother and formed more freely.
2 The hair at the back of the head is peculiarly fine and sharp, consisting of flat meshes, curved now upwards,
now downwards.
3 The left eyelid wholly ancient, the right partly so ; the eyelids were rather heavily formed.

128 KRESILAS
From the strongly inflated chest, however, it seems more probable that a runner is
represented,1 and that the body was originally bent more forwards so that the right
foot would rest on the ground. The glance is directed towards the goal. The turn
of the head is amazingly fresh and lifelike. The arms from the elbows were probably
both extended ; the right certainly so, as is proved by a puntello above the hip.
The original was of course of bronze, and did not require the hideous support
between the legs.
To find a statue of an athletic victor which is purely a portrait is very note
worthy. We know from Pliny that, in Olympia at any rate, those only who had
won three times were allowed the privilege of a portrait-statue.2 Now at the time
to which the original of this statue must be referred occurred the three Olympian
victories of the celebrated runner— known to Plato, and through him to later writers
— Krison of Himera, who won in the three successive Olympiads 83, 84, and 85
(B.C. 440). It is quite possible that his portrait has survived in this statue, which
must in any case have been made soon after the date of Krison's third victory, or
only a trifle later, since its style is nearly allied to that of the Perikles.
In the whole range of fifth-century art there are no works with which this
Vatican statue is more closely and narrowly connected than with those referred to
Kresilas. We must therefore be bold to rank it among them, and to recognize
in it, as compared with the wounded man, an advance on the part of the artist in the
direction of harmony, repose, and a softer technique.

IV. The Amazon.
Besides the Perikles and the wounded man, the wounded Amazon, the third
of the celebrated works by Kresilas mentioned in Pliny, also survives in copies.
Among the several Amazon types preserved in statues, there are two which
represent the heroine as wounded. One of these has been generally attributed to
Polykleitos, for reasons so sound that they need not be even discussed ; the second
type however needs to be studied all the more closely, inasmuch as the wound in
this case is in a far higher sense the fundamental motive of the statue, and accounts
for the title volnerata. O. Jahn had claimed this — the so-called Capitol type — for
Kresilas, and it is an error of later times to have diverged from his view. The
external probability that in the numerous copies of this type we have the volnerata
of Kresilas is increased almost to certainty — to such certainty as is possible in these
matters — by comparisons of style.
First, however, a few words must be said concerning the much-debated question
of the ' Ephesian Amazon statues.' Polykleitos, Pheidias, Kresilas, and Phradmon
were the four sculptors who, according to the well-known passage in Pliny, simul
taneously and in competition made each an Amazon for the temple of Artemis at
Ephesos. This statement has met with undeserved contempt, though it should
rather be considered as confirmed from the fact that copies of precisely four statues
of standing Amazons still exist, which on the one hand are clearly to be referred to
four different artists, and, on the other, are evidently closely connected by identical
1 Cf. the statues of the Pal. Conservat. Bull. Munic. 1876, Tav. 9, 10 (Helbig, Museums, 573 — 575), which
seem to me to represent runners, not wrestlers. The prominent edge of the chest shows the hard breathing.
2 Statuae iconicae, rightly referred by Scherer to portrait features of the face, De Olympionic. Statuis, diss.
Gott. 1885, p. 9 seq.

THE AMAZONS 1 29
measurements,1 by a general similarity of conception and of dress, and by their
belonging to the same period of art.
This important fact is in my opinion beyond all doubt. In the first place, four
types are preserved, and not three only, as is generally supposed. True, the fourth type

Fig. 52. — Amazon in Villa Doria-Pamfili. (Wrongly restored as an Artemis ; the right arm, the left Irom the
middle of the upper arm, the legs from the knee downwards, the dog, are modern.)
exists only in one copy, and that a mere torso (Fig. 52) ; it stands in the Villa Pamfili,
and has been commonly and erroneously ranked among the replicas of the Polykleitan
type,2 from which, however, it is absolutely distinct. It certainly comes nearer to that
1 Length of foot in the three types = 30 cm. 2 Michaelis, Jahrb. d. Inst. i. 16, H.
S

130 KRESILAS
type than to any of the others, but it is somewhat severer, and designed with more
uniform symmetry.1 This would be in favour of an attribution to Phradmon, who,
being an Argive, must naturally have had points of contact with Polykleitos. The
fact that only one copy of Phradmon's Amazon has survived, and that this shows his
statue to have been the weakest and least original of the four, not only bears out the
judgment recorded by Pliny — according to which his name figures last in the list of
competitors — but also explains the exiguity of his fame.
It has further been denied that these four types could have arisen in one and the
same period, but so far as I can see without a shadow of proof. The Mattei type is
said to be distinctly later than the others, yet the treatment of the dress with its
girding, as well as of the body, brings it on the whole within the same stylistic period,
while it bears every mark of fifth-century art. In fact, we have already noted (p. 30)
that the stylistic treatment of the dress is very closely related to that of the torso of
the Medici Athena, i.e. of the Promachos. Unfortunately, none of the copies of this
type preserves the original head ; for the head of the Petworth replica, which Michaelis
took to belong to the statue, is undoubtedly foreign to it,2 and may not impossibly
belong to the Phradmonian Amazon.
It has also been denied3 that Pheidias and Polykleitos could have been working at
the same period, although we have the classic testimony of Plato to the fact.4 In the
Protagoras (p. 311 C), as is well known, he brings in Polykleitos and Pheidias as con
temporaries, and in another passage in the same dialogue (p. 328 C) he speaks of the
sons of Polykleitos, and says expressly that they are of the same age (^Xt/acorat) as
the sons of Perikles, who, as appears from p. 319 E, is thought of as still living. There
is no doubt therefore that, in the eyes of Plato, Polykleitos belonged to the age of Perikles,
and that his life and activity coincided with that of Perikles and Pheidias. Now Plato,
it is true, was not unfrequently guilty of anachronisms in matters of detail ; he seems
to have confused the first and the second stay of Protagoras in Athens, and not to have
known the exact time when a certain comedy was performed. But it is impossible that
he should have placed a celebrity of the first order like Polykleitos in quite the wrong
period. The supposition of Robert,6 that Polykleitos worked about 435 — 390, and that
Plato in this dialogue, written probably before and certainly not much later than 390,6
referred the sons of a still living man back to a remote age, is still less conceivable.
No stronger witness than that of Plato could be required to prove that Polykleitos
was a contemporary of Pheidias, even though he doubtless survived him.
Therefore, far from there being any grounds for doubting Pliny's story concerning
the Amazons, the copies preserved of the four types show that there is every reason
to believe it. The only improbable part of the anecdote is the statement that the
artists worked competitively, and, as -self-constituted judges, determined the respective
merits of their work. It is however easy to see how such a story may have arisen :
1 This is especially clear in the folds between the legs and the compact kolpos with its straight lines. The
motive was so far similar to that of the Polykleitan Amazon that the right upper arm was raised, the left lowered.
But she certainly is not wounded, and the left hand does not seem to have rested on anything. The chest, which
is quite covered by the chiton, corresponds in its broad outlines to the Polykleitan.
2 As Loschcke and I ascertained at Petworth in 1888. Cf. Arch. Anz. 1890, p. 164. The fillet corresponds
to the remains of the head of the Pamfili statue. The character of the head (Michaelis, Jahrb. d. Inst. i.
p. 27, 'Typus III.') would well suit the Argive Phradmon and the Pamfili statue. Michaelis lays stress on
the severe but not melancholy expression of the face.
3 Robert, Arch. Mdrchen, p. 100 seq.
4 Cf. Scholl, Sitzungsberichte d. Bayr. Akad. 1888, i. 42, note I. 5 Arch. Mdrchen, p. 98 seq.
6 Bergk, Griech. Litteraturg. iv. 440, dates the dialogue as early as 407 B.C. Christ, Platon. Stud. p. 46,
as late as 387 B.C.

THE AMAZONS 131
the four statues probably formed together one single offering, and, like similar groups
of the fifth century and even later times,1 they presumably stood side by side upon one
large plinth, although each was a work complete in itself. In addition to the general
inscription referring to the whole set, the artist's inscriptions would be placed under
their respective statues. The names being celebrated, there was only a step to the
invention of the anecdote of the certamen, which was possibly based on some earlier
verdict, current in the Argive school, ascribing artistic pre-eminence to Polykleitos.
This offering would scarcely be made by the Ephesians themselves, but by some
rich man who tried to get the most prominent artists of the day. The subject, single
figures of Amazons, is fully explained by the great importance of these heroines in
the legends of the Ephesian temple. According to the earlier tradition, it was they
who founded the sanctuary,2 who set up the miraculous image of the goddess, and
celebrated her in the dance.3 Later a particular Amazon, Otrera the wife of Ares,
was named as founder of the temple.4
Since two of the Amazons are represented as wounded, it was thought that the
commission must have stipulated for the heroines to be represented as fugitives and
seeking refuge in the sanctuary, thereby recalling the tradition that the Amazons
pursued by Dionysos, and later by Herakles, fled as suppliants to the temple and
there obtained quarter. Not only however are two of the Amazons represented
unwounded,5 which would alone suffice to disprove the proposed theory, but even the
statues of the two wounded Amazons can scarcely be reconciled with the supposed
situation, for they have not in the least the character of suppliants ; whereas it is the
seeking refuge with the goddess that is the kernel of the tradition : the Amazons
were imagined as actually seated upon the altar in the manner customary with
suppliants.6 Finally, that legend seems to be quite late ; it is only found in historians
of Imperial times,7 and the story of the pursuit of the warlike Dionysos, on
which especial stress has been laid, was probably merely elaborated from the story,
originating after the campaign of Alexander, of the exploits of Dionysos among
the Indians ; by analogy the god was made to fight with other wild races also,
such as the Tyrrhenians, Lydians, and Iberians.
The commission seems only to have stipulated for single figures of Amazons, not
on horseback, but standing ; they were to be an offering to the goddess, whose
cult and sanctuary the Amazons had founded. With regard to such externals as size
and dress, the artists probably came to an agreement among themselves, in order to
avoid any great want of harmony.
The question was how to make an Amazon, composed as a single figure, charac
teristic. Legend invariably represents the Amazons as courageous combatants ;
but— after a brave resistance — they are always the vanquished. To give effect to
these two essential features, nothing could seem better adapted than to represent
an Amazon with a wound in the breast, as becomes a brave fighter. Another
main characteristic, in the conception of the fifth century at any rate, is that
1 Cf. the monuments commemorative of battles set up by Attalos I. in Pergamon.
2 Pindar makes the Amazons who fought against Athens found the sanctuary of the Artemis of Ephesos
(Paus. vii. 2, 7).
3 Kallim. Hym. in Dian. 237 ; the Amazons set up the bretas.
* Hygin. Fab. 223, 224.
5 The figure referred above to Phradmon is certainly not wounded. 6 Tacitus, Ann. 3, 61.
7 The Ephesians themselves refer to it before the senate in the reign of Tiberius (Tacitus, Ann. 3, 61).
Pausanias cites the same legend in order to refute Pindar, who had ascribed the founding of the sanctuary to the
Amazons (Paus. vii. 2, 7). In Plutarch's account (Qucest. Gr. 56) the Amazons, pursued by Dionysos, flee from
Ephesos, where they had settled, and betake themselves to Samos,

132

KRESILAS

they are daring horsewomen. Though the terms of the commission evidently
excluded representation on horseback, this characteristic could still be suggested :
hence the ' Mattei type ' — the Amazon preparing to leap. The artist remembered
doubtless the epithet iro\vcrK:apd[io<;, the strong leaper,
applied in the Iliad (ii. 814) to the Amazon Myrine.
Kresilas, as Pliny informs us, represented his
Amazon as wounded. Let us now examine more
closely the 'Capitol type,' which must be his1 (Fig. 53).
The restoration of the statue proposed by
Michaelis may, it seems to me, be improved upon in
one important respect. Michaelis restores the figure
as grasping the spear high up with the right hand,
the elbow being only slightly bent.2 He is guided
in this by the Louvre copy, in which the right arm
seems to be for the most part antique. On looking
at the original, however, it seems more than doubt
ful whether the pieces out of which this supposed
antique arm is made up really belong to the statue ;
while, on the other hand, the familiar gem which is
the basis for the restoration of this type gives a
different position of the arm, bringing the right hand
quite close to the head instead of far above it. It
becomes evident, on experimenting with a model, that
the attitude reproduced on the gem is not a caprice of
the engraver, but must have been the original motive.
It is infinitely more natural, and affords a real sup
port, at the same time producing a much finer and
more self-contained rhythm than the restoration pro
posed by Michaelis. In the gem however, for want of
space, the right hand is placed rather too low ; to corre
spond with the traces of the upper arm left on the
Worlitz torso, the hand must have had hold' of the
spear a little above the head (Fig. 5 3). There is
thus presented a wounded combatant, leaning heavily
upon her spear, for the left -leg on which she stands •
does not suffice to support the body ; with her disengaged hand she draws away hej/
dress from the smarting wound, and her head is inclined wearily to that side.
1 The replicas have been carefully collected by Michaelis (fahrb. i. p. 17); to his list should be, added a
head, which he excludes, in St. Petersburg (ibid. p. 18, note 3) ; it has little interest, except as showing the extent
to which a fine Greek original can be disfigured at the hand of a wretched copyist. There is at Dresden also
an old cast of a head of this type, the original of which I am not acquainted with (perhaps = Michaelis m. • cf. c).
It is a fair replica. Finally the head of the term in the Villa Albani (No. 76) deserves notice ; it is nothing
but an adaptation of this type of Amazon to a decorative purpose ; in these terms long hair on the shoulders was
popular, and is accordingly added, little as it suits the head ; the turn of the head is also completely changed (cf.
supra, p. 66, note 5). A similar transformation is seen in a head of the Barracco collection, also probably from
a term. One of the best of the copies, Michaelis '1,' is given in Fig. 54 from the cast ; the head is wrongly set
upon a statue of the Mattei type. It is unfortunately not perfect, the nose, part of the under lip, the chin and
the neck being restored, but even so it is decidedly better than the head of the torso in Worlitz, which hitherto
is the only one well published. The best copy I know is n (Michaelis), in the Conserv. (Pal. ; Helbig, Museums, 579.
2 See the drawing in Jahrb. d. Inst. i. p. 28 ; Helbig, Museums, 503.
3 Remains of a support on head o (Michaelis, loc. cit. p. 18, 33) show that head and hand were connectec',
I have not examined the original.

Fig. 53. — Amazon of the Capitoline type.
(Restored.)

Fig. 54.— Amazon head of the Capitoline type, wrongly placed on the Mattei statue in the Vatican. (From the cast.)

'-1

I34 KRESILAS
The stylistic qualities most emphatically confirm the attribution to Kresilas.
They readily catch the eye on a comparison of the head with the Perikles. Above
all, the eye has the same characteristic shape, elongated and narrow, with thick
heavy lids and strongly marked lachrymal glands, met with neither in Pheidias nor in
Polykleitos. The only difference is that the distance between the two eyes is greater 1
and the root of the nose rather broader. The replica given in Fig. 54 has preserved
in the little folds, which run along the rims of both lids,2 an interesting detail, which
is slurred over in most of the copies. In the hair is seen the same principle of plastic
fulness as in the Perikles.
If the attribution of this 'Capitol type' to Kresilas be accepted, its special
qualities may be next applied to the further characterization of that artist. They
are most clearly seen by a comparison with the Polykleitan statue. The contrast,
the profound gulf that separates them, is well emphasized by Michaelis.3 Kresilas ¦
has selected the attitude and posture of his Amazon, not with an eye to mere beauty, «
but primarily because he felt them to be true and appropriate to a wounded person, <
while Polykleitos is entirely concerned with the beauty of the motive.4 Kresilas enters .
into the feelings of the wounded woman, and works from the spirit to the form.
Polykleitos aims primarily at a beautiful pose and pleasing drapery. That this
contrast exists is undoubted ; at the same time, we must not go too far, and regard
Polykleitos as giving his Amazon a wound on the right breast (Fig. 5 s)6 absolutely
without thought. Certainly the wound would be fretted by the raised arm, and
the pain only increased ; yet the artist may have had in his mind the true
Doric ideal, an heroic maiden heedless of pain ; he may have intended to give
expression to the constant napTepelv united with the fullest K007UOTJ7? in bearing
and dress.
It would thus be his special conception of the subject which induced
Polykleitos to give a beautiful position to his wounded Amazon, making her support
herself as if merely tired,6 with one arm on her head regardless of her wound.
Besides, in the position adopted, resting on the right leg and with the head turned
to the right, the artist has scarcely left himself an option except to place the wound
on the right side, leaving the left completely secondary in importance.
Polykleitos represents the self-controlled masculine woman, retaining her brave
bearing even when wounded. Kresilas represents only the wounded woman, natural
and human, giving way to her pain and trying to lessen it. In the details of the
statues the contrast is strongest in the dress. The Polykleitan Amazon wears only
a short tunic, which leaves the powerful breast as free as possible ; but this tunic
is disposed in the most elegant pleats, and the portion below the hips is treated quite
decoratively. In contrast to this elegance, the simplicity of Kresilas is conspicuous.
1 The distance is 34 mm., a full eye-length.
2 These small folds have nothing to do with the ' bronze technique ' of the original, but are a mark of style.
On the other hand, the fine groove round the lip represents the edge which in the bronze separated the inserted
lips from the rest of the face.
3 Jahrb. d. Inst. i. p. 41 seq. * Cf. Kekule, Idolino, p. 12.
6 The existence of the wound, as Michaelis correctly infers from the material at his disposal, can no longer
be doubted. For the agreement of the copies makes it impossible to regard the wound, identically introduced in
all, as a capricious addition of the copyists. That some of the copies omit the drops of blood is assuredly only
because these were indicated by painting. In the bronze original they must have been rendered plastically. The
bronze statue of the Chimaira at Florence, whose antique Greek origin can now scarcely be doubted, has a wound-
cut plastically indicated with drops of blood, precisely like the two wounded Amazons (on the lion-body a drop
of blood was even especially let in ; it has now fallen out). Cf. also Fig. 50 and p. 126.
6 Michaelis is right in supposing the supporting pillar to be part of the original. The copies beam him out.
The whole attitude of the figure requires a support ; without one it would fall,

THE AMAZONS 135
Here again his first aim is to be natural. The folds show no artistic arrangement,
no attempt at decorative effect, but only a conscientious endeavour to indicate the
character of the material. Although the copies differ considerably in the dress, all

Fig. 55. — Head of Amazon in Lansdowne House. (From the original.)
reproducing it carelessly,1 the general character is unmistakable. Note, for instance,
the simple straight folds falling from the left breast, the kolpos below the belt, the
lower edge of the tunic cut off in a straight line, and broken only in the middle where
1 The two Capitoline statues correspond fairly well, but Michaelis c is the better copy. In the Colonna
copy the folds of stuff are altered, and made more like those of the Polykleitan statue. The Torlonia copy
has less severe drapery than the Capitoline.

136 KRESILAS
the stuff is a shade drawn up in dainty folds,1 and lastly the neat pleats proper to fine
linen (cf. supra, pp. 29 and 46), which play a large part, especially in the lower
portion of the tunic. The artist did not, like the creator of the Mattei type, under
stand how to combine with these pleats the livelier motive of actual folds, and his
Amazon appears in consequence the more archaic;2 he sought to make his drapery
not rich and pleasing, but only simple and true.
Another token of the feeling of Kresilas for sincere characterization is the
rider's cloak which he gives to his Amazon, to the complete disfigurement of the
back view, so pleasing in the other statues. For it is as the cloak, appropriate to
horsemen in the fifth century, that we must interpret the piece of drapery which
is fastened round the neck in front, and falls down behind in heavy folds. As we
learn from vases,3 this cloak — which was introduced from the north — was of heavy
material ; and the hard, straight, almost ungainly folds about the neck in front
must be taken therefore as true to life. The Parthenon frieze, it must be owned,
represents the rider's cloak much more pleasingly. It is, finally, a delicate trait
to make the wounded Amazon press a corner of the cloak to her side with her left
elbow, as though she were cold.4
On the Polykleitan statue the waving of the hair has nothing individual or
characteristic ; it was simply the mode of the time, and we find it given, especially
in the Peloponnesos, to the most different women. The waving hair of this Amazon
of Kresilas is, on the contrary, something original and eminently characteristic, that
we find nowhere repeated ; at once warrior and horsewoman, she has cut off the long
locks of her thick hair and simply gathered up the shortened ends into a knot — a
mode evidently adopted without regard to appearance, but for freedom of movement
and to save time and trouble.6
The mouth with Kresilas is more in the plane of the face and blended in a way
true to nature with the adjoining parts, while with Polykleitos the lips protrude and
are more strongly formed in order to retain a clear plastic shape. Again, with Kresilas
the cheeks, and still more the nostrils with the adjoining parts, are more fleshy and
living, and the forehead has more modelling than with Polykleitos, who makes
everything flat and abstract. In like manner, Kresilas makes the breast more
fleshy and breathing, while Polykleitos lays special stress on the strong bony
structure. Finally, both heads, in accordance with the general style of the
century, show nothing of that contraction of certain muscles caused by pain ; they
try rather to express the suffering only through the general bearing. Yet here again
there is a difference. The Polykleitan head, apart from the statue, has absolutely
no indication of suffering ; in the Kresilaian head the observer, even looking at it
alone, would at once detect a certain nameless sadness, which is tangibly expressed
only in the slight droop at the corners of the mouth.6
The head of the third Amazon ('Mattei type') is unfortunately unknown. It
would be of the greatest possible interest to bring it into comparison, for from all
that has been said it follows that, as O. K. Miiller maintained, we have in this statue
1 This detail, which occurs in the Polykleitan statue, may be seen in Michaelis c ; in b, f, and h this part is
restored. 2 If the Amazons were not all made in the same year, this one, and not the Polykleitan, should be the earliest.
3 Cf. e.g. the interior of the Geryones Kylix by Euphronios.
4 The Worlitz figure seems not to have reproduced this motive.
D Cf. the cameo in Comm. in hon. Th. Mommseni, p. 479 ; the intaglio Cades, iv. A, 118 ; further, a beautiful
carnelian in Berlin, Gemmen, Inv. 1933 b (head to the right ; fairly exact copy ; indication of the mantle under
the neck). 0 cf Michaelis, loc. cit. 45. 26 seq.

THE AMAZONS 1 37
the Amazon of Pheidias. The little that, in addition to its mention by Pliny, we
know from Lucian of this work goes to confirm this conclusion. Lucian describes
it, evidently to distinguish it from other Amazon statues, as the one leaning upon a
short spear, ttjv eirepeiBop,evr]v tg> BopaTiw. Now, as a fact, the Mattei Amazon leans
upon a lance,1 preparatory to swinging herself upon her horse. It is true that the
' Capitol type ' also leans on a spear, but in this statue the wound is so evidently
the leading motive that it would naturally give the descriptive title, while in the
Mattei type, on the contrary, it is precisely the strong anticipatory leaning of both
arms upon the lance that first strikes the eye and is the most noticeable feature. It
has indeed been said2 that the lance of the Mattei type could not be described as a
BopaTtov ; but is that term any more applicable to the spear of the ' Capitol type ' ?
And could a spear on which an upright figure would lean be anything materially
smaller than that required by our Mattei Amazon ? By BopaTtov, then, Lucian must
have meant a lance of this kind ; and this is not very surprising, for he was probably
mentally contrasting it with the Macedonian lances 5 J metres long. Nor can any
exception be taken to the expression eirepeiBofievr) ; for the Mattei Amazon is leaning
just as much as the wounded Amazon, and is on the point of leaning yet more
heavily for the spring, employing moreover both hands in the action.
The Amazon of Pheidias differs from all the others in its distinctive motive ;
even the Amazon of Kresilas, resting as she does on one leg with the other drawn
back, has less affinity with the Pheidian than with the Polykleitan type. ;The ,
Pheidian Amazon is preparing to leap, and holds the spear almost as required for
the actual spring, but the left hand has not yet tightened its grasp, and will have
to take a little higher hold of the spear. The right arm is restored by Michaelis at
rather too high an angle ; it would be more natural for it to come more forward ; the
spear, as the Mattei gem indicates, was held obliquely, so that the point came farther •
to the front. The left foot rests lightly on the ground, as if feeling for a firm footing.
On this foot the run would be started, and from it the spring be made on to the
horse — whose head must be thought of as facing — the right leg being passed over the
horse's back as she swings herself into the seat.3 In a similar way the ' Diskobolos
taking up position ' 4 stands on the left leg and feels tentatively with the right before
transferring to it his whole weight. And as in the Diskobolos a whole series of
movements must intervene before the throw, so in the Amazon before the leap. She
must first raise the spear, tighten the grasp of the left hand, and take the run ; then
set the spear on the ground again and make the spring. Her attention is naturally
1 Michaelis should have emphasized what he says only dubiously (loc. cit. p. 45) ; a lance is proper to the
warlike Amazon, and not a leaping-pole, such as the boys used in the gymnasium and palaestra when learning
to ride (as on the Attic Kylix, Arch. Zg. 1885, Taf. II, 'p. 183 ; cf. also Holwerda, Jahrb. d. Inst. iv. p. 39).
On the Natter gem the rim cuts off the design immediately above the hand, so that the point of the lance does
not appear. The reproduction in Overbeck, Plastik, 3rd ed. i. 393, is wrong, for it gives the staff as ending
off at the top ; in the Natter design it is cut through by the rim, and thus incomplete. This gem, a convex root-
emerald (plasma di smeraldo), seems moreover, according to Natter's statement, to belong to a certain class of
stones especially in favour at the time of Caesar and Augustus, affording far and away the most numerous
and faithful reproductions of statues. This class of stone was, so far as I know, never counterfeited in the last
century. On these grounds, though I know the gem only from Natter, I feel able to answer quite positively for
its genuineness. It is and remains the solid basis for the reconstruction of the statue. A recently unearthed
bronze statuette in the Museum at Verona is known to me only through a photograph (Arndt-Bruckmann s E. V.
No. 8) ; this suffices to show that it is an unmistakable forgery ; it is one of a whole series of forgeries known
to me, which display just such defects in the casting as the statuette ; the head belongs to no definite style, and
is quite without the character of the antique : in the right hand the forger has copied the restored remains of a
bow from the Mattei type. 2 Cf. Michaelis, loc. cit. p. 47.
3 Cf. the Kylix, Arch. Ztg. 1885, Taf, 11. * Helbig, Museums, 330; cf. suira, p. 90.
T

138

KRESILAS

wholly directed to the left. With this accords the carriage of the head, which is
not to the right, as Michaelis was deceived into thinking by the head wrongly placed
on the Petworth statue, but a little to the left. This is fully
proved by the antique piece of the neck remaining on the
Petworth statue,1 and also by the half-neck preserved in the
Turin replica.2 Directly it is restored with the correct carriage
of the head, the statue gains extraordinarily in life and ani
mation (Fig. 56).
As regards the lost head, I cannot refrain from a con
jecture which has been gaining upon me, although sundry
external circumstances are against it. May not this type be
preserved in that well-known bronze term which at one time
stood in the Villa at Herculaneum as companion to the
Doryphoros term (Fig. 57)? The carriage and turn of the
head are not, it is true, those required, for the head turns a
little to the right and is set too straight on the neck : we
know however that the copyists were not over particular about
preserving the carriage of the original, when they adapted
heads of statues to the term-form. The Herculaneum head
passes indeed as a replica of the Polykleitan Amazon,3 though
it is sufficiently extraordinary that it should ; in reality the
heads have scarcely a superficial likeness, and are fundament
ally distinct, both in the facial forms and in the arrangement
and treatment of the hair. On the bronze head the hair does
not, as in the Polykleitan Amazon, lie smooth upon the skull,
but forms a thick, heavy, waving mass. The hair about
the forehead, too, is not rolled back, but simply combed to
either side. The arrangement at the back is like that of the
Polykleitan head, but the execution is quite different. On the
other hand, we find that nowhere can the hair be better
paralleled than in the works which we have referred to Pheidias : the way it grows
from the parting is similar to the Lemnia. We may also recall in this connexion
the girl's head (Figs. 14, 15, p. 59) in which we recognized a youthful work of
Pheidias ; both the hair and facial forms make it clear that this is only an earlier
stage of the type given in the bronze term. In the latter the facial forms, though
imperfectly and coarsely rendered by the copyist,4 have a distinct Pheidian character ;
the full mouth, with the strongly arched lips, is especially noteworthy. Such a
mouth explains why Lucian went to the Amazon of Pheidias for the o-TopLaTOi;
dpp,oyrj of his ideal beauty. In this particular the term may be compared with the

Fig. 56. — Amazon type. (At
tempt at a restoration.)

1 Cf. Arch. Anzeiger, 1890, p. 164.
2 In the Turin statue (Michaelis, e) the lower half of the neck is antique, and shows a slight turn to the left.
The statue — only the torso is antique — was a careful copy executed on a reduced scale, about three-quarters,
because of the costly material (green basalt). This is exceptional, for most copies of statues, other than colossal
works, are either of the same size as the original or quite small statuettes. In the other copies the neck is
missing. The supporting figure from Luku in Athens (Exp. de la Morie, iii. 88 ; 1. Sybel, Catalogue, 442 ;
Durm, Gr. Baukunst, 2nd ed. p. 259), which adapts the type to a decorative purpose but copies it badly, makes
the head look almost straight out, but keeps a slight turn to the left.
3 Michaelis, loc. cit. p. 16, I. Cf. P. Wolters, Gipsabg. p. 233.
4 The term is probably by an artist inferior to the one who signed his name on the shaft of the corresponding
term. The hair is good, but seems to have been cast and put on separately. The face is poorer. The eyeballs
are restored, as is the case in most heads from Herculaneum.

IX.
Head of Perseus.
BRITISH MUSEU.M.

Printed by Giesecke & Devrient., Leipzig (Germany).

Fig. 57.— Bronze terminal bust of an Amazon from Herculaneum (Naples). (By permission of Messrs. Bruckmann, Munich.)

140 KRESILAS
two heads in which we conjectured an Aphrodite and an Eros by Pheidias (Figs.
20, 21, p. 67 seq.) The facial forms are entirely different from the Polykleitan type:
there are, for example, none of the retreating planes so characteristic of both the
• Amazon and the Doryphoros of Polykleitos. Now, if the stylistic qualities of this
'bronze term point to Pheidias, it is probable that it reproduces his Amazon, to
• whom the powerful structure of the face (the Lemnia looks quite soft and delicate
• by contrast) is eminently suitable.
^ The Mattei type may then be conjecturally restored with a head of the type of
the Herculaneum bronze, and in it may be recognized a work of Pheidias (Fig. 56).
The dress of the Mattei type has certain definite features that mark its period.
As was seen above (p. 30), it is especially closely related to the Promachos (B.C.
445 — 440), a work of the Pheidian school. The attribution to Pheidias is thus con
firmed. But now the possibility of more exactly dating the Amazons is brought
within our reach. Taking the Parthenon as standard of comparison, we find that
they coincide, not with the figures of the pediments, but with the later metopes.
»The Amazons all display a manner of indicating fine linen by parallel folds which
•is not seen either in or after the Parthenon pediments. Kresilas adheres most to
the bare defining of the material ; Polykleitos lets this fall into the background, and
aims rather at beautiful though over symmetrical folds. The Mattei type, on the
other hand, contrives to unite the folds indicative of the stuff with genuine folds, in
a rich but natural manner which recalls the linen chiton of the woman on the
Parthenon metope, South XXIX (Michaelis).
? The Ephesian Amazon statues would thus date roughly about 440, that of Kresilas
coming close to his Perikles. For Polykleitos the date is especially significant,
for it gives us a work by him twenty years earlier than his gold-ivory Hera. The
Amazon of Pheidias was a brilliant achievement of the most brilliant period of the
master's career. The wealth of motive in the dress, which finds a parallel only in
the Parthenon, and seems to anticipate the pediments ; the original arrangement of
the garment — the more lifelike and natural because quite unsymmetrical — so well
adapted to the situation ; the framework of the body, so free from stumpiness or
heaviness ; finally, the clear and freshly conceived motive — all tend to give to the
work the bold and untrammelled note which has erroneously been taken as sign of a
later origin.
Finally, as to the way in which the four statues stood in the temple at Ephesos,
no more satisfactory arrangement can be devised than to place the Amazon of
Pheidias next to that of Kresilas, and on its left ; the supporting legs of each statue
would thus be on the outside, the two spears on the inside, and both heads would be
turned inwards. The Amazons of Polykleitos and Phradmon, too similar to look well
side by side, would then come at either end. Assuming this arrangement to be
correct, it might almost be suspected that the anecdote of the evaluation of the statues
by the artists merely grew out of the order in which they were placed : first, to the
left, would come Polykleitos, then Pheidias and Kresilas, and last of all Phradmon.
We know of only one other figure of an Amazon famous in antiquity — the
statuette of Strongylion, so dear to Nero. It too appears to be preserved in a copy ;
for the attribution 1 of a bronze statuette of an Amazon on horseback to Strongylion
is extremely probable, since that artist was celebrated for his horses, while the style
of the bronze points clearly to a late fifth-century original.2
1 Apparently first expressed by M. Hoffmann, Philol. 1865, 402 ; cf. also Overbeck, Plastik, 3rd ed. i. 476,
note : r4- '* Especially in head, hair, and drapery.

ATHENA FROM VELLETRI 141
The fact that there are more replicas of the Amazon of Kresilas than of any of
the others speaks highly for the appreciation in antiquity of the qualities peculiar to
this work. It has, in effect, a directness of sentiment and force of conception lacking
in the others. Kresilas, living at a time when Pheidias dominated the artistic life of
Athens, was yet able to preserve a perfect independence and individuality.

V. The Athena from Velletri.
In presence of the head of the Amazon attributed to Kresilas, Michaelis 1 felt
reminded of that famous type of Athena preserved in the Albani bust in Munich, and
in the statue from Velletri in Paris. He is right ; for this work also may be brought
within the range of Kresilas.
The admirably preserved colossal statue found in a Roman villa at Velletri, and
now in the Museum of the Louvre (Figs. 58, $g),2 is the one complete copy remaining
of a magnificent composition which is generally taken as the most perfect expression of
the character of Athena.3 The goddess stands in a majestic attitude ; she rests upon
the left foot and draws up the right, which she has moved rather to the side. This side
ways position of the free leg constitutes a difference between her attitude and the
simple walking motive of the Polykleitan Doryphoros. The same pose, designed
apparently to give greater breadth to the lower portion of the body, and to lend a
certain repose in the movement, was seen in the Farnese Athena in Naples (Fig. 26),
referable to Alkamenes. The garment, as in the Lemnia and the Parthenos, is the
Doric peplos, and it is similarily girt with a snake : it is not however open at
one side, but sewn up ; the gorgoneion on the aegis is almost exactly similar to
that in the best copies of the Parthenos, though the actual aegis is much narrower
in front, and resembles rather that of the Athena on the west pediment of the
Parthenon. The border of snakes has of course that richer form introduced with
1 Jahrb. d. Inst. 1886, i. 27.
2 Frohner, Notice de la Sculpture, No. 1 14. The two hands and the lower half of the right forearm are
restored ; the rest of the right arm is antique, but has been twice broken ; at the right elbow there seems to be a
bad join, the forearm was certainly more bent. The arm with the nude part under the armpit is antique,
made in a separate piece. The nose is intact. The statue is made of coarse-grained ' Thasian ' marble.
There are two copies of the torso in the form of statuettes : (a) Broadlands (Michaelis, p. 225, No. 31) ; (b) in
the Pal. Conservat. Rome. The head by itself is preserved in several replicas : (a) the famous Albani bust
at Munich (Glypt. 92), the head of which is thought to be better than that of the Paris statue ; certainly the work
is rather more animated and less dry, but for fidelity in detail the statue may claim the advantage. This appears
from a comparison even of the accessories ; such as the aegis, which in the statue displays a much richer trimming
of snakes, and the gorgoneion, which in the statue is more severely and evidently more faithfully modelled. In
the form of the eyes and of the mouth too the Velletri replica is the more reliable ; the mouth is more austere and
less soft and fleshy than in the Albani bust ; (*) in the Berlin collection (Skulpt. 79) ; it is of inferior workmanship,
yet it agrees in the main with the statue in the formation of eyes and mouth, and in this respect is better than the
Albani bust ; (c) poor and hastily worked replica in St. Petersburg (Guedeonow, Mus. de. Se. 176 ; (d) in Lans
downe House (Michaelis, p. 469, No. 93); (e) Brit. Mus. (Anc. Marbles, i. 1) ; (/) in Madrid (Hiibner, No. 92).
Finally there is a seventh reproduction, without value for the knowledge of the original, but interesting in another
respect— i.e. (g) it is the head from the monument of Eubulides in Athens (Ath. Mitth. 1882, Taf. 5), originating
therefore somewhere about the middle of the second century B. c. It is not a copy in the same sense as the others,
for the artist intended the work to pass as his own, and only fell back upon an older original from want of inventive
faculty. Incapable however of entering into its special character, he brought in contributions from his own
style, and his reproduction is thus only a travesty of the original. At that period real copying, as understood in
Roman times, was unknown, as is proved by the Pergamene imitations of older statues (cf. p. 27) : these are never
close copies. For the monument of Eubulides cf. the researches of Milchhofer, which settle the question (Arch.
Stud. H. Brunn dargeb. 1 893, p. 44 seq. )
3 Ottfr. Miiller (Handb. § 369) describes the ideal of Athena from this work.

Fig. 58.— 'Pallas de Velletri' (Paris)

ATHENA FROM VELLETRI

H3

the Parthenos (cf. p. 1 1). The snakes along the upper border of the aegis are an inno
vation.1 A cloak of the same heavy woollen material as the peplos lies over the left
shoulder and is wound round the body below the waist, falling in a great three-
cornered drapery. The artist has, it seems, borrowed this mode of wearing a cloak
from male figures like Zeus ; 2 and to it is due a good deal of the majestic impression
produced. The head is nearly erect, and has a scarcely perceptible turn to the side
of the supporting leg — a turn which enhances the repose and grandeur of the figure.
With a lance ifl her right hand, held obliquely and grasped high up, not in the least as

Fig. 59. — Head of Athena from Velletri.

if she required support, she displays the serene dignity of her godhead without any
touch of pathos. In strong contrast to the right arm, the left lies close to the side
with the hand held out, as if to carry an object of a certain weight. It has been
conjectured that this must have been a Nike which Athena, goddess of victory,
would here, like the Parthenos, carry as her co-ordinate daemon ; and, in effect,
this conjecture has been confirmed by an Athenian bronze coin (PL VI. 30),3 which
evidently reproduces the type of our statue. This coin is significant, too, as teaching
1 Cf. Puchstein, Jahrb. d. Inst. 1890, v. 85, note 20.
- For the rare instances in early times of Athena wearing a mantle, cf. Roscher's Lexikon d. Myth. i. 696,
3 Imhoof-Blumer and Gardner, Num. Comm. on Pausanias, PI. Z. 22, p. 133.

144 KRESILAS
that the original of the statue stood in all probability at one time in Athens ;
for it is known that the types of the gods on these later Athenian bronze coins
were as a rule taken from famous works of art in the city. The fact that later
the Athenian Eubulides took this statue as model is also in favour of this
supposition. The general qualities both of dress and head prove beyond doubt that, as is
almost universally admitted, the statue must belong to the great second half of the
fifth century. Pheidias himself has been thought of,1 but our present knowledge
of that master quite excludes this notion. The Velletri type is quite foreign to
all the representations of Athena with which we are familiar in Pheidias and his
school, and, on the other hand, there is absolutely no ground to take it, as some
have wished, as Polykleitan.2 Nor can so thoroughly individual a work be brought
under the rubric of a school.
The original, as already noted, stood probably in Athens. The snake girdle and
the arrangement of the folds about the girdle show further that the artist must have
been acquainted with the Parthenos, while the narrowness of the aegis suggests that
the figure can scarcely be earlier than the Parthenon pediments. On the other hand,
the treatment of the lower folds of the chiton, between the cloak and the feet, shows
that we may scarcely venture to exceed that date ; 3 indeed, from the folds alone
the statue would have to be dated much earlier. They are in reality treated in a
pre-Pheidian manner : they are very uniform and not deep, and the ridges are for
the most part simply rounded ; the style of Pheidias, even in the Lemnia, was more
advanced. On the right side, the seam which holds together the peplos is indicated
in the same realistic manner as in an extant original statue (probably Parian) of the
severe style (about 460 — 450).4 The dress, finally, is not allowed to fall over the foot
of the supporting leg, as in later Pheidian works. The head contrasts strongly in
shape of helmet and type of countenance with that ideal of Athena which the
influence of Pheidias and his circle impressed upon the Athens of the second half of
the fifth century, and which, as we learn from many minor works of art, had a wide
spread popularity up to the times of Alexander.5 The Pheidian Parthenos and her
derivatives have full, well-rounded features, surmounted by the round Attic helmet,
which like the hair is treated decoratively. Our statue, on the contrary, wears the
Corinthian helmet, and displays a complete neglect of the decorative element in
helmet as in hair ; the bony structure of the face is clearly apparent, the expression
is full of mature gravity, as beseems the thoughtful goddess.6 The artist found the
elements which he thus embodied ready to hand in the Peloponnesian art of the
severe style ; 7 by deepening and expanding them, he created a type which has never
been surpassed. The forehead is high in the middle (the only statue that comes near
to it in this respect is the Farnese Athena), and broad above the eyes. The cheek
bones are not very prominent, but the chin is strong and bony, and, in profile, the
1 Thus Wieseler, Denkm. d. Kunst, ii. 144 ; Botticher, Verz. d. Abgiisse, No. 672.
2 So Wolters, Gibsabg. p. 552, 225 ; also Kekule, Ann. d. Inst. 1868, p. 318, by false analogy with the
' Hera Farnese ; ' for the latter work cf. infra, p. 223, n. I.
3 Cf. Puchstein, Jahrb. 1890, p. 85, who calls it a ' little later than the Parthenos.'
4 In the Villa Ludovisi, Schreiber, 29 ; Helbig, 889 ; Brunn-Bruckmann, Denkm. No. 357. Braun (Ruine
und Museen, p. 594) noticed the same. Cf. Arch. Studien H. Brunn dargeb. 1893, p. 81, note 62.
6 Cf. Roscher's Lexikon, i. 697, 700 seq.
6 Cf. Feuerbach, Griech. Plastik, ii. 23 ; describes the Albani bust as ' pure abstract thought in embodied
marble.' 7 Cf. Roscher's Lex., loc. cit. The relief of the Akropolis (suira, Fig. 4) shows Peloponnesian influence
in the head.

ATHENA FROM VELLETRI 145
under jaw is very marked. The hair is smoothed back in the plainest and most
unpretentious way.
It again follows from all this that the artist, living in Athens during the thirty
years of Pheidian supremacy, must have preserved a perfectly independent line, which
left him free to give a characteristic colour, derived from Peloponnesian influences,
to his conception of Athena.
Sundry details lead yet further. If a cast of the head of the Velletri statue be
compared with the Perikles bust, the similarity is quite striking in many particulars,
but especially in the formation of the eyes ; 1 both have the long narrow slit and the
singularly thick, substantial lids. To this must be added that the helmet of the Athena
resembles that of the Perikles down to the smallest detail ; and since the Corinthian
helmet admits of great diversity of treatment in the detail — as is proved by examina
tion of other heads of Athena and of strategoi, in all of which the helmet differs
somewhat from that on the two heads under discussion — this resemblance may serve
as clue to guide us to Kresilas.
The relation of the head to the Amazon has been already justly observed by
Michaelis. It appears quite clearly in the eyes, in the mouth, and in the profile ; the
delicate modelling of the brow, too, is similar ; parallel with the clear-cut eyebrows
there is a flat depression which disappears in the centre where the forehead becomes
a trifle prominent. The simple compact folds of the woollen riding-cloak worn by
the Amazon admit of direct comparison with the cloak of Athena ; further, the severe
pleats in the linen chiton of the Amazon — the uniformity of the folds and the level
termination of the lower edge, the fundamental sacrifice of all mere decorative effect
to the true rendering of the stuff — are characteristics that reappear in the woollen
garments of the Athena. A certain greater freedom and boldness in the execution
of the Athena may be explained by its being later than the Amazon : it cannot how
ever have been much later, and must have been executed while the artist still lived
in Athens, previous to B.C. 430.
We may reasonably expect to find some trace of so signal a work in literary
tradition, and I believe, in effect, that it is actually mentioned by Pliny among the
works of Kresilas. Directly after naming the wounded man and the Perikles of Kresilas,
Pliny makes an additional statement and, drawing apparently from a Roman source,2
mentions in terms of the highest admiration a Minerva and an altar in the harbour of
Athena, in the great and splendid sanctuary of Zeus Soter — naming as the artist a
certain wholly unknown Cephisodorus.3 That sanctuary, the Disoterion, has been
reasonably assumed to be contemporary with the superb laying out of the Peiraieus
by Hippodamos in the days of Perikles,4 and it is only natural to suppose that the
great cultus-statue and the altar were set up at the same time, presumably by an
artist from the entourage of Perikles. Now the original of the Velletri statue would
be admirably adapted to Athena Soteira, the powerful and wise protectress, who
would certainly be represented in a colossal statue and would very probably carry
a Nike, to whom sacrifices were offered in conjunction with Zeus and Athena ; 5
1 The other copies of the Athena have lost the characteristic formation of the eyes. This is especially the
case in the Albani bust ; the Berlin copy is better.
2 Cf. Oehmichen, Plinianische Studien, p. 151, who suggests Varro or Mucian.
3 Thus the MSS. The modern reading Cephisodotus is a mere conjecture. Pliny treats of the two
Cephisodoti in quite another connexion under the heading of those artists who eptsdem generis opera fecerunt.
4 Wachsmuth, Stadt Athen, ii. 141 seq. ; cf. i. 560.
6 Cf. Wachsmuth, loc. cit. ii. 144. Pausanias (i. I, 3) mentions a sceptre and a Nike for Zeus, only a
lance for Athena, but this does not exclude a Nike. Milchhdfer, too, in Arch. Studien H. Brunn. dar^ebr.
J893> P- 4^, note 2, interprets the Velletri statue as Soteira, but I think he is quite wrong in detecting in it the
style of Kephisodotos. U

146 KRESILAS
while it would be excellently appropriate for Eubulides to have derived his Paionia,
the goddess of healing, from the Soteira, the saving goddess. It seems not impossible
that Pliny intended this addition for Kresilas — from the list of whose works it is
divided off only by the word Cepkisodorus. It has perhaps taken the place of some
work by Cepkisodorus which stood originally under his name.1 With Pliny's method
of writing, this is only too possible ; and it may at least be surmised that the Athena
of Kresilas, which we have been describing, is no other than the Soteira of the
Peiraieus. What is however certain is that this noble creation of Athena affords a new
proof of the capacity of Kresilas — by independent absorption in his task — to work
from the spirit to the form, and so to create the permanent and the enduring.

VI. The Diomede.
The series of works that may be linked on to the Perikles and the Amazon can be
considerably enlarged. A chief addition — a work of extraordinary force of invention
— is the Diomede of Munich. Flasch and Brunn have recently emphasized the close
connexion of this work with the ' Alkibiades' of the Vatican {supra, p. i27,Fig.5l), and
point out the ' absolutely identical artistic spirit.' 2 And I know, on the other hand,
from personal communication, that Loschcke and Studniczka have anticipated me in
the conjecture that we have here a work of Kresilas.3 The confidence with which I
include it in my list is increased by this community of opinion with other scholars.
The reproductions of the statue in Figs. 60, 61 are taken from the latest casts,
from which the restorations have been removed f only the lower end of the garment
in front remains to be discounted. There is a replica in the Louvre,5 but it is very
inferior to the Munich statue, and displays the carelessness of the copyist, especially
in the head, where the variations from the Munich replica are of no import to
the knowledge of the original, being purely due to neglect of detail — the hair in
particular is flatly and dully rendered. The head has been broken off and unskil
fully replaced with the help of plaster, but it belongs indubitably to the statue —
which has the further advantage of being preserved to below the knees. It can be
seen that the left leg was drawn back in the walking attitude ; by the side of the
1 A few paragraphs further (xxxiv. § 79) Urlichs has noticed that the Autolykos of Lykios seems to have got
among the works of Leochares. The Tyrant-slayers are introduced by Pliny in the same book under Praxiteles
(34, 7°) and under Antignotus (34, 86). It is probable that both notices are wrong, and that the ' Tyrant-slayers '
belong to Antenor and Kritios (cf. Benndorf, Annali, 1867, 306).
2 Flasch, Vortrdge in der 41 Philologenversammlung, 1891, p. 9; Brunn, Bayr. Sitzungsber. 1892, p. 660,
673- 3 This was communicated to me by Loschcke in the summer of 1890 at Bonn, where on seeing the cast of
the figure I became convinced that it belonged to the circle of Kresilas.
4 Cf. Brunn, Glypt. 162; Bayr. Sitzungsber. 1892, p. 651 seq. ; Brunn-Bruckmann, Denkmdler, No. 128.
The statue first appears in Lafrerie, Spec. 77 ' Romae,' placed in a niche. It shows here an older restoration, which
is to be seen also in Bracci, Memorie d. Incisori, i. Tav. 23. The left foot is wrongly restored as if placed
flat on the ground, and a cuirass is used as support for the right leg. The Nike on the left hand proves the
identity, liracci calls it 'gladiator victor ' in the possession of Verospi. The present restoration was probably
made for the Albani collection. An older cast of the bust, probably made when the statue was in the Musee
Napoleon, is now in the Ecole des Beaux- Arts, Paris (No. 2739) ; another copy of the same is in Bonn (wrongly
designated by Welcker as ' Biiste im Vatikan ').
5 In the Salon du Mars Borghese (No. 2138) ; Frohner, Notice, No. 128 ; Bouillon, iii. Statues, PI. 2, 3 ; Clarac,
PI. 314, 1438. Photograph edit. Giraudon, No. 1402. Kalkmann is wrong in supposing (Gesichtsprop. p. 34)
that the body is not genuine : there is no doubt whatever that the head and body belong together.

THE DIOMEDE

H7

supporting leg the copyist has used a tree-stem as a prop.1 The naked body
corresponds closely with the Munich statue, but sheath and sword-belt are omitted.
Since the belt in the Munich statue is treated in a manner quite at variance with
the general style, and is thoroughly Graeco-Roman in character (it is shaped

Fig. 60. — Diomede in Munich. (From a cast with the modern restorations omitted.)

like a scarf, fastened with effective loops, and finishes off in a fringe), the assumption
that it is an addition of the copyist is on the face of it probable ; it is yet further
assured by the Paris replica. In this replica the two little flat cross folds which
jn the Munich statue appear on the garment, above the remains of a support, are also
1 The upper part is antique.

1 48 KRESILAS
omitted ; folds of this kind are also peculiar to Graeco-Roman style. The whole
treatment of the drapery on the shoulder is somewhat simpler in the Paris replica,
with less striving after effective depth and with more severity of treatment. In this it
is evidently the more faithful of the two. From the absence of the sword, however, it
need only be concluded that the sword of the original was not fashioned like that of
the Munich replica, and not necessarily that there was no sword at all, for it is known
positively that in the earlier Greek works objects of this kind were always put on
separately. A real belt and sheath, put on to the original bronze, presented the
copyist with no art-form that could be imitated : if he wished to introduce that addition
into his copy at all, he had to invent one for himself, as did the Munich copyist.
It is impossible to know whether the Paris replica may not have had a sword-belt of
some other material put on, for the falling portion of the drapery, which might have
shown traces of the fastening, is not antique.1 The copyist of the Harmodios (Naples)
hit upon a middle course, replacing the real sword-belt of the bronze original by a
painted one, and adding a sheath of other material. An instance, corresponding to
the Munich Diomede, of the plastic fashioning of the sword-belt in a form selected
by the copyist is met with in a replica of the Borghese Ares at Dresden,2 which has
a sword-band adorned with sprays, omitted in the other copies ; it is to be assumed
that here also the original wore a real sword-belt.
The Munich copyist has thus sought to heighten the effect of the original, and
to suit it to later taste, by the execution in marble of the sword-belt and by the slight
change in the dress. Other details fit in very well with this. The hair is very care
fully worked, and, far from its being smoothed over through carelessness, as in the Paris
copy, there is an attempt, just as in the folds, to give by deeper cutting an effect of
light and shade stronger than in the original. This is proved by a third copy, known
only from an old cast of the Mengs Coll. in the Dresden Museum (Fig. 62). This
cast, though in the form of a bust, is obviously taken from a statue. The head, as
can still be recognized, has been broken off and replaced. I have at last, after a long
search, found a trace of the lost original : the cast is evidently from the statue which
Cavaceppi {Race. i. 9) describes as being in London in the possession of ' Enrico Jen
nings ' ; and it was doubtless taken when the figure was still in Rome.3 According to
Michaelis,4 the Jennings collection was sold later by auction, and the statue in question
— described by Cavaceppi as an ' Atleta ' — came into the possession of Lord Cadogan.
Doubtless it is still in England. Now this third replica seems to be the best preserved
and the most faithful in style. The legs appear to be in the main antique ; they
have at any rate the correct walking attitude and the right proportions ; beside the
right leg a tree-stem again appears as support. The garment seems to be perfectly
preserved in this replica only ; it hangs to the knees in simple broad folds. Both
arms are apparently restored from the elbows. The sword-belt is again absent, which
strengthens the supposition that in the Munich replica it is an addition. The head
can be closely compared with help of the cast. At first sight the hair looks very different
from that of the Munich replica, but on a closer examination it is seen that — with very
slight variations— it corresponds lock for lock, and that a difference exists only in the
1 In front only the upper part lying on the shoulder and breast is antique ; at the back the falling end.
2 Becker, Augusteum, Taf. 35.
3 Winckelmann knew the figure, and mentions it with the Albani statue (now in Munich) in Hist, of Anc.
Art. Bk. v. chap. 5, § 35 (=tr. Lodge, ii. 399) on account of the crushed ears. His statement that the Jennings
statue was formerly in the Palazzo Verospi is wrong. The Jennings statue came from Cavaceppi. According to
Bracci, it was the Albani statue that was formerly in the Palazzo Verospi.
4 Anc. Sculpt, in Gr. Brit. p. 93, note 242.

THE DIOMEDE 1 49
working ; in the Munich head the marble is deeply dug out round each lock,
while the numerous flatly chiselled inner lines of the cast are replaced by fewer
and more deeply cut lines. The difference is naturally far more obvious in the thick
hair of the head than in the little flat curls of the beard. There can be no doubt
that in the rendering of the hair the Dresden cast is more faithful to the original than
the Munich statue ; the latter is evidently a further development of some original in
which the hair must have looked very like that of the Dresden head. The piece of
drapery, which in its main outlines' corresponds to the Munich copy while exhibiting
the simpler treatment of the Paris one, and the absence of the plastic sword-belt,
witness further to the greater fidelity of this Dresden replica.
Moreover, it is highly probable that on the Munich head the furrows on brow
and cheeks are cut deeper than they were in the original ; every feature seems
more strongly and sharply marked, the opened lips are more compact, and the
expression of strained energy is thereby enhanced. It is however most improbable
that this expression should have been entirely introduced by the copyist, or that it
could have been so completely foreign to the original as it is to the Dresden head.
The latter seems rather to be the work of an inferior artist, who, while faithfully
copying the forms of the locks of hair, failed — like the majority of copyists — in the
more delicate forms of the countenance, and contented himself with a general dull
rendering of the main points. Thus the modelling of the forehead accords perfectly
as to essentials in the two heads, but in the Munich head it is far more delicately
and sharply executed. The more subtle modulation at the root of the nose1 and in
the cheeks must be regarded, it is true, as exaggerated, but not as foreign to the
original. On the other hand, in the Dresden head the eyes, although rendered
superficially, are archaic in shape, and so doubtless more faithful to the original ; the
Munich copyist had tried to modernize them by rounding the eyeballs, by curving
the upper eyelid, and by the general clear-cut formation of the lids. In the Dresden
cast the formation of the eye is similar to that of the Athena from Velletri and
to the Perikles : on the lower lid may be seen a fine incision, on the upper a more
strongly marked fold.2
The chief result of this critical examination of the replicas3 is to show that
Brunn's arguments4 against the reference of the composition to the fifth century are
unsubstantial ; for the shoulder-belt and its tassels, the working of the chlamys and
the hair in the Munich copy, on which he bases his objections, have been shown to be
later introductions of the copyist.
We now turn to the examination and explanation of the composition.
Brunn's interpretation of the Munich statue is, in my opinion, indisputable : it
represents Diomede carrying off the Palladium from Troy.5 The left hand bore the
idol, which, hewn from the same block of marble, was joined to the body below the
1 On the original of the Dresden cast the nose was probably restored.
2 The Munich copy of the Diomede bears the same relation to the Dresden copy as the Borghese copy of the
Anakreon does to the other copies, which are more exact in details (cf. supra, p. 60).
3 Besides the three mentioned I know of no other full-size reproduction of the statue. Two other replicas
in statuette size: (a) Terme Museum, only a torso; (b) Berlin, Skulpt. 515, torso restored. Of the head there
is a possible replica in Madrid (Hubner, No. 189). [In the spring of 1893 I saw a full-size torso in the Terme,
with sword-belt and sheath ; phot, in German Inst, at Rome. — E. S.]
4 Bayr. Sitzungsber. 1892, p. 656 seq.
5 The objections adduced by Flasch (loc. cit. p. 9 seq. ) against Brunn's interpretation carry little weight.
F. interpreted the statue as a 'boxer' on account of the swollen ears, and almost in the same breath as a warrior,
because of the sword, and on account of the lance with which he proposes to restore the statue as a Doryphoros. '
In the end he leaves it quite uncertain which of all these interpretations he himself adopts, nor does he give any
explanation of the naked sword,

Fig. 6 j.— Head of the Munich Diomede. (From the cast.)

Fig. 62.— Cast at Dresden of a replica of the Diomede. (Original presumably in England.)

152 KRESILAS
left shoulder by the great support on the drapery.1 The lower portion of the right
arm, which hangs down and was presumably turned a little outwards, was joined by
a support to the thigh, and the hand grasped the naked sword ; the empty sheath
hangs at the left side.
This interpretation2 is decisively confirmed by the fact that the motive of the
statue, as numerous monuments prove,3 became typical of Diomede the stealer of the
Palladium, and this not merely in later times, for one of the monuments dates back,
as will appear, to the epoch in which the statue itself originated. The motive in all
its details can be fully and wholly explained by this interpretation alone ; and when
other figures are found represented with the same motive, a transference from the
Diomede to them, and not the reverse, must be supposed.
The hero has accomplished the bold deed, has torn away the miraculous image
from the sanctuary, and is in the act of retreating. It is now a question of keeping
his booty and repelling assaults from others. To this end, pausing in his stride, he
turns his head to the side with a quick energetic movement ; he scents danger, and
is on the look out with strained attention, ready with his drawn sword to defend
himself at any moment. The entire motive of the statue — the stride, the head turned
to the side away from the supporting leg, the sword held lowered and ready in the
right hand, the precious booty carried in the left hand — all take their rise necessarily
out of the situation represented. The beauty of the motive, the effective contrast of
the two sides, have not been invented for their own sakes ; they are not formal but
purely practical in nature, and have an entirely objective purpose.
This purpose can be yet more closely defined ; for the statue has reference
evidently to the tradition that Diomede was threatened as he retreated by the envious
Odysseus, who came upon him in pursuit from behind. Diomede, with his back still
to him, is warned by the gleam or the shadow of the sword in the moonlight, draws
his own sword, and Odysseus, discovered, relinquishes his design. This tradition,
treated in the little Iliad, is of Argive origin ; 4 it celebrates the heroic king of Argos,
who alone carried off the true Palladium and conveyed it to his birthplace, where he
was afterwards worshipped in a common cult with Athena ; on feast days, his cultus-
symbol, the shield, was carried in procession together with the Palladium. From
this alone it would be probable that the original of our statue stood at one time in
Argos ; the supposition is made certain from the reproduction of this very statue on
an Argive coin of the time of the empire, when celebrated works of art were so
readily copied.5 It was therefore undoubtedly made for Argos, and for the cult of
Diomede there established.6
Another work of art derived from the statue is scarcely less interesting to us.
It is an Attic vase-painting of the last decade of the fifth century,7 representing
^ Brunn's idea of a small Palladium of bronze fastened to the marble support (Bayr. Sitzungsber. 1892, 653
seq.) is, I think, untenable, as being contrary to all known procedure on the part of copyists.
2 Kalkmann agrees with Brunn (Gesichlsprop.p. 34). His theory, however, that the Diomede formed part
of a group is quite untenable (cf. my remarks in Berl. Phil. Wochenschr. 1894, p. 1 142).
i Cf. Chavannes, De Palladii Raptu, Berliner Dissert. 1891, pp. 4, 6, 15, 23, 24, 25.
4 For the tradition of the legend and its origin cf. Chavannes, loc. cit. pp. 42 seq., 78 seq.
5 Coin of Antoninus Pius, Imhoof-Blumer and Gardner, Numism. Comm. PI. K, 44, p. 39 ; cf. Chavannes,
loc. cit. p. 5. The right forearm is a little more raised. [PI. VI. 32 is reproduced from a still more distinct example
of this coin ; it belongs to M. Imhoof-Blumer, to whose courtesy I am indebted for the impression.— E. S.]
This removes Flasch's objection that ' a statuary representation of Diomedes in the period to which the
original belongs is problematic'
7 Naples, No. 3235 A. (Heydemann); Mon. d. Inst. ii. 36; Overbeck, Gallerie, Taf. 24, 19. Cf.
Chavannes, loc. cit. p. 6 seq. The numerous restorations have never been noted ; these are—all of the Odysseus

THE DIOMEDE 153
Diomede with the Palladium almost exactly according to the scheme of the statue ;
the sole differences are immaterial, and lie in the arrangement of the chlamys, in the
introduction of the petasos about the neck, and the wreath on the hair. The painter
has however utilized the motive in a form of the tradition first introduced to the
Athenian stage by Sophokles,1 according to which Helen assists in the theft. The
picture represents the two heroes, Diomede who carries the Palladium, and Odysseus,
both with drawn swords and engaged in a dispute, which Helen appears to be
soothing. This vase-painting, which is for the rest a fresh testimony to the direct
influence of the Attic stage on contemporary vase-painting,2 must either be derived
together with the statue from a common source, such as a wall-painting, or else the
figure on the vase is derived from the statue. The last appears the more probable,
partly because the design is so entirely statuesque in character, and partly because the
figure does not fit in very well with the whole scene, but is conceived rather as a
single figure. This view can only be reconciled with our previous conclusion that the
statue was destined for Argos, on the supposition that the artist worked in Athens,
where the design could easily have passed into the workshops of the vase-painters.
In later antiquity again this statue was utilized in representations of the theft of
the Palladium — as on the beautiful Spada relief3 — precisely because in it had been
found the most pregnant expression for the deed and character 4 of the hero.
Now, though the original must have belonged to the free style of the fifth
century, it is equally clear that it comes under none of the chief tendencies familiar
in this epoch. It does not belong to the narrow Polykleitan circle ; for the contour
of the face and the hair, as also the general bearing and conception, are quite
different. As little does it conform either in head or body to the Attic style of the
Parthenon frieze and pediments and similar works — except in so far as the energetic
turn of the head recalls the Promachos and the Dioscuri of Monte Cavallo — or to
that Attic manner chiefly known from the Munich and Florence athletes.5
On the contrary, it brings us again to Kresilas. The similarity of the beard to
that of the Perikles (especially in the Vatican replica) is startling. We recognize (in
the Dresden cast) those long-slit, heavy-lidded eyes, familiar from the Perikles and
the Amazon ; and again (Munich replica) the marking off of the brow from the nose
by two vertical depressions as in the Perikles. And the likeness of the mouth to that
of the Amazon at once catches the eye, in spite of the difference in the expression.
The hair also — in this respect only the Dresden cast can come into question — is
similar to that of the Amazon and the Perikles ; it consists of heavy, plastically full
(except the upper half of the body and the hands), the shaft of the lance and the handle of the sword
painted above the sheath. Originally the sheath was empty, and Odysseus held the drawn sword in his
right hand. He was of course bearded. In the Helen a piece of the middle of the body, the left arm, the
back of the head, the nose, and the forehead are new. The cloth hung up is quite out of harmony with the
style of the vase, and is also modern. The Diomede is all antique. The inscriptions are given correctly in
Heydemann's facsimile, wrongly in the text and illustrations. The style of the vase is that of Aristophanes and
Erginos, which begins about 430 B. c.
1 In the Aduaivai ; cf. Chavannes, loc. cit. p. 51 seq. 2 Cf. supra, p. no, and Arch. Anz. 1890, p. 89.
3 Schreiber, Hellenist. Reliefbilder, Taf. 7; cf. gems (Chavannes, loc. cit. p. 15), Mon. d. Inst. vi. 51 D ;
Arch. Epigr. Mitth. aus Oester. iii. p. 40.
4 The characterization is carried to the verge of portraiture ; hence Frohner, Not. 128, explained the Louvre
replica as a Roman portrait ; cf. Brunn, Glypt. 5th ed. p. 217. Lately Winter, Jahrb. v. 1890, p. 167, went so far as
to try to trace the original back to Seilanion, on the ground that the Munich Diomede offered points of resemblance
to portraits (e.g. the Plato) by that artist. Brunn in his latest essay on the statue, Bayr. Sitz. Ber. 1892, p. 663 sqq. ,
merely tries to show that the art of Seilanion, like the Diomede, marks a transition from the style of the fifth
century to that of Praxiteles and Lysippos. Previously however (1891) Flasch had thoroughly vindicated the
fifth-century character of the body, and his results are confirmed by the analysis given above.
5 Cf- Mm. d. Inst. xi. 7 ; Rom. Mitth. 1892, p. 81, Taf. 3. Cf. infra, pp 259-262. X

154 KRESILAS
masses, with shallow inner lines carved upon them ; the crown is deeply embedded.
The hair of the so-called Alkibiades (Fig. 51) is very similar ; but there it is evidently
copied from the person pourtrayed, while in the present instance the artist (as will
presently appear) submitted, in the whole arrangement of the close-cropped hair, to
the influence of an older heroic type.
On the ground of the treatment of the hair, the Diomede should be placed in
the same period as the Perikles, the ' Alkibiades,' and the Amazon, i.e. circa 440.
Yet other considerations —the dress, the structure, and the modelling of the various
facial forms — would suggest it to be perhaps somewhat earlier. The dress — abstrac
tion always made of the modernized details in the Munich replica — is closely related,
in its simple true defining of the thick woollen material and in the singular force
with which the folds are rendered, to the Velletri Athena.
The body seems at first sight to resemble the Polykleitan canon ; and it does so,
both in the broad outline and in the details j1 yet a close inspection brings marked
differences to light. The whole body is more compact and more strained ; all the
forms display more tension, more readiness for the energetic manifestation of force.
In Polykleitos, for instance, the lower line of the chest is graduated more gently and
harmoniously — one might say, more schematically — than is the case in the Diomede,
where its projections and depressions produce far richer modulations. In the Diomede
also, as in the other works attributed to Kresilas, is to be found the harsh passage of
the deltoid into the shoulder. The straight and oblique abdominal muscles, too, are
rather more hardly marked off than with Polykleitos, and even the navel, notwith
standing its general similarity, is less flat and more substantial. All these differences
bear witness, however, to the indisputable connexion with the characteristics observed
in the earlier of the works attributed to Kresilas. The Diomede continues their
tradition, tending in the direction towards Polykleitan forms.
The modelling in the face of the Diomede is singularly rich, even allowing for
the fact that in the Munich copy it is probably exaggerated. The bony eminences
of the brow are strongly emphasized. Their junction with the nose has already been
cited as especially Kresilaian. The interciliary region also is perceptible though faint,
since the whole lower half of the forehead projects strongly. The root of the nose,
i.e. the point where the frontal bone and the nasal bone meet, is treated with especial
delicacy and richness, and is carefully marked off from the adjoining parts. The
treatment is similar in the Perikles, though not so advanced. The bridge of the
nose is narrow.2 The delicate, natural shape of the no:e affords a good contrast to
the schematic shape affected by Polykleitos. The nostrils are of singular energy
in their swollen cartilaginous formation. Finally, the spare cheeks and the parts
about the mouth are richly modelled. On the other hand, in the strong emphasizing
of the oblong, angular shape of skull may be recognized a tendency towards the
Polykleitan canon.
All these observations tend to prove that the Diomede most probably originated
in the period circa 440 — 430. This is the epoch to which we referred the Velletri
Athena, a work with which the Diomede corresponds in every respect. It will be
remembered that in the Athena the Polykleitan influence was apparent in the
walking attitude.
This result admirably fits in with the required condition, that the artist of the
1 E.g. the formation of navel, abdomen, and pubes.
In the Munich copy the nose is antique. A characteristic point is that the distance between the inner
corners of the eyes corresponds not to the full eye-slit length, but only to the length without the tear-gland.

THE DIOMEDE 155
Diomede made in Athens a statue destined for Argos. The later removal of
Kresilas to Argos, which we dated after 430, would thus be prepared for by earlier
relations with that town, attested both by the fact of the commission and by the
influence of the Argive school.
Finally, it is sufficient to indicate in what a pre-eminent degree the Diomede
exhibits that Kresilaian power of individualization which characterized the Amazon.
Of the facial forms, the thinness of the cheeks is especially to be noted, as well as the
unusually broad and powerful chin, the strong structure of the lower jaw, and the hard
prominence of its corners, which it had been customary to conceal under a fatty
stratum of skin. All these details express the rough obstinate force of the son of
Tydeus. The cropped hair and the sprouting whiskers x serve to define the powerful
youth, so do the ears swollen from the boxing-match (this detail undoubtedly
belonged to the original : the Munich and Dresden replicas exhibit it, and only the
indifferent Paris copy omits it. This allusion to skill in the roughest and most
dangerous of sports is as appropriate to our hero as it is to Herakles).2 The ear is
for the rest very carefully modelled : it has a long thin lobe, clearly separated
from the rest of the ear.
It was natural that so forcible a creation as the Diomede should make a strong
impression on its own and on later times. The earliest monument derived from it, the
Attic vase, has already come under our notice. A statue of Ares, referred to in
an earlier connexion (p. 94, Fig. 41), would appear, from the whole attitude and
bearing, the arrangement of the garment and the sword, to be also derived from the
Diomede. The reverse is certainly not the case ; for in the Diomede every detail has
its reason in the situation represented, while in the Ares we only get a beautiful
motive used without special significance — whence also the less emphatic turn of the
head. The dull and meaningless drapery of the Ares likewise denotes an artist
of the second order.
It will be seen presently that the Diomede seems to have stirred up emulation
even in the circle of Polykleitos.
To later, perhaps only to Roman, times pertains the transformation (preserved
in a charming bronze statuette found at Zurich) of the Diomede into a Hermes.3
The winged cap and shoes denote Hermes ; but the entire motive, down to the
garment on the left shoulder, is borrowed from the Diomede ; even the close-
cropped hair, the shape of skull, and the expression retain something of the
prototype. On the other hand, a larger, much mutilated bronze statuette in Berlin appears
to be an original of the good Hellenistic period.4 The Diomede forms the basis for a
portrait conceived heroically. The garment is omitted ; but the prototype can be
traced not only in the motive but also in the forms of the body, and — notwithstanding
the portrait-character and the different hair— even in the countenance, which has
the same angular jaw and slight whiskers.
Finally, the Diomede, like so many celebrated Greek statues, had to serve as
model for portraits of the Roman emperors. A good instance is a statue of Augustus
1 Cf. Brunn, Bayr. Sitzungsber. 1892, 654 seq.
2 As far as I know, the first dated representation of Herakles with swollen ears appears on the coins of
Euagoras I. (410— 374)— cf. Roscher's Lexikon, i. 2163, 5 ; yet there are many instances which may be earlier.
3 Mitth. d. Antiqu. Gesellsch. in Ziirich, Bd. xv. Taf. 5, 23, and xvii. 7, p. 133, No. 57 (Benndorf).
4 Antiquarium, Inv. 7419, height 0'20, from Asia Minor. The figure is cast hollow with thin walls ; square
patches are let in, as often occurs on larger Greek bronzes. Arms and legs were cast separately, and are now
missing. The surface is much rubbed.

156 KRESILAS
in the Vatican j1 the garment is there altered to suit the Roman fashion. Lastly, it
seems to have been present to the mind of the artist of the superb bust of Caracalla
in Berlin,2 which recalls it in bearing and expression, in chlamys and sword-belt,
and even wears the same whiskers.

VII. The Medusa Rondanini.
To the surprise doubtless of many, the Medusa Rondanini — that noble and
unique creation which once so inspired Goethe in Rome— does not belong, as has been
supposed, to later Greek art, but is inseparably linked to the series of works now under
discussion and associated with the name of Kresilas.
The Rondanini mask, now in Munich 3 (reproduced from the cast in Fig. 63),
is admirably preserved ; save for some insignificant bits of the snakes and of the
hair, only the extreme point of the nose and the edge of the left nostril are modern.
The mask is cut straight off at the back, and has been fastened in modern times
upon a square plaque. The existence of two replicas preserved in Rome, now in
the Museo Torlonia,4 witnesses to the celebrity of the work in antiquity. They have
no background ; in modern times they have merely been placed upon busts. Two
Medusa masks in the Vatican 5 and a colossal one in Cologne 6 are similarly cut off at
the back and unprovided with a background : they have however at the most only a
very distant dependence on the Rondanini mask, imitating it in externals but trans
lating it into the late half-effeminate and half-pathetic manner.
The circumstance that these masks were worked without a background suggests
that they were intended to be fastened by the purchaser against a wall — that is to
say, to hang simply against the wall of a house. They certainly could not, as has
been suggested,7 have stood in any definite relation to the architecture, for in that
case they would have been worked into the architectural member they were intended
to adorn, and it would be difficult to explain the existence of exact replicas.
The latter suggest rather a famous Greek original, copied for its own sake and
absolutely independent of varying architectural surroundings. This original must
have been cast in bronze, to judge from the network of snakes, so little suitable
to marble, and from the smoothness of the surface of the face. It will be felt,
too, how much more effective the work would be in bronze than in marble. The
best (No. 294) of the two Torlonia replicas is significant, because it renders the hair
generally without the deep undercutting of the Rondanini replica, and is rather more
severe and evidently more faithful in the details. Thus the little curling lock at the
side near the left eye is formed quite in the archaic manner with a tightly rolled end.
In other respects the faces show that the Torlonia replicas are both inferior works.
The general place of the Medusa in the history of art, notwithstanding the gross
error of judgment in which I had myself concurred,8 is unmistakable on a close
1 Gall, delle Statue, No. 262 ; Bernouilli, Rom. Ikonographie, ii. I, Taf. 3, p. 58. The head unbroken.
2 Skulpt. 384 ; Mitchell, Select, from Anc. Sculpt. PI. 20.
3 Glypt. No. 128; Brunn-Bruckmann, Denkm. No. 239; Brunn, Gdtterideale, p. 60.
4 Museo Torlonia, Taf. 74, Nos. 294, 296.
5 Pistolesi, Vatic. Descr. iv. 13 ; Helbig, Museums, 10. 6 Friederichs-Wolters, Gipsabg. 1598.
7 Brunn, loc. cit. , and Verhandl. d. Philologenversammlung in Dessau, p. 76 ; Gdtterideale, p. 60 ; Dilthey,
Annalid. Inst. 1871, p. 228; K. Botticher, Erkl. Verz. d. Abgiisse, No. 793.
8 In Roscher's Lexikon d. Myth. i. 1724. The coin of Seleukos (Gardner, Types, PI. 14, 6) compared
by J. Six, De Gorgone, p. 73, is quite different, and is probably at the most a distant Hellenistic derivative of the
Rondanini type. The hair is Lysippian.

1 58 KRESILAS
inspection of the replicas. The stylistic treatment of the eyes and their adjoining
parts (such as the sharp edges of the eyebrows), of the hair, and of the mouth is
never met with at all so late as the fourth century. The work is of the free style of
the fifth century — a result to which the development of the art type of Medusa does
not in the least run counter.
The Medusa head with beautiful, undistorted features appears on an Attic vase
as early as about the middle of the fifth century,1 not it is true in the form of a
mask or of a gorgoneion, but as a severed head with the neck, in the hand of Perseus.
A great artist must be at the back of this conception (cf. infra, p. 200). Following
on this come several gorgoneia with the severed neck ; they exhibit the same type of
pure beauty, and from their style they must be dated in the second half of the fifth
century.2 The endeavour was to dispense with all external tokens and to define the
Medusa by the expression alone, mostly by wide-opened eyes and bristling hair.
But now our artist invested the old traditional gorgoneion with beauty of
feature. In doing so he did not entirely give up the severe character of the mask:
he also retained the two snakes which encircle the head and twine into a knot
under the chin, and which had been a familiar feature in the distorted gorgoneia
of the type immediately preceding ; 3 and he adds to the head the further attribute of
two wings, similarly borrowed from more archaic art, though it was not common
there,4 and had still a certain air of innovation. For the rest he is in substantial
agreement with the other works of the epoch that represent the new ' beautiful ' type
of Medusa.5 He too gives her big wide-opened eyes and short bristling hair. Only,
to give full effect to the wings and snakes, he had to let the motive of the hair fall
more into the background, whereas other artists of the time lay chief stress upon it.
He too, like most of the others, lets the hair lie smooth and well arranged in the
centre over the forehead ; but he makes it stand out at the sides and frame the
countenance as far as the region of the ears ; the ears themselves he omits, as is usual
in the 'beautiful' type. On the other hand, the character he strives to give to the
mouth seems to be peculiar to himself. He throws into it the chief expression.
While the others either make the mouth beautiful though quite ordinary,6 or else
recall the older type in a mechanical and discordant way by introducing into the
face the projecting tip of the tongue,7 our artist has contrived, without impairing the
beauty, to retain something of the older conception, by making the mouth unwontedly
broad and with parted lips, showing the upper teeth. The discovery of the place
occupied by our Medusa in the development of the type affords a key to the right
understanding of the intention of the artist.
Above all, it exposes as false the ordinary notion that the Medusa is conceived
as dying. Goethe it was who first discovered in the mask ' the agonized stare of
1 Annali d. Inst. 1881, Tav. F. Style of the Orpheus vase (50th Berliner Winckelmannsprogr. 1890, Taf. ii.)
For the date see ibid. p. 162.
2 Thus the terra-cotta relief, Arch. Anz. 189 1, p. 122, Fig. 17 a; the Panofka tile, Terrak. v. Berlin, Taf. 62,
I ; bronze masks in Berlin Antiquarium, No. 7484.
3 Frequent on the gorgoneia of the ' middle ' type ; so on the shield of the Parthenos (British Museum
copy) and on the shield of the Athena of the Nike balustrade ; on the aegis of the Albani Athena (Fig. 29, p.
79) and of the Munich Athena (Brunn, Glypt. No. 86).
4 Cf. Roscher's Lsxikon, i. 1722 se/., where several certain examples of the older fifth-century type are
given. 5 See Roscher's Lexikon, i. 1721 seq. For the gem with Solon's name cf. Arch. Jahrb. 1888, p. 310 ;
the monuments which in Roscher I placed in the beginning of the fourth century should probably be dated in
the fifth. For important new evidence cf. the terra-cotta in Arch. Anz. 1891, p. 122, Fig. 17 a.
6 E-g- the bronze attachments for the handles of pails, Berlin Antiquarium, Inv. 7484, and Roscher, 1722.
Cf. the terra-cotta mask, Arch. Anz. 1891, p. 122, Fig. 17 a.

MEDUSA RONDANINI 159
death.' x He was followed by many others— e.g. Friederichs,2 O. Jahn,3 Dilthey,4 and
Kekule,5 who expanded on this theme. The representation of a dying, defeated
woman stiffening in death is, however, far removed from the intention of the whole series
of works with which the Rondanini mask must historically be ranked. What these
works really do is to substitute for the wild brutality of the older type 6 a tranquil
human conception ; for the grim look of fury, a fixed wide-opened eye that
fascinates the beholder by its daemoniac power — and this effect is heightened by the
bristling hair. By the powerful chin and the broad open mouth our artist has
depicted yet more intimately the wild force and constraining strength of the daemon,7
and at the same time has suggested the powerful voice which tradition attributes to the
gorgons. Above all, he alone has known how to invest the whole expression with a
freezing horror.
It was not till a much later date — till the epoch after Alexander — that the type
was created of the agonizing gorgon, looking out with blank despairing gaze.8 Yet
not even at this time is she ever represented as exhausted or dying, and those
gorgoneia still preponderate which depict the force, the wrath, and the wild anger of
the grim daemon, though in the pathetic and realistic fashion of the time.
This historical survey exposes what was incorrect in Brunn's 9 interpretation of
the Rondanini mask. Brunn did not, certainly, fall into the error of seeing in it a
dying creature : he describes the. general expression admirably as a ' cold stare,' but he
attributes this fixity to an ' architectonic petrcfaction of the form,' the mask having,
according to him, served an architectonic purpose. The objections to this view have
already been raised. But the so-called tectonic character of the mask, its severely
symmetrical composition within an outline of almost geometrical simplicity — an
inverted triangle — is nothing more than one of the proofs for the relatively early
date of the work. For this character is common to all earlier gorgoneia without
exception. It is most strongly marked in the earliest period, and is afterwards
gradually modified,10 till it completely disappears in the Hellenistic epoch, when the
severe full view is also commonly given up.
It is true that our mask is distinguished from among works of the same time and
epoch by its severe lineal structure : this structure is however not selected for any
external purpose, but only as lending itself to the mental expression. The mask
has only to be compared with others to appreciate how considerably the severe
structure contributes to the daemoniacal expression : even as the serried ranks
of an army produce more effect than the same troops in loose array — so it is
with art-forms.
1 Schriften der Gothegesellsch. Bd. ii. Tageb. iiber Briefe Gothe's aus Italien, p. 240 = Lat. Reise, Rome,
25 Dec. 1786. Cf. April 1788, where he notes 'the discord between death and life, pain and delight
(' Zwiespalt zwischen Tod und Leben, zwischen Schmerz und Wollust ').
2 Bausteine,No. 672 ( = Fried. -Wolters, Gipsabg. 1597). 'At the moment of turning to stone ' ('im
Moment des Erstarrens').
3 Aus der Alterthumswissenschaft. p. 278 ('im Tode erstarrend ' ; 'eine lahmende Kalte . losch
den letzten Lebensfunken aus'). 4 Annali d. Inst. 1871, p. 220 seq.
5 Entstehung der Gdtterideale, p. 25 seq.: 'unterliegt in Trotz und Schmerz.'
6 Dilthey, Annali, 1871, 220, takes for granted that the archaic gorgon was represented as dying. It is
however easy to prove that the old type came into existence without any thought of death. On the contrary, the
type was, although inappropriate, adopted for the scene in which Perseus kills the gorgon. Cf. Roscher's Lexikon, i.
1 701 seq.
7 Meyer noticed (on Winckelman's Geschichte d. Kunst, v. 2, § 20) that the forms incline to the wild and
terrible ('zum Wilden und Schreckenden '.). s cf_ Roscher, loc. cit. 1724.
9 Glypt. 5th ed. p. 164 seq. ; Verhandlungen d. Philologenvers. in Dessau, p. 76 ; Gdtterideale, p. 59 seq.
10 Cf, Roscher's Lsxikon, i. 17 19, 15 seq. In the older period the circle is the fundamental form,

160 KRESILAS
The mask terminates above in a broad horizontal, which exerts a gloomy
massive effect upon the whole. Thence the lines converge downwards in the
manner of an equilateral triangle ; at their point of intersection the terrible ex
pression of the whole culminates in the half-opened mouth with the snake-knot
beneath. But above the wings broaden levelly out — surely and inevitably, as
when a bird takes its quiet flight through the air,1 does the spectator feel
the ghostly head approach, while beneath the shadow of the wings hiss the
snakes.2 We have endeavoured to explain the Rondanini mask by the period and the
subject ; but complete comprehension is only possible with the recognition of the
personal note, of that quality in the work which belongs, not to the nature of Medusa,
but to the personality of the artist who created her.
After all that has been said, no detailed proof is needed to show that this
personality is the one with which we have been occupied. The broad likeness—
the likeness of a child to its parents — strikes the eye at once. Among the single
features, the eyes again first claim attention. Although the Medusa type required the
eyes to be wide open, the artist has given them the heavy thick lids noted in all his
heads, nor is the little fold on the lower lid omitted, while the strongly marked lachrymal
gland also belongs to his system of forms. In the centre of the forehead the smooth
interciliary region is again indicated ; to the sides the lower part of the forehead
is markedly prominent, but without being separated off by depressions, as in the
Diomede, from the sharp edge of brow — a difference based on the character of the
heads : in the one everything is concentrated towards the centre, and in the other a
massive breadth dominates the whole design. A further point of agreement is the way
in which the nose is formed, with a narrow rounded bridge rising gently in the
middle and very lifelike nostrils, although its lower part is essentially different, being
made broader to lead on to the unusually broad mouth, while the nostrils of the
gruesome creature are strongly inflated. And again it is indisputable that the
drawing of the mouth accords with the works already examined, and in particular
with the Diomede. Finally, the hair about the forehead closely resembles that of
the Athena in arrangement and treatment.
We find, then, in the Medusa all the characteristic details of the work of Kresilas,
and, more than this, we find that general type so difficult to define in words. It is,
finally, a fresh witness of the artist's power of entering into the heart of his subject,
and of making the outer form expressive of the inner qualities. It must, of course,
be reckoned among his later works.
Being a work of the fifth century, the mask could not have served a mere
decorative purpose ; it must have been set up in some public building and have
had a religious significance ; something like the ' golden ' (probably brazen)
gorgoneion on the outer side of the south wall of the Akropolis (Paus. i. 21, 3);
or like that older stone one at Argos, ascribed to the Kyklopes (Paus. ii. 20, 7).
Precisely Argos, the home of the legend of Perseus and Medusa, might well have
given this commission, and our Medusa may be thought of as a possible offering
in the sanctuary of Demeter Chthonia at Hermione, for which an inscription
1 The wings are not sunk as Brunn ( Glypt. ) and Kekule (Entst. d. Gdtterideale, p. 26) assert. In the Verh. d.
Philologenvers. zu Dessau, p. 76 (Gdtterideale, p. 59), Brunn himself says that the wings are raised in a
threatening manner. I only know one instance of a wearily sunk pair of wings, i.e. a Medusa head on a gem of
doubtful authenticity (from Coll. Blacas, now in Brit. Mus. Catal. 1253 ; King, Anc, Gems, ii. PI. 20, 6 ;
cast in Cades, CI. ii. F, 56).
2 The snake-heads are restored, but evidently right on the whole,

THE PETWORTH ATHLETE.

101

shows Kresilas to have worked. Since the Topyetr] Kecpa\rj, according to ancient
conception,1 had a ghostly existence in the underworld and was subject to Per
sephone, the Medusa mask might well be set up in a sanctuary of the goddess
of the lower world. It has been seen that the mask was intended to be fastened
to any wall. Pausanias (i. 2, 5), in the description of a Temenos of Dionysos at
Athens, reports that the mask of the Dionysiac daemon Akratos was attached to
the wall of the sanctuary; and, according to Pliny (xxxvi. § 13), there was a mask
of Artemis, an archaic work of Boupalos and Athenis, on the wall of her temple
in Chios. The creation of Medusa also influenced succeeding art, were it only by its
external scheme, which was repeated with more or less similarity in so many later
gorgoneia ; as to its spiritual character, that could only be copied, never imitated.

VIII. Statue of an Athlete at Petworth.
As last and latest link in the personal growth we have been studying, there
remains to be added the head of a youth adorned with the victor's chaplet, which
is preserved in four copies. The finest is at Petworth, in the collection of Lord
Leconfield ; 2 it is reproduced in Fig. 65 from a photograph taken from the
original by permission of the owner ; Fig. 64, from an old cast in Dresden be
longing to the Mengs collection, shows the profile. Nothing is restored in the head
except the tip of the nose. The neck is almost entirely preserved, but it is cut sharply
off and set on to a coarse modern nude bust. The head is probably identical with one
published by Count Caylus in 1736.3 This youth with the curling hair has wound
about his head the victor's chaplet, taenia or mitra.4 The ears, which are not swollen,
and a certain refinement and gentleness in the whole form, would indicate that he is
no hero of the glove and the pankration, but has conquered in a different way —
either in a running match or in the combats of the Pentathlon, where skill rather than
brute force was required. The chaplet is not twisted in a knot at the back, but wound
round the head, with the ends tucked in and pushed through the fillet on either side
above the temples. This method of fastening, though very practical and doubtless
often employed in real life, is yet nowhere else represented in plastic art. The artist
1 Nekyia of the Odyssey, xi. 634 ; Aristoph. Frogs, 475. Cf. Roscher's Lexikon, i. 1703 ; Max. Mayer in
Jahrb. d. Inst. 1892, p. 201.
2 No. 24, Michaelis, Anc. Marb. in Gr. Brit. p. 609 ; Specim. of Anc. Sculpt, i. 30. The marble is fine
in grain, the nude parts polished.
3 Caylus, Rec. d'Ant. ii. PI. 48, 2 ; p. 142. The engraving (reversed) certainly gives a head of this type.
According to Caylus, the neck was cut off below and fitted on to a Roman draped bust which he severely
criticizes. He says further : ' Ce buste ilait dans le cabinet de M. le chancelier de Pontchartrain.' Probably it
had been recently sold and sent to England. The Petworth collection was being formed at the time Caylus wrote
(1750— 1760; cf. Michaelis, loc. cit.) For the new possessor the Roman bust may have been replaced by a
nude Greek one. The other replicas are : (a) a poor copy, lately at the art dealer Abbati in Rome— see Bull. d.
Inst. 1867, 33 (Helbig) ; Mon. d. Inst. ix. 36 ; Annali, 1871, 279 (Conze) ; Brunn-Bruckmann, Denkm. No. 84
(from the cast), (b) A fragment of the right half of the head, found at Treves— see Hettner, Die Rdm.
Steiiulenkmdler des Provinz. Mus. zu Trier, No. 695. According to Loschcke, apud Hettner, loc. cit., the head
comes rom a relief. Loschcke informs me that the relief belonged to the incrustation of the Thermae, where
famous athlete statues had been copied in relief on the scale of the originals, (c) In the Palazzo Riccardi in
Florence, left of the doorway leading to the staircase — Dtitschke, ii. 182.
1 The woollen fillet which was given to the victor in addition to the wreath was called p-irpa in the earlier
period (cf. Pind. 01. 9, 84 ; Isth. 4, 62 ; Bdckh, Expl. p. 193) ; as was also the similar fillet used in symposia
and worn by Dionysos (cf. Samm. Sabouroff, Taf. 23), Y

1 62

MYRON

here evidently desired to vary for once the ordinary arrangement. The locks of hair
fall so freely over the chaplet that at the back they almost conceal it. On the crown
of the head and also on the left side
above the chaplet is a rectangular
broken surface, which proves the
right arm to have rested on the
head with the hand hanging over
on the left side.1 The head is
slightly inclined to the left. The
expression is one of complete re
pose, to which the motive of the
arm on the head, as in the famous
statues of Apollo and of Dionysos,
would further contribute.
In an athlete, it is true, this
attitude is open to another interpre
tation. It occurs in representations
of athletes cleaning themselves
with the strigil : the youth places
one hand above his head in order
to scrape himself under the arm
with the other, as for example in
a statuesque figure from a fifteenth-
century Attic grave-relief.2 But
the tranquillity of this head, and
above all the festive wreath of the
victor, exclude the notion of a
youth scraping or anointing him
self, and still more of course the
idea of a youth exercising. The
in repose, in which case the left
The leaning attitude, adopted for

Fig. 64. — Profile of athlete. (Petworth Coll.)

victor must therefore have been represented
arm would also require a certain support.
the gentler divinities, would of course be quite out of keeping for an athlete.
It might therefore be suggested that, resting firmly on the right leg, he supported
himself lightly with the left hand upon an athletic weapon, the akontion or
short spear, something in the manner of the young athlete on a Spartan relief.3
This would denote him a pentathlete or conqueror in the five combats, as already
hinted. In the relief the athlete is further characterized as a pentathlete by the
1 This broken surface occurs both on the Petworth and Abbati copies. Conze was the first to interpret it
correctly. 2 Friederichs- Wolters, Gibsabg. 1017 ; for the date cf. Samm. Sabouroff, i. Introd. p. 41, note 9. Here the left
hand is laid on the head ; cf. Annali, 1S62, Tav. M. Note also the copy of a statue on a wall in Pompeii>w«.
Mitth. 1888, p. 200, fig. 2, where the right arm lies on the head; the action is Lysippian in character' The
motive of the Skopasian head of an athlete in Athens (Annali d. Inst. 1876, Tav. G ; Friederichs- Wolters
Gipsabg. 1300) is not quite clear ; the right hand rests on the head, and the left, which probably held the strigil, is
close to the head. In the head we are now discussing this cannot be the motive, as in that case the rough surface
on the top of the head would have to be much larger. Cf. a relief on a marble seat in Turin (Dtitschke iv 311) •
between Ionic pillars is a figure (evidently in imitation of a statue) of a youth holding his right arm over his head
and slinging round his neck a sword, the belt of which he holds in his right hand. (Dtitschke took the sword for
a bow. )
3 Arch. Zeitg. 1883, Taf. 13, 2 ; p. 228 (Milchhdfer).

THE PETWORTH ATHLETE 163
addition of the springing weights in the right hand, as the spear alone might in this
case be open to misinterpretation. Of this there would be no danger in a statue
with the victor's chaplet, and bearing besides its appropriate inscription. An athlete
statue thus reconstructed would be briefly described in the late Greek art jargon

Fig. 65 —Head of an athlete. (In the collection of Lord Leconfield, at Petworth.)
as a ' doryphoros,' like the famous athlete of Polykleitos, who also carried the
short spear, the akontion and not the doru, in his left hand.1
1 This is evident from the careful copy on the Berlin gem, Tdlken, Kl. iv. 249. It belongs to the same class
of stones as the Natter Amazon gem. In Ath. Mitth. iii. p. 292, note 2, I laid stress on the incorrectness of
calling an athlete a Doryphoros, but I now think that the interpretation of the Doryphon of Pliny as statues of
victors is too probable on other grounds to be invalidated by these considerations. Cf. infra on Polykleitos, p. 228.
—The length of the athlete's casting spear was, as we know from the Spaitan relief and many vase-paintings,
about the height of a youth,

164 MYRON
Now, although Pliny x mentions a Doryphoros among the celebrated works of
Kresilas, it would be presumptuous to ascribe this work to Kresilas solely on
this ground ; it is the style which definitely points to his authorship. This style
has been generally described as fourth-century Attic.2 But the head is certainly
not of the usual Attic type — its fellow would be looked for in vain on the numerous
Attic reliefs — and, more than that, it has all the marks not of fourth- but of fifth-
century work. This may be recognized in the treatment of the hair alone, with
its separate tight curls and the arrangement over the ears, and is seen still more
decisively in the eyes and adjoining parts. Comparison with the fourth-century Attic
type of youth, so admirably shown on grave-reliefs, or with the Skopasian athlete-head
in Athens, who also rests his hand upon his head, brings home the complete
contrast of epoch more forcibly than words can describe it. Everything is different,
but the most readily appreciable difference lies in the stylistic treatment of hair
and eyes.The familiar indications are easy to recognize : the eyes are long and heavy-
lidded,3 with strongly marked lachrymal glands; the two vertical depressions4 start
upwards from the angle formed by eyebrows and nose ; the root of the nose has the
formation more especially pointed out in the Diomede and the Medusa, but its
modulations are more delicate ; the nose has the narrow rounded bridge with the rise
in the centre like the Diomede ; the modelling of the forehead is almost identical with
that of the Diomede, but somewhat flatter and daintier. We recognize Kresilas again
in the lower part of the face with its rich modelling, though he makes it softer and
more refined, and also a little shorter in proportion to the nose, than he had done
hitherto. All these forms are but the direct continuation of what was observed in the
Amazon. Since however in the Riccardi replica {supra, p. 161, note 3 {c)) the cheeks
and the parts about the mouth'present a much simpler, harder, and severer appearance,
owing to the absence of the detailed modelling of the flesh given in the Petworth head,
and the bony structure is more prominent, it is just possible that the Petworth head,
like the Munich copy of the Diomede, represents a slight intensification of the
original. The ear with the long narrow lobe is similar to that of the Diomede.
Finally, the hair again envelops the skull in plastic abundance;5 its main motive
consists in the little tight curls with twisted ends already studied in the Perikles. In
some places, as for example in the chaplet above the left temple, may be recognized
the old severe primitive form ; but the hair as a whole has become more mobile, freer
and more elegant, and its tangled irregularity is yet more natural. In the centre
above the forehead the hair is slightly parted ; but this only serves to accentuate
its capricious character, for it falls quite unsymmetrically over the chaplet, and the
1 Cf. p. 115. Only one other Doryphoros is named, that of Aristodemos, probably a Peloponnesian artist
who stood in close relation to Lysippos, and who seems to have enjoyed a general reputation for his statues of
athletes (Overbeck, S. Q. 1605).
2 Michaelis, loc. cit. : ' Attic ... no doubt of the fourth century.' He recalls the Diadumenos which Kalli-
stratos describes and assigns to Praxiteles ; but, apart from the fact that this Zmppairis only shows a general
acquaintance with the usual Diadumenos motive, and is therefore worthless (cf. Jahrb. d. Ver. d. Alt.-Fr. im
Rheinl. vol. xc. p. 65 seq.), the motive of the youth has nothing to do with the Diadumenos.— Conze, loc. cit.,
fixed the time more exactly about 400—350 B.C., and Helbig (Bull. 1867, 33) recognized a stage preparatory to
the later Attic school. Only Brunn, apud Julius, Annali, 1875, p. 31, dates the head correctly in the fifth
century, since he compares it with the sculptures of the Parthenon.
3 The length of eye-slit is the same as in the Perikles head (36 mm. with tear-gland, circa 31 without, height
1 1 mm.) The mouth is only i| eye-length without tear-gland (46 mm. )
4 These are distinct in all the replicas.
6 Although the Riccardi copy only indicates the hair, it well reproduces this plastic character,

KRESILAS AND MYRON 165
chaplet itself diverges from strict symmetry in allowing the two strips to overlap and
the ends to show on the one side above, on the other below the band. Evidently
the artist, like his contemporary the older Praxiteles, had fallen under the spell of
asymmetria.

IX. Relation of Kresilas to Myron. — The Riccardi Head. — The Diskobolos and kindred
Heads. — Pythagoras of Rhegium. — Myronian Portrait-keads.
Among contemporary works, those which have just been associated with the
name of Kresilas form a conspicuous group ; and this would be yet more obvious had
we the bronze originals instead of marble copies, which are all more or less indifferent
and inaccurate. For all the works that have been quoted seem to be without
exception copied from bronzes, and it is only among the bronze workers that Pliny
names Kresilas.
The development of the formal elements in our artist corresponds in general to
that observed everywhere in the second half of the fifth century. Yet the style of
Kresilas has met with no proper continuation, only with an occasional imitation in
late antiquity of certain details.1 While the personal style of Pheidias carried all
before it to the farthest corners of the Greek world, the art of Kresilas remained
an individual art, confined within narrow limits. It does however point backwards ;
the style of those works attributed to Kresilas hangs on by a thousand threads to
earlier manifestations, and it is at once an attractive and fruitful task to examine the
soil in which the individuality of the Kresilaian works had its rise.
The inquiry must start from a work that is unmistakably to be recognized as the
forerunner of the Diomede. Only head and breast are preserved, in a copy in the
Riccardi Palace at Florence, which is of bust-form (Fig. 66).2 Like the Diomede, this
youthful hero wears a garment on the left shoulder, falling in perfectly simple folds.
He also has the close-cropped curling hair and the swollen ears, though, as the
garment proves, he is no athlete, but, again like the Diomede, a hero noted for his
1 A signal instance of this kind of imitation is afforded by the famous Pourtales Apollo and the Apollo
from the Baths of Caracalla, now both in the Brit. Mus. (Overb. Apollo, p. 141, Nos. 5, 6 ; cf. Brunn, Gdtterideale,
p. 84 seq. ) From the similarity in their proportions and main features, there can be no doubt that the two heads
are merely different versions of one and the same original, while from the qualities common to both it is evident
that this original was not materially earlier than Alexander. Now the copy from the Baths of Caracalla displays
exclusively the forms proper to that period : the eye is deeply recessed and exceedingly pathetic (the god is
supposed to be sunk in musical inspiration), the hair aims dexterously at the most realistic treatment. In a word,
there reigns complete harmony between the conception and the stylistic forms. The fidelity of this copy to the
lost original is attested by the existence of an exact replica in the Palazzo Giustiniani (Ov. Apollo, p. 142, No. 7 ;
until the statue on which this head is, is properly cleaned; it is impossible to tell whether they belong together or
not). In the Pourtales head, on the contrary, we note an irreconcilable contradiction between style and
conception : an artist enamoured with the style of Kresilas has evidently attempted to introduce the formal
qualities of that master into a head of totally different style. Accordingly, the deeply recessed and pathetic eyes
have been transformed into Kresilaian eyes with strong prominent lids and overshadowed by sharp angular brows ;
in the hair conventional little curls with twisted ends replace the naturalistic fall of the loop of hair over the
forehead, and the loop itself hangs more over to the front , further, the mouth is sharply outlined, and the brow
has touches that recall the Amazon.
2 Dtitschke, Zerstr. Bildw. in Florenz; Heydemann, Mitth. aus Ober- u. Mittelitaliens , Taf. 6, p. 101 ;
Friederichs-Wolters, Gipsabg. 458 ; Brunn-Bruckmann, Denkm. 361. The adaptation as bust is ancient ; right side
of the breast with the sword-belt is new. The original is now in one of the rooms of the palace ; its place in the
cortile has been taken by a cast. There are two replicas of the head: (a) in the Pal. Conserv. (No. 5) in Rome —
poor, restored, and worked over ; (b) in Berlin, Skulpt. 472 (Fig. 67 on p. 167) — better, though superficial copy ;
modelling of the forehead good ; the whole of the back of the head is restored. In addition to these two, Arndt
thinks that a bust in the Loggia Scoperta of the Vatican which lias been turned into a Hermes is a further third
replica.

Fig. 66.— Head of a hero (Palazzo Riccardi, Florence)

THE HERO RICCARDI

167

strength. There is a further resemblance in the broad, forcible chin, in the slightly
opened mouth, the heavily lidded eyes, in the modelling of the forehead and the
manner in which it is framed by the hair. Yet all these forms are essentially harder
and more archaic : the hair, for instance, still lies upon the skull like a heavy cap,

Fig. 67. — Replica ot the Riccardi head (Berlin).

instead of growing from it ; no depression as yet indicates the actual crown, and the
separate masses, unnatural in their extreme smallness, are only as it were carved on
the surface, almost without relief ; it is only in the mass that the hair has any plastic
effect. Moreover, the skull is higher at the crown, and its outline is rounder — not so
angular as in the Diomede. Finally, the hero's attitude and bearing was simple and

1 68 MYRON
constrained, as compared with that of the Diomede, which vibrates with energy. He
stood quietly, with his upper arm lowered, and his head turned slightly to the right ;
the conception lacks that expression of conscious energy and mental strain which
animates the Diomede. Kresilas must have known the original of this work, certainly
twenty to thirty years older than his, and must have stood in close relation to a
master whose creation he was further to develop. Now by the help of the extant
copies of the Myronian Diskobolos it becomes possible to determine with comparative
certainty who this older master was : it was Myron.1
The head of the Diskobolos of Myron is known to us, not only from the Massimi
statue,2 in which it is attached to the body, but from at least three separate copies.
The first of these three heads is the one that has passed from Steinhauser's possession
to the Museum of Bale ; 3 it has been considerably restored ; the second head, in the
collection at Catajo, is intact save for the front part of the nose, which is modern
(Fig. 68) ; 4 the third head, which is at Berlin (Fig. 69),5 though much worked over
and restored, is yet of considerable value. A comparison of these four replicas of
the head shows that the copyists allowed themselves great freedom in the execution
of detail, and especially so in the case of the hair. It is evident that in this respect
the Berlin head presents by far the closest and most faithful copy of the original :
the elaborate detail of the hair is obviously intended to imitate the chiselled bronze
locks of the severe period, and cannot possibly be explained as an invention of the
copyist. On the other hand, the greater freedom and lightness of treatment in the
case of the other heads is doubtless to be ascribed to carelessness or neglect on the
part of the copyist. The case is identical with that of the two replicas of the Riccardi
type mentioned above. The copyist of the Catajo head has treated the original of
Myron with the greatest freedom, keeping only to the general character of the close-
cropped hair, and indicating the detail boldly and openly according to his own ideas.
He has even chosen to modify the characteristic wavy dent, made by the outline of
the hair over the forehead. The Massimi copyist was more careful, though he too has
for the most part dispensed with the delicate little fringed curls which encircle the
head, or has at any rate simplified them to a great extent. It is only about the
forehead that he has represented them at all elaborately ; and even here he is content
to suggest them without working them out. Behind the ear and on the neck he has
omitted them altogether, and substituted simpler motives. In this one respect the
Steinhauser head, which reproduces the little tight curls behind the ear and in the
neck, is almost as faithful a copy as the Berlin head. With the help of the latter
it becomes possible to realize what care Myron must have bestowed on the hair of
his Diskobolos. On the upper part of the head the individual locks only slightly
turn up at the ends, while on neck and brow they form a fringe of tightly twisted
curls. In the Riccardi bust the hair is treated in precisely the same manner, and
produces the same singular effect of a cap fitting close to the head — similarities that
lead one to infer that Myron was the artist of both works. In the case of the
1 Wolters (Gipsabg. 458) was the first to recognize this, though he certainly went too far in speaking of
' vollstandiger Uebereinstimmung,' and in considering the Riccardi head to be almost u. replica of the
Diskobolos. 2 Reproduced from the old well-known photograph, Collignon, Hist, dc la Sculp. Gr. i. PI. xi. ; Brunn-
Bruckmann, Denkmdler. An excellent small plaster reduction of the original can be purchased in Rome. The
Munich Bronze (F. -W. 453) is late Roman ; only the motive — not the head — derives from Myron's statue.
3 The head was first recognized by Helbig, Bull. d. Inst. 1870, 12. Cf. Kalkmann, Prop, des Gesichls,
P- 74- 4 Diitschke, No. 699; Arndt-Biuckmann, Einzelverkauf, Nos, 54, 55. 5 Beschi. d. Ant. Skulpt. 474.

THE DISKOBOLOS OF MYRON

169

Riccardi head, however, the hair is richer and more plastic in its effect ; the curls of
the fringe are somewhat larger ; and the line of hair upon the forehead falls quite
simply and naturally without forming the dent noted in the Diskobolos.
In the shape of the face there is considerable agreement among the different
replicas of the Diskobolos, and if we compare them further with the Riccardi head the
general resemblance of the two types is very striking. It lies especially in the formation
and modelling of the forehead, in the straight, sharply defined, and projecting eyebrows,

Fig. 68.— Head of Diskobolos (Catajo).

in the spring of the nose and its naturalistic formation, and in the full lines of the lips.
Such points of difference as are worth noticing rest chiefly on the obvious endeavour
to give a different character to the two personages represented. The pentathlete who
is hurling the discus is of comparatively slight build, and consequently the lower part
of his face is less powerful and more rounded than that of the confident hero, with
the swollen ears, pourtrayed in the Riccardi bust. Other differences show, as do those
noted in the treatment of the hair, that the hero is a somewhat later work by the
artist of the Diskobolos ; for instance, whereas the lips of the Diskobolos, though not
tightly set, are closed, in the Riccardi head the lips are parted ; further, the expression z

170

MYRON

of the Diskobolos is still constrained, while that of the boxer is more open. Again,
the ears in the Riccardi head are placed somewhat lower ; the skull is somewhat less
high, its upper line inclining more to the horizontal, and the head when looked at
from above appears a trifle longer and narrower than that of the Diskobolos. These
differences are after all insignificant, and in the main the two heads are very similar.
Finally, we must notice the eyes. Even the Diskobolos has somewhat heavy thick lids,
that lie in a natural manner on the eyeball ; still, the Riccardi head shows considerable
advance in the attempt to give substance and a natural look to the upper lid.
If the Riccardi head is a copy of an original by Myron, executed at a somewhat
later date than the Diskobolos, it follows that Myron must have been the master who
influenced Kresilas in his Diomede. And it further becomes evident that Kresilas,

Fig. 69.— Head of Diskobolos (Berlin).

both in the representation of detail, such as the eye and hair, and in his attempts at
expressing individuality, merely developed what Myron had begun.
It has, however, been lately urged by P. Herrmann that the Riccardi head
differs from the Myronian type, and is to be classed with the head of a youth from
Perinthos, now at Dresden (Fig. 70).1 The only point I can admit in this contention
is that the Perinthos head can have nothing to do with Myron. P. Herrmann quite
correctly points out that it differs from that of the Diskobolos in having a shorter
crown, which when seen from above appears rounder, but he should have also noticed
that the Riccardi head has a still longer crown than the Diskobolos, and, further,
that the two last works are linked together, as we saw, by a thoroughly
characteristic treatment of the hair, while in the Dresden head the locks of
1 Ath, Mitth. 1 89 1, p. 313.577. PI. IV. Y,

PYTHAGORAS

1/1

hair are curled all over the head with considerable uniformity, in a manner
differing totally from the Myronian, and recalling on the whole the hair of the Lapith,
who is being bitten by a Centaur, from the west pediment of Olympia (a comparison
made by Herrmann himself).
Another still more essential difference between the Riccardi and Dresden heads
consists in the formation of the eyes. Herrmann rightly criticizes the prominent
angular lids of the Dresden head, which produce a wholly unnatural effect, ¦ as though
they possessed no power of move
ment.' This is precisely the manner
characteristic of the art of Kritios
and Nesiotes, and of the sculptors
of the Olympia pediments ; and one
of the most precious results obtained
from studying the extant copies of
the Diskobolos is the knowledge
that Myron treated the eyes in quite
a different way. The lids of the
Myronian statues lie naturally on
the eyeball, and their edges are
neither prominent nor broad nor
angular, as in the Dresden head. In
this respect the Riccardi head marks
a further advance upon the Disko
bolos. We find a similar treatment
of the eyes to the Myronian in the
works ascribed above (p. 54) to
Hegias, the master of Pheidias, and
to Kalamis (p. 81). Finally, the
lifeless ' perfectly horizontal line' of
the mouth of the Dresden head
offers a complete contrast to the life
like modelling of the mouth of the
Riccardi bust, which in this point
too is so clearly in Myron's style as
we know it from the Diskobolos.
The Perinthos head belonged to
the statue of a victorious pankratiast,
and there are a few heads of athletes which are really allied to it : first among these
is the boxer in the Louvre j1 next, and in a somewhat later manner, an athlete of the
Giardino Boboli, formerly erroneously taken for Harmodios f and, lastly, a statue of an
athlete in Lansdowne House.3 The expression, as well as the round, rather short
crown, the hair, the flat eyes, and the wide horizontal opening of the mouth common
to all these heads, seem to me to link them together indissolubly, and to mark them

Fig. 70. — Head from Terinthos (Dresden).

1 Mon. d. Inst. *. 2 ; Annali, 1874, Tav. I ; phot. Giraudon, 1207. Cast of the head in the Ecole des
Beaux-Arts and in Bonn.
2 Mon. d. Inst. viii. 46 ; Arndt-Bruckmann, Einzelvcrk. 96 — 98 ; Dtitschke, Zerstr. Bildw. in F/oreuz, No.
77. Cf. Berlin. Philol. Wochenschrift, 1888, 1448. Separate casts of the head and of ihe torso in the Ecole
des Beaux-Arts, at Paris, No. 2844.
3 Michaelis, p. 446, 36. The head belongs to the statue, and resembles that of the Louvre boxer, especially
in the eyes and forehead, though the statue itself is later in style.

172

MYRON

off sharply from the type created by Myron. The statues in the Louvre and the
Giardino Boboli 1 apparently belong to the heads set upon them : they represent the
athletes in violent movement. The artist, who evidently was a man of note, since
his statues were afterwards copied, must have been a contemporary of Myron,
inasmuch as he vied with him in representing athletes in violent exertion ; but his
system of forms has more in common with Kritios and the sculptures of Olympia
than with Hegias and Myron. There is no artist to whom all the characteristics
noted would apply so well as to Pythagoras ; the Perinthos head possibly gives us a
copy of one of his earlier athletes,
produced about B.C. 480.2 Thus
we should at last have obtained
approximately reliable material
for recovering the style of this
remarkable master, of whose
celebrated athletes some copies
must have survived. In the whole
range of Greek sculpture, how
ever, the series of statues just
discussed seem to me the only
ones that can be brought into
connexion with him.3
It follows that the well-
known head from the I nee Blun-
dell collection (Fig. 71) must be
classed, not with the Perinthos
head as Herrmann had it, but
with the Riccardi head, and is
therefore also Myronian.4 The
hair agrees minutely with that
of the Riccardi bust, even the
little tapering locks about the
forehead being practically iden
tical in both. The eyelids are
also thoroughly Myronian, and
differ totally from those of the Dresden head ; indeed, they already have some
thing of the manner of Kresilas. Further, the formation of the part between
upper lid and brow deserves close attention, as it marks an advance upon the
1 The neck in both is a modern restoration ; yet the marble and the workmanship of head and body seemed
to me, after repeated examination, to correspond exactly in character and movement. Moreover, on the left-hand
side of the upper part of the head of the Louvre athlete is a great square puntello which indicates that the left
arm was raised above the head ; this same motive is displayed by the torso. The Boboli statue too is naturally
that of an athlete in active movement ; the left arm is parrying, the right striking out.
2 Herrmann and Arndt also thought of Pythagoras in connexion with the Perinthos head (Ath. Mitth.
¦891, 333)- Concerning the athletes of Pythagoras and the probability of their attitude being one of movement,
cf. Reisch, JVeihgeschenke, p. 44.
3 Further, a much-mutilated statue of a youth in Olympia (Arch. Zeit. 1880, 51, erroneously described as
Apollo— solely because of the headdress) may, I think, be regarded as a Roman copy after Pythagoras. It
belongs to the close of the severe style ; the upper part of the body is slightly twisted, the head inclined towards
the side of the free leg ; the drapery is falling from the left arm. In close relation to it comes a torso of a youth
in the Lateran (No. 52).
4 Arch. Ztg. 1874, Plate 3. Kekule was the first to pronounce the head Myronian. Cf. Friederichs-Wollers,
Gipsabg. 459.

Fig. 71. — Head in the collection at Ince Blundell Hall (Lancashire).

MYRON'S TREATMENT OF HAIR 173
Diskobolos, and is far more realistic than in the Dresden head. Ouite different
from the latter also is the full sensuous mouth. Yet just as the Riccardi head,
although it recalls the manner of Myron, differs totally from the Diskobolos in
expression and character, so does the Ince Blundell head differ in certain particulars
from both. The features are more closely packed, the modelling throughout is
rounder, and the cheeks fuller.1 The shape of the head resembles the Riccardi head,
though when seen from above it presents a longer, narrower ellipse. On the forehead
the transitions between the different planes are effected with peculiar softness and
delicacy. The full mouth indicates a powerful and sensuous nature, but the expression
of the face is sombre and gloomy. This is accounted for by the swollen ears : the
man is a boxer, and, as I have shown elsewhere, it was a favourite device to represent
this class of athletes with a sombre look.2 The head has doubtless retained much of a
real model. The Ince Blundell and Riccardi heads belong together to the same period, but
are later than the Diskobolos. The strikingly different individuality of the three heads
need not perplex us ; for from what artist should we expect such variety sooner than
from Myron, who multiplicasse veritatem videtur ? It is specially interesting to learn
from the Ince Blundell head that, side by side with the usual broad type of face,
Myron also had a narrower type with a more delicate root to the nose, inasmuch
as we discovered that the works of Kresilas also exhibit both types.
We are now in a position to understand an adverse criticism which Pliny passed
upon Myron : capillum quoque et pubem non emendatius fecisse quam rudis antiquitas
instituisset. The Berlin copy of the Diskobolos bears witness to the archaic cha
racter of the hair in this famous masterpiece, and we cannot wonder that the critic,
judging from his Lysippian standpoint,3 should have viewed with contempt this
manner of treating the hair, and merely seen in it a sign of rudis antiquitas. The criti
cism recorded by Pliny has, however, generally been taken to mean that Myron bestowed
no care or pains upon the hair — an interpretation which is entirely contradicted by
the Berlin head. As a fact, in comparison with the artists of his time, Myron treats
the hair with exceptional richness and naturalness. How keenly he appreciated its
varying effects is evident from the trouble he has taken to distinguish between the
little curling fringe of hair and the masses on the top of the head.4
Closely bound up with this ancient criticism upon Myron's treatment of hair is
another that accused him of neglecting to express the ' sensations of the mind ' :
corporum tenus curiosus animi sensus non expressit. This is perfectly intelligible in the
case of the Diskobolos : the quiet impassive expression of the head clashes with the
violent movement of the body. The critic was quite justified in his remark, judging, as
he did in the case of the hair, from the standpoint of the fully developed art of Lysippos.
But he would not have been justified had he wished to criticize from the historical
standpoint ; for the absence of emotional expression was merely a characteristic of all
1 This difference comes out clearly in the measurements : in both heads (the Ince and the Riccardi) the
length of the lids (34 — 35 mm.) and of the lower part of the face, and the distance from nose to brow (70 mm.
in each case), are identical ; but the distance between the inner corners of the eye differs markedly (Rice. 34, Ince
29 mm.), as also the width of the mouth (Rice. 51 — 52, Ince 44J mm.)
2 Cf. Olyinpia, vol. iv. Die Bronzen, text p. 10 seq.
3 Cf. my treatise Plinius u. seine Quellen, p. 69 seq. (ix. Suppl. vol. of Fleckeiseu' s Jahrbuch).
4 The pubes of the Diskobolos is treated with just the same care : it is arranged in four rows of small, flat
curls, elaborately worked in a manner suggestive of the most careful chiselling ; those of the upper rows are
comparatively larger and less curled than those of the lower ones. The pubes seems most accurately rendered in
the torso of the Capitol (Helbig, Museums, 446). The affinity to the hair of the Berlin head is striking in this
case. Next in accuracy comes the London copy ; the Vatican statue is less exact.

Fig. 72.— Head in Brescia. (By permission of Messrs. Bruckmann, Munich.)

PORTRAIT-HEADS BY MYRON 175
work of the earlier part of the fifth century, though it was less striking in the case of
figures in repose than in those represented like the Diskobolos in violent movement.
Yet if we compare the head of the Diskobolos to other works of about the same
date, such as those athletes we saw reason for attributing to Pythagoras, or to
the so-called Apollo on the Omphalos, or to the sculptures of Olympia, we shall
indubitably find that a more refined and spiritual life pervades it than is the case
in any of those other works. But it is precisely because this head transcends by its
spiritual refinement that we realize the limitations that separate it from works of
the free style — that we realize all the more keenly a certain fixedness and dulness
in its expression. And this anomaly may have influenced in some measure that
unjust ancient criticism. We shall see, however, that Myron himself at a later stage
rose beyond his own earlier limitations (see p. 181).
Before passing to works which lead us gradually away from the Diskobolos
to a somewhat later period in the artist's career, we must consider one head which
exhibits the stage of art that immediately preceded the Diskobolos. It is the head
of a youth in Brescia (Fig. 72),1 whose affinities with the head of the Diskobolos
seem to me unmistakable, although everything about the head points to a cruder
and an earlier artistic phase. The origin of the characteristic dent formed by the
hair on the forehead of the Diskobolos is explained in the Brescia head, where
it forms an angle instead of a curve, and practically amounts to a parting over the
forehead. The treatment of the hair itself is very similar in both heads : on the
raised, modelled masses the individual masses are delicately carved ; they represent
short, tangled hair, with a tendency to curl thickly at the ends ; a narrow circlet
presses into this soft mass. Moreover, the head in its general outline, especially in
the powerful development of the skull at the back and in the line from the circlet
to the neck, resembles the head of the Diskobolos, except that it is not quite so
high. On the other hand, the head is inferior to that of the Diskobolos in the
modelling of the forehead (which is also lower), in the rendering of the eyelids
(especially of the part between the upper lid and the eyebrow), of the mouth, and
of the whole lower part of the face. The form of face so vividly recalls the
type of Hagelaidas as it appears in the figure copied by Stephanos,2 that in spite
of the difference of expression it appears to derive from it. The points of affinity
with the Diskobolos would lead one to suppose that the Brescia head is a copy
of an early work belonging to Myron, in which he combined the type current in the
Hagelaidas schools with the treatment of hair, the shape of head, and the expression
individual to himself. This result is specially interesting, because it seems to
confirm the tradition that Myron was a pupil of Hagelaidas — a question to which
we shall have to return.
Fig. 73 reproduces a head in the Villa Albani.3 It is obviously a portrait,
the unusually wide mouth and thin parted lips 4 being distinct traits of individu
ality. Hair and beard are short. All these characteristics, joined to the erect
1 Diitschke, Oberital. vol. iv. No. 336. Length efface 0-14. Earlier notices : Conze, Arch. Anz. 1867, IOo>
(who calls the beautiful head ' a wretched copy'); Kekule, Annali, 1865,62; Gruppe des Kiinstlers Menelaos,
p. 40, 4 (Benndorf) ; Heydemann, Mitth. aus Oberital. p. 29, No. 44, 3.
2 This resemblance had also struck Conze, loc. cit.
3 La Villa Albani descritta, No. 744, ' Pericle,' evidently so called on account of the similarity of beard
and hair to the Perikles of Kresilas. Brunn, Bull. d. Inst. 1851, 88, proposed to recognize Peisistratos, but cf.
Helbig, Museums, 834.
4 The width of the mouth (51 — 52 mm.) measures i| the full length of the eye (29 — 30 mm. from corner tn
corner). Nose to arch of brows 60 mm.

Fig. 73.— Portrait-head in the Villa Albani. (From the cast.)

Fig. 74.— Portrait-head in the Hermitage. (From the original.)

178 MYRON
carriage of the head and its straightforward glance, produce an impression of great
energy. The hair frames the forehead in a manner that recalls the Riccardi head
and the head at Ince ; on the top it is left blocked out in masses which only
slightly curl at the end, while over brow and neck it spreads into a profusion
of ringlets in the true Myronian manner. Quite in Myron's style too are the
form of the eyelids and the tear-glands, the forehead that bulges out towards the
root of the nose, the round cranium, and the slight depression in the neck. It is
evident however that the work is somewhat later than the Diskobolos, and equally
evident that in it we have the immediate forerunner of the Perikles and of the
Vatican ' Alkibiades.' The form of the moustache, which leaves the upper lip un
covered, and the shape of the beard recall in particular analogous traits in the
Perikles. Since there were plenty of occasions for the private dedication of portraits
in the days of Myron, and since we have absolutely no nearer data to go upon, it is
naturally impossible to name the personage pourtrayed.
A head in the Hermitage (reproduced from the original, Fig. 74) x has a still
more marked individuality, though in point of style it is closely akin to the preceding.
The treatment of hair and beard, the heavy eyelids with the strongly defined tear-
glands, the rounded though elongated shape of the head, with the slight depression in
the neck, are all obviously the same in both heads. The outline of the profile of the
Hermitage head is further surprisingly like that of the Massimi Diskobolos. This
head appears to stand in much the same relation to the Albani portrait that the
Ince Blundell boxer does to the Riccardi. It belongs to the small-faced type,2 in which
the corners of the eyes are nearer together, the root of the nose and bridge more
delicate, the mouth smaller. These common characteristics receive a singularly
animated and individual expression by the addition of a number of interesting
details. First we note the mouth, with full, parted lips, very different to the thin
lips of the Albani portrait ; then the moustache with its curling ends, the rich beard,
elegantly parted in the middle, thus leaving the line of the lower lip uncovered. The
forehead is modelled3 like that of the Ince Blundell boxer. The eyes do not look
straight out, but the gaze inclines a little to the left. The whole expression is
indicative not so much of prowess, as in the Albani portrait, but rather of cleverness,
of worldliness, and of a certain refined sensuousness. We can scarcely doubt that we
possess in this head a copy of a work by Myron, belonging to the same period as the
Ince Blundell boxer, and affording fresh proof of the artist's remarkable power of
reproducing individuality.
We can associate with this portrait a work whose subject is taken from the circle of
the heroes— a head of Herakles over life-size, which was found in the Villa of Hadrian,
near Tivoli, and is now in the British Museum (Fig. 75).4 The individuality of the
hero is thoroughly accentuated, yet the characteristics of style are essentially those
which we observed in the head from the Hermitage. The eyes are unusually big and
round, rather than long— a trait which especially distinguishes Herakles in archaic art ;
1 Gue'de'onow, Must'e de Sculp. Ant. No. 68.
2 Clear length of eye 36—37 mm., without the gland 32 mm. ; this last measurement is also that of the
distance between the inner corners of the eyes. The width of the mouth amounts to \\ the length of the eye
without the gland (48 mm.) Nose to the arch of the brows 71 mm. (= twice the length of the eye). Length
of ear 62 mm.
3 Unfortunately not clearly enough indicated in our reproduction.
4 Well produced in Specimens of Anc. Sculp, i. 9, 10. Less well in Anc. Marbles of the Brit. Mus. i. 12.
My notice of the head in Roscher's Lexikon, 2163, 10, was influenced by this untrustworthy reproduction. Only
the front part of the nose and the edges of the ears have been restored.

MYRONIAN HERAKLES

[79

on Attic vases of the severe style, for instance, dating about the time of the Persian
wars, he is represented with large, wide-open eyes.1 Lids and glands are rendered as
on the Albani and Hermitage heads. The edges, too, of the sharply moulded lids
terminate in a little wrinkle, as in the Riccardi head. On the sharp edge of the brow
the hair of the eyebrows is slightly indicated — a singularity which is not, I think, to be

Fig. 75. — Herakles in the British Museum.

ascribed to the copyist, but to the artist's search after realism, or after some touch
that should emphasize the powerful nature of the hero. Hair and beard arc quite
short, and arranged in small tight curls. The vases teach us, again, that these short
ringlets belong to the typical conception of Herakles in archaic Attic art. The
moustache alone is smooth, and is treated similarly to that of the Albani head. The
framing of the forehead by the hair resembles the heads already mentioned, with this
1 Cf. Roscher's Lexikon, i. 2161 seq.

l8o MYRON
difference, that the forehead in this case is lower and wider, and that just over the
centre the little locks are brushed up, obviously in order to heighten the impression
of strength. At the sides and over the temples, however, the hair of the Herakles
is treated similarly to that of the other Myronian heads we have been discussing.
The modelling of the forehead is also the same, except that the forms are more
pronounced. Wc should also notice two furrows in the forehead, which start upwards
from the point where the line of the eyebrows meets the root of the nose ; this is
another detail in which the art of Myron appears as the immediate predecessor of
that of Kresilas. The nose, too, is interesting in this respect. Enough of it is
preserved to show how realistically the ridge, which in this case is naturally made
wide, broadens out in the middle. The lips are full and slightly parted, as in the
previous heads ; the line of the under lip is left uncovered by the beard, as in the
Hermitage portrait. The lower part of the cheek-bones is strongly defined under the
beard. The ear, however, is not represented as swollen. The copyist has made the
curls of hair stand out in very realistic fashion by means of under-cutting — doubtless
imitating in this particular the sharply chiselled bronze of the original. The shape of
the cranium is round, and the depression that usually marks the transition to the
occiput is omitted, owing to the unusual prominence of the muscles of the neck that
begin at this point. Thus at every point in this head we note the attempt to emphasize
the character of the hero, and the conception of him that was thus established inspired
all subsequent representations of Herakles, even down to the age of Lysippos. The
unnatural size of the eyes was, however, abandoned at a later date. The relation of
this head to the heads previously discussed is so close, in spite of certain deviations
necessitated by the different nature of the subject, that I incline to attribute it also to
Myron. We know of as many as three statues of Herakles from his hand : of these,
one made a short sojourn in Rome,1 and another remained there.2 It is probable
therefore that in our store of copies more than one Herakles after Myron may be
preserved.3

X. Statues by Myron. — Diskobolos and Marsyas compared. — Kindred Works.
So far we have dealt solely with heads which can be traced to Myron : we now
turn to whole statues. It would be strange indeed if among the numerous copies which
have been preserved there were not found a considerable number after statues of an
artist who, precisely in the age in which the copies were made, was always named
among the most eminent of sculptors, along with Pheidias, Polykleitos, Praxiteles,
and Lysippos.
The two statues of which we have undoubted copies, the Diskobolos and the
Marsyas, afford a firm basis for further investigation. In both the body is represented
in violent exertion, in both the energies of body and of mind are concentrated in one
dominating movement. The prevailing note of Myron's artistic temperament was
doubtless love of movement, though he cannot have always had an opportunity for
indulging it.4 His statues of the gods must have been represented in tranquil attitudes,
according to the fashion of his time. Coins- show us the Athena in the Marsyas group
1 Strabo, xiv. p. 637, 6 = Overbeck, S. Q. 536. " Plin. xxxiv. 57.
3 See infra, p. 202, with regard to a probable second Herakles after Myron.
4 Cf. Kekule, A'opfdo Praxil. Hermes, p. 15 seq.

DISKOBOLOS AND MARSYAS

181

standing calmly,1 while the Riccardi bust proves that Myron had represented one hero
at least in an attitude of repose. The same marked powers of individualization that
we noted in the treatment of the heads make themselves felt in the bodies of the
Diskobolos and of the Marsyas. The first shows us a well-bred youth, carefully
trained and developed in the palaestra, executing a studied movement with graceful
precision. The second statue shows us a wild, sinewy creature of the woods,
accustomed to move by bounds and springs, ignorant of any training, tormented by
unbridled passions, and at this moment a prey to curiosity and fear alike. The
contrast to which we allude runs through the whole of each figure, and may be
observed from head to foot.2 Both hair and beard are in the case of Marsyas treated
in an exceptionally characteristic manner ; the great moustache is rough and bristly ;
it is not connected with the rest of the beard, which is itself divided into two distinct
parts, consisting of the hair flowing from the chin and of the whiskers. The hair of
the head not only bristles up in front, but grows also on the neck in short, flame-like
wisps that go up instead of down. This arrangement of hair is affected by our
modern clowns, and suits the excitable nature of the droll Seilenos much better than
those heavy masses of hair flowing downwards from the back of the head that were
generally given to Satyrs in Myron's time.3 With the exception, however, of the front
wisps, the hair lies close to the small head, the round outline of which passes into the
neck without the break noted in other Myronian heads. Thus Myron, who had given
to the Diskobolos the long crown of the aristocrat, gave to the Seilenos the small
round head of the plebeian.4
The Marsyas undoubtedly bears the stamp of a later date, and we can see that
the character of Myron's works varies considerably according to the time of their pro
duction. Further, while in the Diskobolos the expression of countenance is quite
unaffected by the violent movement of the body, in the Marsyas expression and
movement are in perfect harmony. Astonishment and fear combined are admirably
expressed in the head. That ancient criticism about the want of continual expression
in Myron applies, like that about his archaic treatment of the hair, properly only to
the Diskobolos and kindred heads, and not to the Marsyas ; and it would be unfair to
allow it to influence us in our judgment of Myron's artistic capabilities.
The more marked the differences between the Marsyas and the Diskobolos, the
more must we insist on their common elements. To these belong above all the
severity and the spareness of the forms in both statues. Even in the Diskobolos, who
has, it is true, a somewhat fuller body, there is nothing of the nature of soft flesh ;
indeed the forms, in their severe sobriety, may almost be accused of hardness and
meagreness, though the spirited action of the statue helps us to overlook this defect.
1 L. von Sybel, Athena u. Marsyas, Bronzemiinze des Berl. Museums, Marburg 1879.
2 Even the pubes is treated characteristically. In the case of the Diskobolos the hair is cut straight off
above and arranged in the same little curls as in the hair of the head (cf. supra, p. 173) ; in the case of the
Seilenos it takes the form of larger, coarser, curls, and spreads in a triangular shape high up the belly.
3 The nearest analogy to the head of Marsyas and its flat short hair is presented in the Argive bronze,
reproduced (very inadequately) Rom. Mitt. 1889, p. 170, which is undoubtedly later than Myron. This bronze
justifies the supposition that in his Marsyas Myron possibly allowed himself to be influenced by Argive models.
4 It is only the head of the Lateran statue that can be considered as an approximately faithful copy of the
original. The replica Coll. Barracco, PI. 37 and 37 a, which is usually taken as better (see Friederichs- Wolters,
Gipsabg. 455, and more recently Collignon in Mc'langes d'Archdol. et d'Hist. 1890, p. 118 sqq. PI. 2 ; Hist, de la
Sculpture Grecque, i. p. 468), proves on closer examination to be decidedly inferior. The copyist of the Barracco
head has obviously taken exception to the simplicity and hardness of the original, and tried to improve upon it ;
he has introduced richer and more flowing but more conventional lines ; the typical characteristics dwelt on
above, such as the treatment of the moustache, its separation from the rest of the hair of the face, the peculiar
treatment of the hair of the head — all this is abandoned and made commoner.

182 MYRON
If one imagines the Diskobolos standing upright in repose, and then compares it with
the so-called Apollo on the Omphalos, this peculiarity comes out very strongly. The
Apollo has full, soft flesh ; the Diskobolos, flat, spare forms. In this point Myron comes
into intimate connexion with the group of works which I have felt obliged to ascribe
to the school of the Argive Hagelaidas.1
In the Diskobolos, further, the obvious difficulties caused by the violent bend of the
body are by no means quite overcome : this is especially striking in the representation
of the lower part of the trunk.2 Moreover, the treatment of the planes, which remind
one of work in relief, may be taken as a sign of confusion of ideas, due probably to the
fact that just about this time figures in motion were being transferred from relief to the
round. The Marsyas, on the other hand, is already conceived throughout for execu
tion in the round.
We accordingly find that the ancient orators name him among the last masters
of the severe style.3 The culminating point of his career would be reached about
465 — 450, a view which has lately been confirmed by the discovery that his son Lykios
was already an artist of independent position in 446 B.C., or at any rate at a period
previous to the building of the Propylaia.4 Thus Myron flourished before the mighty
artistic revolution which found expression shortly after the middle of the century in
the sculptures of the Parthenon, where only a few metopes preserve that severer style
which we have noted as a characteristic of his works.5 Our next task is to find out
whether some conception can be arrived at, through extant copies, of the statues of
Myron that stood in repose.
Among extant statues there is only one which I could at once bring into
immediate relation to the Diskobolos. It is a Mercury in the Vatican, which,
though fully deserving the epithet ' bad ' given it by Brunn ° (for it is an extremely
poor copy), is yet of considerable significance (Fig. 76).7 The family likeness of the
head to that of the Massimi Diskobolos is self-evident : it is particularly striking in
the expression and the shape of the lower part of the face, with its full mouth and
round chin. The framing of the forehead by the hair resembles the Riccardi and
kindred heads, and the forehead itself, in spite of the poorness of the copy, preserves
the modelling characteristic of Myron.8 The profile shows substantial agreement with
that of the Diskobolos ; even the ears have the characteristic wide shape, with the small
lobes. The hair is arranged in little curls, which in the original would doubtless appear
fuller than in the copy. The upper eyelids are somewhat heavy. The turn of the
head, and the quiet upright pose of the figure with hanging arms, correspond exactly
to what we should imagine was the pose of the statue from which the Riccardi bust
was copied. The weight of the statue is thrown on the right leg, leaving the left leg
comparatively free, though it is not drawn back, but is planted with full sole on the
ground. The somewhat severe, spare, and wiry forms correspond on the whole with
those of the Diskobolos. The feet, for instance — in the rendering of which the copyist
1 Cf. 50th Berl. Winckelmannsprogramm, p. 134 seq.; supra, p. 52 seq.
2 Cf. Jul. Lange, Fremstilling af Menneskeskikkelsen (Abh. d. Kopenhag. Akad. 1892), pp. 394, 397, 462.
3 Cicero and Quinctilian in the oft-quoted passages (Overbeck, S. Q. 600—602).
4 Lolling, AeXrtov, 1889, 181 seq.
" I should not like however to bring either these or the metopes of the Theseion into direct relation with
Myron himself. Cf. supra, p. 46. 6 Brunn, K. G. i. p. 613.
? Vat. Gall, delle Statue (Not. 417) ; Mus. Pio-Cl. iii. 41. The head had been severed from the body, but
certainly belongs. Nose, both forearms, and the piece of hanging drapery are restored.
8 Unfortunately not visible on the photograph from which our illustration is taken.

Fig. 76. — Mercury in the Vatican,

1 84 MYRON
has succeeded best — show in their vigorous shapeliness, and the realism with which
even the veins are indicated, a special affinity to the feet of the Diskobolos.1 On the
other hand, the chlamys, which is fastened on the right shoulder with a great round
button, cannot have been part of the original statue, and was evidently added by the
copyist ; for the character of the folds is purely Roman — indeed, it may be definitely
stated that the addition of the chlamys is a favourite device of a certain group of
copyists of the later Empire. For instance, it was given to the later replicas of the
famous ' Meleager ' after Skopas ; the earlier copies are without it.2 Some copyists
also gave a chlamys to a statue of Apollo of which the original must have been very
like the bronze Apollo Sabouroff in Berlin.3 These instances could easily be multi
plied ; the last one is particularly interesting, because the chlamys of this Apollo
agrees so closely with that of our Hermes that one might almost assume both statues
to have been executed in the same workshop at the end of the second or the beginning
of the third century A.D.4
Further, the wings on the head,5 which are attached in a manner both thoroughly
inorganic and thoroughly non-Greek, are certainly an addition of the copyist, just like
the wings that appear from time to time on heads imitated from Polykleitos. In
Roman times it was not uncommon to turn Greek figures of youths into statues of
Mercury. It is now impossible to determine whether the original by Myron represented
a hero, a mortal, or Hermes (without the wings on the head). Pose and attitude
have the merit of being natural and unstudied. The youth stands at ease, though
not at rest; ready at any moment for energetic action. This is quite a contrast
to those attitudes of rest that seem to have been the ideal of the art of the Pelo-
ponnesos. A few other statues may be grouped about the Vatican Mercury. In the very
same room of the Vatican — unfortunately, in a very unfavourable light — is a statue
of a bearded man (Fig. J7)e restored as a Poseidon ; in pose and in attitude, as also
in the spare, attenuated forms of the body, it agrees very closely with the Mercury.
Both arms hang in the same way, and the head has the same turn.- Although the
copyist, who belonged at the earliest to the age of the Antonines, has, like the artists
of the sarcophagi executed at that time, worked the hair elaborately by means of the
drill, it is still clear that the type of head is of the fifth century, and on close
examination traits may be detected which recall the technique of Myron. Hair and
beard, in spite of the modernized appearance due to the drill, still bear witness to the
1 The pubes is also identical.
2 List of replicas, Rom. Mitth. 1889, p. 219 (Graf). The copies with the chlamys (which moreover varies
in each) all belong, so far as I can tell, at the earliest to the later Empire ; the Vatican and the Borghese statue
in particular are characteristic works of the age of the Antonines. The beautiful Berlin torso without the
chlamys is undoubtedly older.
3 Skulpt. No. 51 ; Overbeck, Apollon, p. 226, 5. For the Sabouroff bronze see Sabouroff Coll. PI. 8— 11 ;
Overbeck, loc. cit. p. 227.
4 Cf. also the very similar chlamys of the statue of the youthful Marc. Aurelius mentioned above, p. 92, note 4,
and other portrait-statues of the same period. Visconti also (Mus. Pio-Clem. iii. PI. 41) says that the Mercury is
certainly not older than the age of the Antonines, possibly much later : he quotes in support the inscription on
the plinth, which however need not necessarily be of the same date as the statue.
5 The right wing is ancient.
Gall, delle Statue, No. 394 ; Mus. Pio-Clem. i. 32; Clarac, PI. 743, 1796; Overbeck, Runs/myth. Part ii.
Poseidon, p. 287, No. 14, p. 289 seq. ; for the head, cf. p. 267, No. 10 ; Atlas, PI. XI. 9, head in profile ; XII.
35, the whole statue. Both arms from the middle of the upper arm are restored, as also the lower parts of the legs
and the attributes, fish and trident. The portion of the shaft of the trident against the upper part of the arm
is ancient ; its squareness led Visconti to recognize in it the remains of a trident. However, as the fragment of
shaft had been separately attached, we cannot feel certain that it belonged to the original.

Fig. 77.— Statue restored as Neptune (Vatican).

li H

1 86 MYRON
severer style of the original, in the simple, heavy strands that curl up at the ends.
The semicircular framing of the forehead by the hair, the vigorous modelling of the
forehead which projects towards the eyebrows, and the slightly open mouth are all
familiar to us. The hair is brushed up from the forehead. Certainly this statue is no
portrait, but an ideal type, in which power and majesty are united. The statue may
well represent a god, perhaps Poseidon, as Visconti conjectured.1
Even if the head has lost by the hand of the copyist something of its original
severity, the statue, whose connexion with the Vatican Mercury is obvious, yet serves
to give us a notion of the dignified manner in which Myron represented the gods. The
conception is really godlike and quite distinct from human types ; the same power of
individualization makes itself felt here that was noticed in the other works attributed to
Myron. The attempt to individualize the gods must have been quite novel in those
days ; for instance, the hair that rises up from the brow of the ' Poseidon ' had not yet
become a fixed characteristic of the sons of Kronos, and was apparently now intro
duced for the first time.
There is another statue very closely related to our Vatican Neptune ; I mean the
statue to which a beautiful head in the Berlin collection,2 reproduced in Fig. 78,
belonged. If we disregard superficial mannerisms due to the copyists, the general
likeness of the head to that of the Neptune is very striking. The hair of the massive
beard, without being deeply carved, produces a rich and varied effect, vividly recalling
the beard of the Marsyas ; the moustache, with the simple, straight lines and its clear line
of demarcation against the upper lip, is akin to the moustache of the Albani ' Peisi-
stratos ' and to that of the Perikles. The hair — exactly like that of the Vatican Neptune
— is treated after a method which disappears in the second half of the fifth century (cf.
supra, p. 19) : it radiates from the crown of the head towards a circlet that con
fines it ; below this circlet it falls loosely forward, forming a rich crown of locks ;
on the neck it is just left long enough to cover the nape. The hair rises up on the
centre of the forehead to show the growth, and then falls away gently to either side ;
in the Neptune, however, the side locks are, so to speak, forced to the sides instead of
flowing of their own accord. Possibly the first arrangement was intended to suggest
a mild and the second a vigorous nature. At the point where the root of the nose
meets the forehead there occurs the same depression which we noticed in heads
by Kresilas. In order to emphasize the nature of the god, the forehead, as compared
with heads of athletes, assumes a more powerful and pronounced form, though the
main lines remain the same. The cranium is less developed at the back, but this is a
point which I incline to charge the copyist with ; it is not impossible that he was
limited, as was not unfrequently the case, by the size of his block of marble. The
well-defined eyelids with the delicate little wrinkle, the lachrymal glands, the slightly
open mouth — all recall the manner of Myron. The body to which this type of head
belonged forms, as we shall see, an inseparable group with the Vatican Mercury and
Neptune. We recognize a replica of the same head in a statue at St. Petersburg from the
Campana collection (Fig. 7c/).3 Though this statue is poor enough in itself, it is
1 Undoubted Poseidon statues, like those in Scherschell (Overbeck, Kunslm. Atlas, PI. XII. 34), may be
conceived as later developments of the type under discussion. Cf. also coins, Miiller- Wieseler, D. A. K. 3rd ed.
ii. 72d- 2 Beschr. d. Ant. Skulpt. No. 158.
3 Guedeonow, Mus. de Sculpt. Ant. 314; d'Fscamps, Marbres de la Coll. Campana, PI. 14. The head had
been severed from the body, but my examination of the statue has convinced me that it belongs. There are only
trifling restorations at the point where it joins on : the left side of the head, the right arm, together with the staff
of Asklepios, the left forearm with the flat cup, are restored.

1 88 MYRON
invaluable for our purpose, inasmuch as body and head belong together. The general
pose is much the same as that of the Mercury, and the same spareness is conspicuous
in the body, though the copyist has evidently tried to modify it. The god — for there
can be no doubt that it is a god we have before us — wears sandals, and a mantle,
which is flung over the left shoulder and covers the whole lower part of the body, but
in such a way as to leave the origin of the thighs free. This whole arrangement of the
mantle, and especially the portion of it that falls from the shoulder, bears witness to
the severity of the original. The simple straightforward lines recall the drapery of the
Riccardi bust. The depth of the folds between the thighs may be due to the copyist,
who altered the shallower forms of the original, to suit the taste of his time. In the
arrangement of the drapery about the lower part of the body we recognize the hand
of the artist on whom Kresilas modelled himself, when he executed the mantle of the
Athena of Velletri. On the other hand, drapery like that of the Zeus from the
eastern pediment of Olympia (a statue which in other respects is intimately related
to the one under discussion) leads to the pictorial draperies of the Parthenon. So we
may recognize two contemporary tendencies, and we now understand that Kresilas's
treatment of drapery was historically a development of the Myronian manner.
The most probable interpretation of the statue is that it represents Zeus — a view
brought home to us by comparison with the afore-named Zeus from the eastern
pediment of Olympia. Both statues agree so well in their main features that one is
tempted to conjecture that our type, with its quiet, simple lines, was known to the
artist of the Olympia pediment. The line formed by the mantle on the right hip is
just the same in both, the principal difference being that in the Olympian Zeus the
mantle has slipped off from the left shoulder and has been wound round the arm. The
interpretation as Zeus is further confirmed by the great likeness to a statue in the
Louvre,1 which is known to represent Zeus by reason of the eagle sitting below and
of the recurrence of a similar type on coins.2 In conclusion, the attitude of the statue
is favourable to the view that the left hand which hangs down grasped the sceptre,
while on the right, which is extended, was held the thunderbolt or the eagle.
It must be owned, however, that an examination of the other replicas does not at
first favour the Zeus theory. Of the eleven replicas of the Hermitage statue that
are known to me, four certainly, and a fifth probably — to judge from the remaining
attributes— represented Asklepios ; while the others, though without the attributes of
Asklepios, are equally without those of Zeus.3 Still, in my opinion, it is not probable
that the original represented Asklepios : statues of this god were relatively still
1 Frohner, Notice, p. 65, No. 34; phot. edit. Giraudon, 1224.
2 Cf. the coin of Laodikea, Overbeck, Zeus, Miinztaf. 2, 29; p. 138.
3 The replicas consist partly of statues of the same proportions as the Hermitage statue, partly of statuettes
of varying dimensions. The statues are : 1. Berlin, Beschr. d. Ant. Skulp. 290 (probably intended for an
Asklepios rather than a Zeus). 2. Capitol, lower hall, 41 : feet, arms, whole of the right breast new ; the head of
Zeus does not belong : the flatness and rigidity of the body, especially about the navel, is characteristic. 3.
Louvre, Gall. Mollien, No. 2936 : lower part of the legs new, head does not belong ; the left arm as far as the
middle of the forearm preserved. 4. Rome, Palazzo Altemps : Clarac, 560 A, 1 160 D : Matz-Duhn, 58 : head
does not belong : the lower part of the serpent staff is old : on the feet Roman shoes in place of sandals: the
copyist regarded the stouter covering for the feet as more appropriate to Asklepios : in the body the severity of
the forms is well preserved. 5. Pacetti : Clarac, 545, 1146 : not known to me in the original : the serpent staff is
stated to be restored. The statuettes are : 6. Mus. Chiaramonti, 113 : Clarac, 549, 1 157. Undoubtedly Askle
pios from the votive inscription, serpent staff restored : head does not belong. 7. Mus. Kircheriano : feet and
arms new : head does not belong : poor. 8. Cassel, ii. 15 : head does not belong, forearms new : the severity
of the style is well preserved. 9. Mus. Torlonia : Visconti, Tav. x. 39 : serpent staff old, head modern.
10. Giustiniani: Clarac, 552, 1167 A: the serpent below reported to be ancient. 11. Rome, in the market,
noted by me 1892 : head foreign.

Fig. 79.— Statue restored as Asklepios (Hermitage).

190

MYRON

rare towards the middle of the fifth century, though in the period of the copyists
representations of him were in great demand. Considering that at this late period
there was a rage for old types, it is conceivable that the copyists, for want of old
originals of Asklepios, should sometimes have seized upon originals pourtraying Zeus,
and by a change of attributes have turned them into statues of Asklepios. The
restful conception that marks the older type of Zeus exactly suited the mild character
of Asklepios. Anyhow, the erect, kingly pose of our statue ; its gesture, better suited
to grasp the royal sceptre than to hold the healing staff; its analogies to other
statues of the god ; and, finally, the type of head, to which we shall have to return,
make it clear, in my opinion, that a statue of Zeus was the original from which the
Hermitage statue was derived.
The number of replicas found in Rome x shows not only that the original was
famous, but that it was probably to be seen in Rome ; our analysis of the statue
has led us to Myron as the artist ; and a Zeus by Myron actually was in Rome
in an aedicula, erected by Augustus on the Capitol.2
Another point of special interest about our statue, and one which further
confirms its interpretation as Zeus, lies in the fact that the head exhibits in the
arrangement of the hair and the modulation of the brow all the fundamental
characteristics which were to develop into the famous type of Zeus known from
the Otricoli bust. An historical study of the system of forms of this celebrated head
shows it to belong not to the circle of Lysippos, as was formerly maintained, but
to that of Praxiteles.3 Thus, even as the Hermes of Praxiteles marks the culminating
point of a line of development that begins with the heads of Myron's athletes,4
so it would seem that the bust from Otricoli is the Praxitelean development of a
type of Zeus created in the days of Myron.

XI. The ' Cassel Apollo.' — Argive Influences traceable in Myron.
The three statues which we have considered so far and have referred to Myron
have the same position and attitude. In all probability the statue from which the
Riccardi bust {supra, Fig. 66) is copied formed a fourth in the series. The common
characteristics of this attitude are found in the main in all statues in repose of the
first half of the fifth century. Still, among these works, all of which have the
free leg placed slightly forwards, and the head turned a little to one side, there are
several sharply defined groups.5 Those ascribed to Myron are most closely allied
as regards the pose to the so-called Apollo on the Omphalos, which I refer to
To the twelve statues and statuettes mentioned above, of which only the Petersburg example preserves the
original head, must be added the Berlin head (our Fig. 78) and two replicas of the same : (a) in the Palazzo
Riccardi: Dutschke, Zerstr. Bildw. in Flor. 161 : the head is set up as a pendant to the Riccardi head;
(b) a freer copy in the Palazzo Colonna, No. 12.
2 Strabo, 14, p. 637 ; the statue belonged originally to a group with Athena and Herakles in Samos.
Colhgnon (Hist, de la Sculpt. Gr. i. p. 465) thinks that the Zeus of Myron may be recognized on the coins of
Augustus which were struck after the dedication of the temple of Jupiter Tonans (Cohen, Mid. Imper. 2nd ed. i.
88; cf. Roscher's Lex. d. Myth. ii. 748). This is a mistake. The va^Kos that Strabo mentions is certainly not
the famous temple of Jupiter Tonans ; further, the statue of Jupiter Tonans on the Capitol was by Leochares
according to Pliny (xxxiv. 79) ; the motive of the statue as shown on coins confirms this statement, and excludes
Myron, since it is not till the second half of the fifth century that statues are represented resting their arm high
against the sceptre. 3 Cf. infra under PraxiteleSi
Cf. Kekule-, Kopf des Pra.xit. Hermes. ° Cf. 50th Berl. Winckelmannsprogr. 1890, p. 149^?.

MYRONIAN APOLLO 191
Kalamis (cf. supra, p. 81 seq.) Together they stand in vivid contrast to the type
which I ascribe to the school of the Argive Hagelaidas, where the advanced free
foot is more markedly unincumbered, the shoulder droops a little, and the head
is inclined. With Myron as with Kalamis the upper part of the body and the head
are more upright, the whole presentment expressing power and energy. As regards
the rendering of the bodily form, however, we have already seen that Myron differs
essentially from Kalamis, inasmuch as he — to compare especially breast and
abdomen — exhibits a flatter, sparer style, which seems to betray his connexion with
the Argive school.
In the Apollo — generally known as the 'Apollo of Cassel,' from its principal
replica1 — we possess a statue which is unmistakably connected with the Myronian
works we have just been studying, and which at the same time shows us the artist under
the influence of the Argive type as regards the position of the body.2 The figure
rests on the left leg ; the left forearm is stretched forward in a horizontal position
to hold an attribute (in this case a bow), the right hangs idly down, and perhaps
held some slight object, such as a twig. The head is inclined to the left. The flat
smooth forms also recall the Argive tendency. On the other hand, the statue differs
from the Argive type in certain deep-seated differences, and its close relation to
the statues ascribed to Myron is no less evident. The head is not inclined forward ;
one shoulder does not droop ; the god looks straight before him, with a look not
of concentrated repose but of a strong and active will. The free foot is placed more
to the front, and is more turned out than is customary in those Argive statues :
this helps to give energy to the pose. The same was the case in the Zeus and the
' Mercury ' of Myron.3 The treatment of the bodily forms also comes nearest to
that of the Myronian statues. We have noted their connexion with the Argive
works ; it remains to point out the differences. Above all, the contrast, so
characteristic of Hagelaidas, between the narrow hips and the over-broad angular
shoulders can no longer be detected ; it is only by comparison with the Myronian
Zeus and 'Neptune' that the Apollo still appears somewhat hard. Further, just
as the turn and pose of the head correspond to that of the Myronian statues of gods,
so also does the relation of the head to the rest of the body; in this respect it
seems also to agree with the Diskobolos ; the head is somewhat larger than is usual
in the other works from the circle of Hagelaidas.4 Further, the muscles are more
1 Known to me are : Entire statues— I. In Cassel, Overbeck, Apollon, PI. 20, 24 ; p. 166, I ; 108, 1 ;
where also see the older literature. The work is hard, but careful in all parts : it gives the impression of great
exactness. Through a mistake in putting the fragments of the statue together a wrong turn has been given
to- the upper part of the body ; the mistake lies in the middle of the left thigh and under the right knee ;
jt can be checked by the aid of the Paris statue. 2. Louvre, Overbeck, loc. cit. 2 = our Figs. 80, 81.
3. Torso, Athens, Central Museum. Finally the following heads : 4. In Athens, found to the east of the
Olympeion, Sybel, 747 ; Overbeck, loc. cit. 108, 3. Careless and feeble work of Roman time, much damaged.
5. From Greece, now in the Louvre, No. 2032 (phbt. Giraudon) ; very probably identical with that noted by
Benndorf, Ann. 1880, 198, 5, at a dealer's in Athens. Poor Roman work. 6. From the Esquiline, Rome,
Barracco Coll. : Overbeck, p. 108, 4 (see our Fig. 80). 7. From Rome, formerly in Barracco Coll., now belonging
to Brunn in Munich. 8. In Naples, set on the term of the so-called Polyhymnia. Mus. Naz. Inv. No. 6393.
I have not seen the following heads noted by others as copies : Overbeck, p. 108, 6, belonging to Herr v.
Keudell ; Schreiber, Villa Ludovisi, 315 ; Dtitschke, Zerstr. Bildw. in Florenz, 520 : ace. to Benndorf, Annali,
1880, 199, probably belongs to the series ; Arch. Anzeiger, 1891, p. 181, 3, in Vienna, University Collection.
2 Cf. 50th Berl. Winckelmannsprogr. p. 135 seq. ; supra, p. 49.
3 The middle part of the feet of the Cassel statue as well as a piece of the plinth is old ; the feet are close
together. The restoration of the legs in the Paris statue should be corrected accordingly.
4 The Argive bronze, published by me in the 50th Winckelmannsprogr., shows that at a later date, circa
B.C. 460, the old Argive canon was improved in Argos also: the head was made larger, and the hips
broader.

Fig. 80. — Apollo of the ' Cassel type (Louvre).

MYRONIAN APOLLO

193

energetic than in this latter school ; the prominence of the collar-bone just under the
neck dimple recurs on our Myronian statues and in the Diskobolos, though it is
not usual in the flat quiet treatment of the early Argive statues.1
The head seems to afford definite proof of the Myronian authorship. We
recognize in it the characteristic lower part of the face with the slightly open mouth
and the form of the mouth 2 itself, not to speak of the chin, which accords very closely
with that of the Riccardi bust. Then the modelling of the forehead and the well-
defined arch of the eyebrows are very like the Diskobolos, while the peculiar parting

Fig. 81.— Head of Apollo.

of the locks in the middle of the forehead is the same as in the ' Neptune ' after
Myron. Further, the heavy tresses curling at the ends in the Apollo are identical
with those of the ' Peisistratos ' (Fig. 73), while the plastic treatment of the beard
must be compared to that of the Marsyas.
1 It is further characteristic that the pubes is fully and powerfully formed, as in the Diskobolos and the
Myronian ' Mercury ' ; in the circle of Hagelaidas it was invariably left out in the case of youthful beardless
figures. The pubes of the Apollo is interesting in itself : it consists of two parts — an upper row of smooth,
scarcely turned up, and very regularly disposed hair, and a lower row immediately about the membrum,
consisting of the same full plastic locks that appear on the brow. I know of no exact parallel to this method
of representing the pubes, though the nearest analogy occurs on Myronian statues : the upper smooth row of hair
is found on the best copies of the Diskobolos, and the lower curled part of the pubes resembles that of the
Marsyas. 2 Next to the Cassel head the mouth should be studied in the Louvre head (supra, p. 191, note 1, No. 2),
which is very careful in this point : the lips are defined here too by incised lines in imitation of bronze technique.
C C

194

MYRON

The points in the Apollo that deviate from the usual Myronian head are only
such as may well have been chosen to suit the character of the god (Fig. 8 1). The most
striking of these is the great length of the lower part of the face, which is not other
wise found in Myron, though it is found in Kritios and Nesiotes 1 (with whose work,
however, the Apollo has in other respects absolutely nothing in common). Now, as
the length of the lower part of the face together with the slightly open mouth, in
which the teeth are visible, contributes materially to produce the impression of an
exalted and unapproachable divinity, I incline to think that the artist chose it solely
on that account. Indeed, the head betrays in every point a studied adaptation of the
facial forms to the production of a definite expression. The shape of the brow is
especially notable in this connexion : it is high in the middle, and broadens out very
considerably at the base towards the temples, where it attains its greatest width
in the strongly projecting eyebrows above the outer corners of the eyes. This
formation is continued by the wide cheek-bones and the broad round skull, which
Myron, as we have already seen, was wont to make characteristic, and not after a
conventional pattern. The eye is correspondingly large, with a long narrow slit. As
the brow is given a overhanging shape, the slant of the eyeball from top to bottom is
very considerable. The art of later days kept to the leading traits of this creation : ,
the broad scheme of the upper part of the face becomes henceforth characteristic of
Apollo, just as we noted that the later ideal of Zeus was based upon the type
created by Myron.
The treatment of the hair is also turned to account to express divine majesty and
power. We have already noticed the ample crown of hair in the Neptune and the
Zeus of Myron. In the Apollo it has a somewhat stiffer character, owing to the work
being on the whole somewhat earlier. This rich crown, consisting of individual locks
treated in a very natural manner, was something new, and our Apollo seems to afford
the earliest example of it. It took the place of the sober roll of hair worn by the
earlier Apollo of the school of Hagelaidas, which has been ascribed to Hegias (p. 52).
Doubtless this Argive Apollo was known to our artist, for he reproduces the main
features of the hair : in the Myronian statue, as in its predecessor, we get the two
twisted locks at the side falling to the upper edge of the breast, together with the
trim coiffure at the back and the part above the forehead. But the favourite 2 roll of
hair of the Peloponnesian artists, with its stiff, smooth regularity, is here dispensed
with, and replaced in front by a wealth of natural curls, at the back by the plait that
was in vogue in the Attico-Ionian school.3
One may indeed pretty confidently suppose that an Apollo by Myron exists
among the marble copies that are preserved to us of older originals. For the
inscription AttoWwv Mvpavos on a headless term in Rome4 shows us that an
Apollo of Myron was known in Roman times, and that its head was highly enough
valued to be copied alone apart from the body. Pliny (xxxiv. 58) also mentions an
Apollo by Myron, which remained for a long time at Rome, until Augustus restored
it to the Ephesians. Moreover, as Verres had despoiled Agrigentum of another
Apollo of Myron, there was evidently more than one statue of this god by this
artist, and those that are known to us in copies need not necessarily be identical
with either of the two which happen to be mentioned in literary tradition.
The provenance of the copies affords no clue whatever to the place where the
1 Cf. the Harmodios and the Boy from the Akropolis.
Cf. 50th Berl. Winckelmannsprogr. p. 128 seq.
J Cf. Ath. Mitth. 1883, 246 seq. (Schreiber). ¦» Kaibel, Inscr. Gr. Ital. p.^698, 1256 a.

Fig. 82.— Apollo head of 'Cassel type' in Barracco Collection (Rome)

1 96 MYRON
original stood. These copies all belong to Roman times, and as usual with copies
doubtless served to decorate private houses, baths, and the like. The same holds
good for the head and torso from Athens ; both came in all probability from the
Roman ruins of the city of Hadrian around the Olympieion ; x and we should not be
justified in concluding from this circumstance that the original was in Athens.2
All that we can say positively about this Apollo, which is avowedly one of the
grandest ideal creations of Greek art, is that it must have been highly prized in
antiquity, seeing the many copies of it which have survived.
On the other hand, it becomes possible to throw light on Pliny's statement
(xxxiv. 57) that Myron was a pupil of Hagelaidas, now that we have clearly detected
in the Apollo the influence of the old Argive canon. We have seen that the artist of
the Apollo was thoroughly acquainted with the epoch-making creations of the old
Argive school, even though his work is at the same time widely different from them ;
in addition to this, we have detected a certain Peloponnesian influence in Myron's
whole treatment of the bodily forms, and, in the case of the Marsyas, even in his
treatment of the head. It is, however, no longer possible to ascertain whether Myron
was really a pupil of Hagelaidas or whether the tradition arose, as in the case of
Pheidias (cf. p. 54), owing to some ancient art critic's observing in Myron traces of
the master's influence. The latter seems to me the more probable. The tradition,
for which Pliny is responsible, is derived from the same good old source (probably
Xenokrates) as that which says that Polykleitos also was a pupil of Hagelaidas8 —
an impossible relationship in its literal sense, since Polykleitos, even dating him as
far back as possible (cf. supra, p. 130), is still too widely separated from Hagelaidas
to admit of it : his master must have belonged to the intermediate generation — a
generation that was however soon forgotten, inasmuch as it achieved no renown ;
in time therefore the head of the later school was made to follow close on the
famous head of the old. All the same, it is true that the canon of Polykleitos was
developed from the old canon of Hagelaidas, as we too may still recognize,4 but the
actual teacher of Polykleitos must have been some artist like that Argeiadas (Lowy,
/. G. B. 30) who was proud to be known in Olympia as the pupil of Hagelaidas. If
the story of the pupilship of Polykleitos is not literally true, it is unlikely that the
similar story about Myron should be any the more true, though in his case chrono
logical considerations do not stand in the way. The traditions must have had their
source in those works which reveal a point of contact between Myron and
Hagelaidas ; who it was that transmitted this influence remains unknown to later
ages. It is however quite certain from monuments such as the Apollo that the story
is no mere capricious invention.
The Apollo enables us to look not only backwards but forwards. It again proves
how entirely Myron was the forerunner of Kresilas. In order to be once more
convinced of this it suffices to consider the stylistic treatment of the hair of the Apollo
1 The statue of a boy, Jahrb. d. Inst. 1893, PI. 4, in which Mayer (ibid.) sees a copy of the Splanchnoptes
of Styppax, was also found in the same region.
2 Nor do the coins of Athens afford any certain clue. The type of Apollo on coins— Imhoof-Gardner,
Num. Comm. PI. CC, XV., XVII. ; Overbeck, Apollo, Aliinztaf. iv. 33— shows close agreement in the general
scheme with our type (left leg bearing weight, left forearm extended with bow, right hanging); another
Athenian coin '(Imhoof, etc., loc. cit. PI. CC, XVI. ; Overbeck, iv. 29) shows much the same Apollo, only seen
more from the front, and for this reason with the supporting leg reversed. I formerly tried to identify this with
the Omphalos Apollo (Roscher's Lexikon, i. 456) ; it could also be referred to our Cassel type (cf. also
Winter in the Jahrb. d. Inst. 1887, 235, note). But this too is quite uncertain, and there is yet another
possibility open to us, which I shall discuss in connexion with Euphranor (infra, p. 356).
3 Plin. 55, 9. Cf. Robert, Arch. Mdrchen, p. 92. 4 Cf. 50th Berl. Winckelmannsprogr. p. 149.

THE PERSEUS 197
(compare the form of the little curls with those of the Perikles), the shape of the eyes
(in the better copies, such as the one in Cassel), and of the mouth with the teeth,
which reminds us at once of the Medusa.
In conclusion, the Apollo invites comparison with the other statues of Apollo of
the same period — i.e. of the years ranging from about 475 to 460. First and in notable
contrast come the two statues, in which we recognized the style of Kritios (p. 54 seq.) and
of Hegias respectively (p. 52 seq.) The one referred to Hegias is under the immediate
influence of Hagelaidas : it is a beautiful work, though somewhat heavy and dull in
tone ; the other is quite independent of that master, and in many respects — especially
in the pose — less perfect, though fresher in conception. Somewhat later follows the
so-called Omphalos Apollo, attributed to Kalamis,1 also independent of Argos ; the
body is a magnificent creation, teeming with force, while the head scarcely rises as yet
above a human type ; it even relinquishes the attempt made in the two preceding statues
to mark the god at any rate by the headdress. It is very different with the two approx
imately contemporary works that follow — an earlier work of Pheidias {supra, p. 49), still
somewhat constrained and recalling his master Hegias, and the work executed by Myron
at the highest point of his development. Both try to go deeper than their predecessors,
both try to mark more clearly the character of the god and to raise it above the human
type, yet both works are absolutely different. Myron's assuredly deserves the prize.
The genius of Pheidias is still under the constraint of youth, and his attempt to express
the spiritual element in the god proceeds rather on the lines of a gentle beauty, while
Myron creates with sure hand true Apolline grandeur and majesty.
In close proximity to these two works we must note finally another Apollo that
proceeds from some second-rate artist 2 who copied the body of Myron's statue, and
for the head modelled himself superficially at least upon Pheidias, though he either
could not or would not adopt the special forms that lend the face its expression. The
form he employs seems to indicate that this otherwise insignificant artist belonged to
the school of Kalamis. XII. The Perseus.
A head of Perseus, which is preserved in two replicas, will be found to afford
the final external proof that the original of the Cassel Apollo and its replicas was
a work by Myron. The general characteristics of the Perseus show that its original
must have been an older work of the fifth century. Now we are acquainted through
literature with one famous statue of Perseus belonging to this period, and with one
alone,3 that of Myron. It stood on the Akropolis of Athens, and is mentioned by
Pausanias (i. 23, 7), and also by Pliny (xxxiv. 57) in his short list of the most
famous works of Myron. Since this statue was undoubtedly copied, there is every
probability that, if we can point among extant copies to a Perseus of the fifth century,
it will be to the statue of Myron.
The two extant copies are a head in the British Museum (Plate IX.), which A. S.
1 Cf. supra, p. 81 seq. ; the attributes preserved in the replicas establish this statue as undoubtedly an Apollo.
2 The statue of the Capitol mentioned above, p. 49; Overbeck, Apollon, p. 275, 4; 112, 1 ; Atlas,
PI. 20, 22.
3 It is true that an anonymous writer — probably Favorinus — in a passage quoted in the Corinthiaca of Dion
Chrys. (p. 106, R. ii. ) speaks also of a Perseus by Pythagoras; but it seems more than probable that this
rhetorician of the later Empire substituted for the name of Myron that of his rival Pythagoras, who is so often
named in conjunction with him. Cf. W. Klein, Arch. Epigr. Mitth. aus Oesterreich, 1883, p. 68. L. Urlichs
(Ueber Griech. Kunstschriftsteller, p. 48) attributes to the same rhetorician yet another confusion, between
Alkamenes and Euphranor.

198 MYRON
Murray identified and published as a Perseus,1 and its recently discovered replica in
Rome (Fig. 83).2 The interpretation as Perseus may be accepted as certain, since
Hermes, who alone could also come into question, might perhaps have worn a petasos
or a winged pilos, but not the distinctive cap covered with scales or little feathers, and
doubtless intended for the wonderful tarn cap of Hades.3 The head of Perseus on an
electrum coin of Cyzicus of the earlyfifth century4 also wears a similar close-fitting round
cap, passing at the back into great wings. On the coin the hair, though treated more
severely, escapes in short curls from under the close edge of the cap, just as in the
marble heads. One might say that the artist of the head of the statue had modelled
himself on a type of beardless Perseus wearing a round winged cap, already current in
the art of Ionia.

Fig.

Head of Perseus (Rome).

The original can be recovered from a study ot the two copies of the head. The
London replica is rather harder, though at the same time it is in some points more
accurate ; the Roman one is a hasty yet intelligent piece of work. The London
copy gives the cap more accurately, but the hair, which is much injured, seems less good ;
the Roman head represents the forms of the face with more delicacy and roundness,
and the lips arc closed, while in the other they are slightly open ; the last is likely
to be more correct.
The connexion between the Perseus and the Cassel Apollo is striking and quite
1 /. H. S. 1881, PI. IX. p. 55.
- Bull, della Comm. Arch. Comm. di Roma, xviii. (1890), Tav. 13, with Klein's text. [An interesting view
of the Brit. Mus. Perseus, both facing and profile, is given by Kalkmann, Proport. des Gesichts, p. 77.— E. S.]
3 On the top of both the Roman and Brit. Mus. heads some further object, not of marble, was attached ; ci.
Murray and Klein (loc. cit. ) Unfortunately the objects visible on the cap of the kneeling Perseus on a Cyzicene
coin of the free style (Num. Chron. 1887, PI. III. 26, p. 89), which comes next in point of time, are not clear.
4 Num. Chton. 1887, PI. III. 24, 25 (p. 88).

THE PERSEUS 1 99
undeniable. Since in the case of the Perseus tradition is in favour of Myron,1 we
gain at the same time a piece of external support for referring the Apollo to this
master. This connexion is all the more weighty and convincing in view of the differ
ence of the characters represented — in the one case a calm divinity, in the other a strong
hero. The two works cannot have been far apart from each other in point of time.
There is a special affinity between them in the treatment of the arch of the eyebrows,
with their powerful development towards the sides, and in that of the eyes themselves ;
the forehead presses heavily above the eyes, the upper lid comes prominently in front
of the lower, and the eyeball slants back.
The longer side locks of the Perseus have the same wavy form with curled ends
noticed in the Apollo. The shorter locks over the forehead and in the neck present
the unsymmetrical tangle so often noted as genuinely Myronian, except that they
seem handled with greater freedom than in those older heads after Myron that have
already been dealt with. The receding lower part of the face, with the full, slightly
opened mouth and the strongly curved lines of the lips, are all equally characteristic.
The points of difference between the Perseus and the head of Apollo serve simply to
express character. To begin with, the Perseus lacks that unusual length of the lower
part of the face ; in the Perseus this part is of normal Myronian proportions, inas
much as its length is the same as the distance between the nose and the arch of the
brows.2 The close stylistic resemblance between the Perseus and the Apollo confirms
the view that the length in the lower part of the Apollo's face was merely a means
of expressing character. The forehead is especially interesting. The modelling,
which in the head of Apollo is so slight that most of the copies have ignored it,
is here exceptionally marked.3 From either side of the root of the nose strongly
marked swellings rise upward in an oblique line in the direction of the points where
the wings are attached, reaching their highest point in the middle. The wings in this
way are, as it were, counterpoised by the powerful forehead, and the expression of
extreme energy is secured. This same end is served also by the broad, bony chin,
and the powerful muscles of the cheeks.
A long train of stylistic development undoubtedly lies between the Diskobolos
with his impassive countenance, his severe, flat hair, and the marked angle of brow and
nose, and the Perseus with his full, flowing locks, straight profile, and rounder, freer
modelling. Yet the Diskobolos contains all the essential forms of the Perseus. It
must also be remembered that the close-cropped hair of the athletes and kindred
figures by Myron must necessarily appear different from the clustering curls — not long,
it is true, but very thick — that Myron gave to the Perseus.
This Perseus must have been a work full of grandeur and energy, as can be best
realized in the London head, in which the whole neck and a piece of the breast have been
preserved. A. S. Murray, with fine insight, inferred a powerful original, and appositely
said of this head that the sculptor had in view ' a first impression not of form but of
action.' The total effect however must have been greatly enhanced by the wings, now
1 Klein now rejects the Myronian authorship, and takes refuge in the theory of a Perseus by Pythagoras — a
work which he had himself proved to be apocryphal (supra, p. 197, n. 3) ; evidently the only statue he thinks of
in connexion with Myron is the Diskobolos, and he forgets the distance that separates its head from that of the
Marsyas. Not only does tradition afford no adequate grounds for assigning a Perseus to Pythagoras, but the
statues of athletes which we have felt obliged to assign to him bear a totally different character to the heads
under discussion.
2 Namely 72 — 73 mm., twice the length of the lids from corner to corner, 36 — 37 mm. Width of mouth
55—56 mm.
3 The Perseus head on the coins from Amisos of the time of Mithradates Eupator also exhibits a remarkably
prominent forehead and large eyes (Brit. Mus. Catal. Pontus, PI. 3, 12).

200 MYRON
missing, which to judge from the attachments must have been of considerable size, and
have given a supernatural expression to the head.
According to Pausanias, Myron's Perseus on the Akropolis was represented
as to et? MeBovaav epyov elpyacrfievo<;. The deed was accomplished. The view
formerly entertained was that he was represented as in flight, running, with the
wallet («t/8to-t?) slung about him. The head teaches us now that he was standing
quietly, probably holding in his hand the severed head of the Medusa. Even without
the help of monumental evidence we must have known that these would be the
main lines of the Myronian conception ; a Perseus running through the air with his
winged shoes might prove an attractive subject for archaic sculpture, which liked to
pourtray Nike in a similar manner, but could never have been so for Myron. The
running Perseus is an archaic type, and moreover one not easily presented without the
pursuing Gorgons. On the other hand, the Perseus at rest is admirably attested as a
type in statuary of the free style. For instance, a similar statue has been repro
duced on a series of Imperial coins from Argos, which explain the violent
turn of the head to the left.1 Perseus held in his right hand the newly severed head
of the Medusa, and turned his head in horror and alarm towards the other side, to
protect himself from the petrifying glance. The kneeling Perseus too, on a vase of
Cyzicus of the free style of the fifth century,2 and on an Attic vase of the time about
4So,s displays this same motive, though with a change of sides, so that the weapon is
in the right, the head of Medusa in the left, and the head of the hero consequently
turned towards the right.
Since the turn of the head on the Argive coin coincides with the head after
Myron, we may imagine the whole statue restored in its leading features on the
lines of this coin. The hero would be represented standing on the right foot, holding
the weapon in the left hand, and the head of Medusa in his right hand stretched' to the
side. On his feet were naturally the winged shoes.
Possibly the statue of the Argive hero was not erected at Athens without
reference to Argos : it might almost be surmised that it bore some allusion to the
close relations between Athens and Argos, which began with the treaty of 461 B.C.4
In any case the statue on the Akropolis of this Argive hero, to whom the goddess
of Athens owed her Gorgon's head,6 and in whom accordingly the tie between
Argos and Athens found mythical expression, would have been a very fitting offering
at a time when politics were emphasizing this tie. To judge by the style, we must
imagine the Perseus to have been created about 450 rather than 460. The statue in
Argos, which is attested by the coins, must have been a replica of the Myronian statue,
which may have been already set up at that date, or perhaps only later.
1 Imhoof-Blumer and P. Gardner, Num. Comm. PI. I. xvii., xviii. p. 35. The type appears unchanged
from Hadrian to Septimius Severus ; a variant, which practically only reverses the sides, appears in a single
instance at Argos on a coin of Sept. Severus (ibid. PI. FF, xxii. p. 159) and at Asine on a coin of Julia Domna
(ibid. PI. GG, xxiii. p. 163). Similar type on the autonomous bronze coins oflkonium. Another ruder type, also
based on some statue, occurs on the coins of the Pontic towns of Amisos, Amastris, Sinope (Brit. Mus. Catal.
Pontus, PI. 3, 13 ; 19, 8 ; p. 99). [The type appears without variation from Hadrian to Lucius Verus, and also
under Valerian (Brit. Mus. Cat. Pelopon. p. 153); the variation on another coin of Septimius Severus, in which the
head of Perseus is not turned away, is probably due to the caprice of the copyist (supra, PI. VI. 33). I regret
that, owing to an oversight which I only noticed after PI. VI. had been photographed, this coin was reproduced
instead of one belonging to the series mentioned by Furtwangler.— E. S.]
2 Num. Chron. 1887, PI. 3, 26 : cf. supra, p. 198, n. 3.
3 Millin, Vases Feints, ii. 3. For a second vase illustrative in this connexion, cf. infra, p. 201 seq.
4 Concerning the significance of this treaty cf. latterly Robert, Hermes, 1890, 412.
5 In Athens quite old vases already attest the connexion between Athena and Perseus as the slayer of the
Medusa. The saying in Euripides. Ion. 987, which has found no expression in art, can never have been popular
in Athens.

THE PERSEUS 2°I
The motive of the Perseus, so far as we have proved it, was certainly the pattern
for the Diomede of Kresilas ; in this case also it was Myron on whom Kresilas
modelled himself. The essential character of the effective attitude and pose of the
Diomede had already been invented by Myron for his Perseus, though Kresilas no
doubt treated this motive with still greater effect.
What can have been the appearance of the head of the Medusa which Perseus
held in his hand ? This was no gorgoneion, no mask, but in all probability a solid
head worked in the round with severed neck, as it is represented also on the Argive
and Cyzicene coins of the latter half of the fifth century.1 Myron was accordingly
free from the severe mask type. On an Attic vase of the middle of the fifth century 2
the head of Medusa in the hand of Perseus is represented as that of a beautiful woman
free from any distortion. This led us to conclude {supra, p. 158) that Medusa must
have been so represented at Athens in the greater arts even previous to this vase, for
the vase-painters never invent such bold novelties for themselves. Since the vase-
painting which represents Perseus with the severed head of the Medusa in his hand
agrees with Myron's statue, and since the motive of the Perseus on the vase is the
same as that which we have recovered for the statue {i.e. Perseus standing on his
right foot with the Medusa head in his extended right hand and turning his head
in the opposite direction), we can have no doubt that the vase-painter was condi
tioned, not only in the whole figure but especially also in the Medusa, by Myron's
work ; that Myron's Medusa represented, not a hideous mask, but a beautiful head
worked in the round.
A. S. Murray, in commenting on the remarkable similarity between the profile of
the Perseus and a type of Medusa that appears on Roman gems, is apparently refer
ring to that type best preserved in the gem of Sosos,3 and less well in the cameo of
Diodotos,4 and other unsigned replicas. The similarity to the Perseus is striking
throughout, not only in the way in which the head and neck are rendered, but also in
the whole line of the profile, and especially in the forehead and eye. Courage, power,
and beauty are expressed here in the same forms as in the Perseus. It is the severed
lifeless head of the Medusa, and no mask ; it has undistorted, beautiful traits ; it is
just such a head as we have conjectured for Myron. On the one hand, the type has
nothing to do with the pathetic representation of the Medusa, common in a later age ;
on the other hand, it is an individual, powerful creation, which cannot be referred to
the gem-cutters, but must be traced to some older original. Therefore it seems to me
probable on all grounds that the original was the Medusa of Myron, held in the hand
of the Perseus on the Akropolis. And so again Myron appears as the forerunner of
Kresilas in his representation of the lifeless head of the Medusa. Myron had the
courage to cut himself entirely adrift from the traditional type of the wild-eyed
Gorgon. Kresilas also has bestowed on the actual gorgoneion mask a delicate beauty
of feature. Our inquiry into Myronian statues has enabled us to form a very different view
of the artist from that generally in vogue. He is no longer the one-sided sculptor of
athletes, interested only in violent momentary exertion, and caring nothing for the
spiritual elements of his figures. Myron has created gods and heroes too, which
1 Num. Chron. 1887, PI. 3, 26.
2 Ann. d. Inst. 1881, Tav. F.— The early Lower Italy vase, Millingen, Vases de Div. Coll. 3, which belongs
to the class characterized above, p. 108 seq., must be mentioned here, since it undoubtedly reproduces an Attic
model : it again shows the severed head and neck of beautiful type.
3 Jahrb. d. Inst. 1888, PI. 8, 18 ; p. 214. For the inscription cf. Arch. Anz. 1891, 136.
4 Jahrb. d. Inst. 1889, PI. 2, 6 ; p. 63 seq.
D D

202 MYRON
for depth of conception and power in the presentation of character surpass all
contemporary work. The ancients, however, were so lost in admiration of his two
figures representing athletes in violent movement, the Discos-thrower and the Ladas,
and of the cow with its fidelity to nature, that, by the side of these, other aspects of
his work became obscured.
If we compare Myron with Pheidias, leaving out of account such points of
difference as are due to the different times at which the artists lived, we find that the
bent of Pheidias is towards the peaceful, the mild, the reflective, the beautiful ; while
Myron's is to energy, power, character, and truth. In Pheidias, too, we find a certain
pictorial instinct, which found expression in the decorative modelling which he
cultivated to so great an extent. Myron has nothing of this, and, further, we know
him only as a maker of bronze statues in the round. In this respect also he was
followed by Kresilas, who likewise worked only in bronze, and held quite aloof from
decorative modelling in other materials.

XIII. Myronian Female Head. — The Herakles Altemps and Kindred Works. —
Asklepios in tke Uffizi.
The foregoing sketch of Myron's artistic personality can be made still more
definite by means of a few other monuments.
Among the ten figures of gods of which mention is made in our scanty literary
tradition of him, no less than three represented female deities. It would be
interesting to know what a goddess by Myron looked like, for. so strong an
individuality as his cannot have failed to impress itself upon his female types also.
Now the head of the Perseus makes it possible to trace back to Myron, conjecturally
at least, a beautiful woman's head, preserved to us in a copy, placed upon a statue that
does not belong to it, in the Giardino Boboli at Florence (Fig. 84).1 The peculiarly
energetic expression, the wide mouth and chin, and the slightly drooping corners of
the mouth present a decided affinity to the Perseus, to whom this head forms the
female counterpart — a goddess of earnest and grave character, certainly no amiable
beauty. A diadem encircles her head ; her hair flows simply back ; it is dressed low
over the forehead and is parted, though it consists of short tangled locks tolerably
freely treated and characteristic of the now familiar Myronian manner. The same is
true, too, of the eyes and their strongly marked lids. The head must, however, be
somewhat older than the Perseus, and belong to about the year 460. In particular
the part towards the temples between the eyebrows and the upper lids is somewhat
harder and less developed than in the copies of the Perseus, and betrays the same
stage of development as the Riccardi head.
Next come a few more works which appear to follow on those that have been
already described as after Myron, and to belong to the second period of the artist's
work. In the court of the Palazzo Altemps at Rome is an excellently preserved
colossal seated statue of Herakles:2 In opposition to the bearded Myronian Herakles
belonging to the artist's earlier days, the Altemps statue appears to belong to his
1 Dutschke, No. 79 ; much knocked about and weather-beaten, nose abominably restored. Arndt-
Bruckmann, Einzelvcrk. Nos. 101, 102 ; cast in the Ecole des Beaux-Arts.
2 Matz-Duhn, 123 ; Clarac, 802 F, 1988 A ; Rom. Mitth. 1889, p. 333, fig. 2 ; Kalkmann, Gesichtsprop.
p. 74, PI. I. II. K. also suggests Myron. Even the left arm with the club is antique. Only a little piece at the
top is new. The upper part of the head is broken, the nose new.

Fig. 84.— Female head in Giardino Boboli (Florence). (By permission ot Messrs. Bruckmann, Munich.)

204 MYRON
later years. The head and hair are closely akin to those of the Perseus. The body
is spare ; the pose is majestic, with a fine, energetic swing, but still somewhat hard
and angular. The hero sits upright, the left arm is raised and supports the club,
the right is stretched straight out. The original probably stood in Rome, for it was
turned to account in the time of Trajan for the representation of a Herakles seated
amid trophies of weapons and holding an akrostolion in his hand.1 It is quite
possible that this original was the very Herakles by Myron that was set up near the
Circus Maximus in aede Pompei Magni (Pliny, xxxiv. § 57).
The second statue which we must name here is the so-called Diomede of the
Palazzo Valentini at Rome,2 in which the stooping posture, as has often been
remarked, recalls the Diskobolos.3 Probably it represents a hoplitodrome, not only
on account of the likeness in the attitude to the Tiibinger bronze,4 but also on
account of the helmeted head, which has usually been taken to be foreign to the
statue, but which seems to me to belong to it. This head, which is much restored
and wrongly set,5 shows kinship to the Perseus.
Among heads belonging to the period of the Perseus may be named an athlete
in the Capitol : 6 the youth was apparently winding round his head a curious
contrivance of thongs.
On the other hand, the statue in Munich of an athlete dropping oil into his
hands,7 the beautiful Florentine athlete,8 and the types akin to these, must be kept
quite distinct from Myron, for they have no immediate connexion with any of his
works. They must belong to other Attic artists of the generation after Myron — i.e.
of the second half of the fifth century.9
Among bearded types of Myron's later period may be mentioned a beautiful
head of the Museo Chiaramonti (Fig. 85).10 The braid of hair at the back, reaching
only from ear to ear, resembles that of the Apollo. The longish hair, coming from the
crown of the head, is brought without a break towards the front, where it rolls up
into little curls and mingles with the short ringlets that surround the forehead. This
rich tangle is repeated in the beard, where the moustache alone flows in smooth
lines, as in the older heads of Myron. The eyelids are strong and marked with
a little crease in the manner so often observed in the Myronian heads. The root
of the nose is narrow. The mouth is slightly open, the expression is exceedingly
distinguished— yet it is not quietly indifferent, but full of vivacity. The head is
rather the representation of a noble hero than a portrait. I can imagine the
1 Cf. Petersen, Rom. Mitth. 1889, 331 sqq. The Altemps statue represents the Greek prototype, andis not, as
P. thinks, a replica of what he calls the Trajan type. The weapons of the latter are wanting, the arms are
different, and head and body alike betray a considerably older style. The replicas of the statuette of the Trajan
type display the usual Hellenistic-Roman character.
2 Matz-Duhn, 1097 ; Clarac, 830, 2085. A cast of the torso alone in the Ecole des Beaux-Arts (No. 3836).
3 Also interpreted as a ' Diskobolos ' by Matz-Duhn.
4 Hauser rightly points this out, Jahrb. d. Inst. 1887, p. ioi, note 24, but on account of the tree-stem he
dissents from the restoration as a hoplitodrome ; this stem however could scarcely constitute an impediment to
the attachment of a bronze shield.
5 Restored : whole neck, back, upper, and front part of helmet, nose, and a part of the lips. The serpent on
the helmet is ancient. The head is set on with far too violent a turn ; still it was slightly inclined to the left :
the relation of the marble of the head to that of the body could only be decided after a thorough cleaning.
0 Helbig, Fiihrer, No. 415 : curiously interpreted as 'Juba,' and by Helbig as 'Barbarian chief with
' negro blood.' The so-called Juba head at Athens has not the faintest likeness to this.
7 Mon. d. Inst. xi. 7. Cf. Kekule, Kopfdes Praxit. Hermes, pp. 13, 18.
8 Diitschke, Uffizi, 72 ; Rom. Mitth. 1892, p. 81 seq. 9 Cf. infra, under Polykleitos.
10 Museo Chiaram. No. 287 A. The terminal bust and the nose are modern. Mentioned as a portrait by
Schreiber, Ath. Mitth. 1883, p. 255.

BEARDED MYRONIAN HEADS 205
Erechtheus of Myron, so much admired by Pausanias (ix. 30, 1), and which no
doubt lurks somewhere among our copies, conceived somewhat after this fashion.1,
Akin is a head in the British Museum (Fig. 86),2 evidently by an artist of the
second rank, who, working about 450, comes between the earlier manner of Pheidias
and the later one of Myron,

¦fifiz±
Wm&
".;- ' <tf.'
^^H|;
w£
§1
ill PS*- ¦
mm
ft-*'" 1
1 F^~*
1 ^p ¦ '
Wb£&'-' \ ^%- -"^H
^K ¦ <$
¦¦¦'¦¦. ¦. '. :¦,.:.-':.'.;
¦B BM
Fig. 85. — Head in Museo Chiaramonti.
Another bearded type that we come to last takes us back again to the earlier
time of Myron. The beautiful Asklepios of severe style in the Uffizi at Florence
(Fig. 87) 3 was always in my opinion one of the most attractive statues of that
1 As the Erechtheus of Myron was highly esteemed by connoisseurs, to judge from the words of Pausanias,
it seems to me more probable that it was a single figure, identical with the statue of the eponymous hero in the
Agora (Paus. i. 5, 2), and not a member of the fighting group on the Akropolis (Paus. i. 27, 4), in which Eumolpos
must have looked very like his opponent. The eponymous hero, however, was certainly conceived as a quiet and
exalted figure : cf. the Erechtheus on the vase contemporary with Myron, Mon. d. Inst. x. 39. Statues of
eponymous heroes probably came into existence in the time of Kimon ; cf. Wachsmuth, Stadt Athen, i. 509 ;
Curtius, Stadtgesch. 117.— For the question of Erecththeus see latterly Kalkmann, Pausanias, p. 192, 2 ; Sauer,
Anfange d. Gruppe, p. 60.
2 Acquired with the Castellani collection 1873 ; place of discovery unknown. The term restored. At the
back of the head a piece that had been separately attached is missing. Fully life-size. A head known to me
only through the engraving in Cavaceppi (Raccolta, iii. 23) seems very like, especially in the arrangement of the
hair on the forehead.
3 Diitschke, Uffizi, No. 198; Clarac, 547, 1 152; Miiller-Wieseler, Denkm. a. K. ii. 771. Cast in the
Ecole des Beaux-Arts.
206

MYRON

collection. I formerly thought to discern in it a creation of the early Argive school,1
on grounds which will immediately be made clear ; but the Myronian character
of the head seems to me now beyond doubt. The head is unbroken, and only the
nose restored; on the other hand, the whole of the right arm with the shoulder
is modern, as well as the left hand with the serpent staff, both legs including
the whole of the right knee and the portion of the drapery that flows freely out
behind. The restorer committed the great mistake of letting the right leg appear
drawn back as in the act of walking. The ancient portions show that the disengaged

Fig. 86. — Head in the British Museum.
right foot was far more probably planted lightly on the flat of the sole next to
the supporting left.
On the drapery of the left shoulder are seen the remains of a hand ; 2 the drapery
too on the left side is less finished than elsewhere : evidently a second figure, now
lost, stood on this side. A small replica in the Palazzo Barberini (Fig. 88),3 of careless
workmanship but well preserved, puts us in perfect possession of the group. It is
Hygieia who thus lays her right hand on the left shoulder of Asklepios.

1 Sabouroff Coll. i. PI. 24. Text, p. 2.
2 Erroneously taken by Diitschke for the remains of a sculptor's point, although the mark had long been
recognized as that of a hand— viz. by the editors of the Galleria di Firenze, ser. iv. vol. i. p. 72 ; after them by Clarac
in the text to No. 1 152 ; further by Wieseler in the Denkm. a. Kunst to ii. 771, and in the official Catal. della
R. Gall. d. Uffizi, No. 209.
3 Matz-Duhn, 51 ; cf. Matz, Bull. d. Inst. 1870, 36. Phot, in the German Inst, at Rome.

Fig. 87, — Asklepios in the Uffizi (Florence).

208

MYRON

The head of the Florentine statue shows that the original of the Asklepios was
an early work of the fifth century. The question now arises whether this original also
formed a group with Hygieia. This must, I think, be negatived. It is clear, in spite
of the poor workmanship, that the Hygieia of the group 1 is based on a type that
hardly came into existence before the fourth century, and is in any case much
later than the original of the Asklepios. Certainly the composer of the little group
has lost all traces of the severe style in his Asklepios, so that one might assume
that the same thing has happened in the case of the Hygieia ; but the characteristic
style of the dress of the latter and the position of the feet are so decisive that,
in spite of the poor execution, we can safely say that her original cannot possibly
have been contemporary with the Asklepios. We gather then that the Asklepios
was composed as a single figure, and
needed no Hygieia : indeed, this figure,
who likewise stands on the left leg and
likewise holds a serpent staff in her left
hand, forms but a sorry complement to
the Asklepios. We have here a fresh
instance of the device common in the
time of the Empire, of making up into
groups figures that were originally un
connected. The little group, to judge
from the workmanship and the shape
of the plinth, was composed in the
second or third century A.D. ; the statue
in the Uffizi, to judge from the manner
in which the hair is treated, must be
ascribed to the age of the Antonines.
The Asklepios, then, must have
been originally intended as an isolated
figure. Its motive, however, remained
unaltered by adaptation to, a group ;
just as the old Argive athletes were left
unchanged when they were grouped
together as Orestes with Elektra or
Pylades,2 or as the Borghese Mars
when he was united with Venus.3 The
motive of the hands in the Asklepios of the Barberini group is so entirely in
harmony with the whole pose of the figure, that we must assume it for the
original. The god is in the act of feeding his snake. In the left hand he holds
a cake or something of the kind,4 and is on the point of breaking off a morsel,
as it seems, with his right to give it to the snake, which is raising its coils on his
left side. In the little group the creature is coiling itself round the traditional
knotted stick of Asklepios, which is leaning up against the figure in a most
improbable manner, and evidently only serves to hold up the snake. Had
the stick been part of the original composition, it must have been supported under
the left shoulder, or the left hand must have rested on it.5 Neither is however the

Fig. 88. — Asklepios and Hygieia in Palazzo Barberini (Rome).

1 The head is modern. 2 Cf. 50th Berl. Winckelmannsprogr. p. 136. 3 Cf. infra, on Venus of Milo, p. 384.
4 Zoega conjectured— probably correctly— a honey-cake ; Matz, a liver.
B This is the case in the Berlin bronze, which is reproduced in Miiller- Wieseler, Denkm. a. K. ii. 772 ; in

ASKLEPIOS 209
case. The staff was naturally indispensable in the eyes of the copyist in marble
as a support for the serpent. We can leave it out of account in the bronze original —
(the treatment of the hair of the Florentine head clearly points to bronze). There
is nothing in the attitude of the god to lead us to expect the existence of the staff:
he does not support himself upon one, but stands firm on the left leg, leaving the
right free, but with the whole sole down — (the Barberini group shows that the
Florentine statue should have been thus restored); he holds the cake for the snake
in the left hand, which is extended horizontally, and moves the right in the same
direction. Snakes lifting themselves up without any support were, as is well known,
a favourite theme of quite early Greek art, both in decorative1 work and in the
round, as for instance on the old Spartan reliefs, notably on one where a
snake is raising itself up in front of a youth who is offering2 a fruit to it. On
a Thessalian coin of the fifth century Asklepios is feeding a serpent that appears
upright in front of him.3 In a statuette from Epidauros4 which goes back to some
fifth-century original the snake is likewise not wreathed about a stick. The bronze
original of our statue must accordingly have had at its side a great snake which,
resting on its coils (compare the snake at the side of the Parthenos), lifted its
head to the level of the god's hand.
Our marble copies obviously represent Asklepios. The addition of Hygieia
makes this plain. It does not however follow as a matter of course that the original
belonging to the first half of the fifth century also pourtrayed Asklepios. A
bearded divinity, feeding a snake, is a figure appropriate to more than one
cult, and to whom different names might be assigned. In the Graeco-Roman
period this type became confined to Asklepios, while the other kindred deities of
health, from whose midst he sprang, retained only a local significance. Our statue
might very well have represented originally some Athenian divinity, akin to Asklepios,
associated with the serpent, a divinity significant only in some local cult. Such were,
for example, at Athens the popular Hero Iatros and the Hero Alkon,5 at Oropos
Amphiaraus, and at Lebadeia Trophonios ; the two last were, we know, repre
sented in exactly the same way as Asklepios.6 The reason for this likeness lies, not
in an extension of the Asklepios type to the others, but in the fact that this type
had originally a wider signification. It might even be that our statue represents Zeus
Meilichios^who was intimately associated with the snake, and was a genuine Chthonian
divinity of benign character.7
In this statue the personality of such a god is very effectively embodied. The
hair combed down from the crown of the head on the forehead and framing the face,
together with the long beard, give to the face an expression of solemn melancholy. In
the turn and inclination of the head, however, and in the kindly mouth, we read
Roscher's Lexikon, i. 636. (Friederichs, Kl. Kunst. u. Industrie, No. 1846.) This bronze, like its replica in Berlin
(Friederichs, 1846a), belongs in design and technique to the 16 — 17th century: the design seems to have been
suggested by the Florentine statue : the artist skilfully allows both hands to rest on the staff.
1 Cf. Olympia, vol. iv. Die Bronzen, No. 906 seq. Text, p. 145 seq.
2 Ath. Mitth. 1877, PI. 25 ; Skulpt. in Berlin, No. 732. Cf. also the great Meilichios snake, ibid.
No. 723, and various other reliefs, e.g. Le Bas, Voy. Arch., Mon. fig. 39 : also the vase 'Eip-riix. apx- 1890, PI. 7.
3 Brit. Mus. Catal. Thessaly, PI. 5, 9. 4 Kabbadias, yXvisrd ipytt. 270.
6 Cf. Deneken in Roscher's Lexikon, i. 2483 ; Sabouroff Coll. i. Introd. p. 20.
6 For Trophonios see Paus. ix. 39, 3,4; for Amphiaraos, Deneken, loc. cit. 2588, and AeXriov dpx. 1891,
p. 89, No. 23. It is far more likely that one of these two and not Asklepios is represented on the Boeotian
vase, 'Etpri/j.. apx. 1890, PI. 7.
7 Cf. Sabouroff Coll. i. Introd. p. 22. A votive relief of the fourth century in the Peiraieus represents Zeus
Meilichios (name inscribed) enthroned with a horn of plenty and a bowl ; the sacrificial victim which is being
offered to him is a pig.
K E

210 MYRON
gentleness and goodness. The action is above all characteristic, in that the god pays
no heed to himself, but feeds his snake, seemingly wholly absorbed in the occupation.
In this we see the tenderness and care he bestows on all that trust in him ; the
nature of the healing god is thus accentuated, and at the same time the snake, which
is his attribute, is brought into close and actual union with him. Only an artist of
exceptional originality can have created this characteristic figure, which differs from
most of the statues of gods of the time in the fact that the god is represented as
engaged in some action. We believe that Myron may have been this artist.
Let us examine the figure more closely. The pose is evidently borrowed from
the canon of Hagelaidas ; the left supporting leg with the right leg placed close to it,
the drooping left shoulder, and the inclination of the head to the same side, the left
forearm horizontally extended and holding some object, the upper part of the arm
lying close to the body, and the broad shoulders, are all traits that recall this canon.
At the same time we can also observe great differences ; quiet and dull repose are
replaced by action and concentrated attention. The head and the upper part of the
body are not simply inclined to the left side, but turned that way ; and the right arm
does not hang down but is in active movement — in a word, everything combines to
represent a definite action.
The fidelity to the canon of Hagelaidas, together with the bold departure from it in
certain respects, alike point to Myron. The head decides the question. The forehead,
the eyes with their rather pronounced lids, the full mouth with the slightly parted
lips, the tangled profusion of curls, and the shape of the head, are all distinctly
Myronian. For criticism of detail, two heads in the British Museum (Fig. 89)1 and in the
Louvre 2 respectively are of weight : the original which they reproduce 3 must have
resembled our Asklepios so closely that it almost seems as if the two heads were
merely variants, due to the hand of the copyist.4 Both are quite erect ; the one is
furnished with a terminal bust, and the other probably terminated in the same way.
It is conceivable that both reproduce our 'Asklepios' head, but altered and posed
erect in order to suit the exigencies of a terminal bust. Still, if a different original
did actually underlie them, it certainly proceeded from the same artist and had
substantially the same significance as our ' Asklepios.'
In both heads the stylistic treatment of the hair and of the moustache produces
an impression of far greater fidelity to the original than in the Florentine statue, where
the finicking character of the work reveals the mannerism of the age of the Antonines.
Both the heads also display to perfection Myron's modelling of the forehead : the
eyelids are pronounced, and the full mobile mouth allows the teeth to show as in the
Cassel Apollo.5 The delicate root of the nose and the line of the bridge of the nose —
preserved in the London copy — are all quite in Myron's manner. Further, in the last-
named head the hair of the eyebrows is indicated ; this detail, which we also noted in the
bearded Herakles (p. 179), may possibly be referred to the original. The treatment
1 Spec, of Anc. Sculpt, i. 16 ; Anc. Marbles, ii. 29 ; Brunn-Bruckmann, Denkm. No. 229. Term with
head, preserved without any restoration.
- Louvre, No. 2055, belong to the Campana Collection, unfortunately very unfavourably placed at present in
the Salle du Mars Borghese. The head is broken at the point where it joins the neck; term restored:
otherwise only the nose is modern. The beard and the locks of hair about the middle of the forehead are much
damaged ; workmanship good and careful.
3 The heads are replicas, the hair agreeing lock for lock. There is a third very poor replica in the Pal.
Orlandini in Florence. Dtitschke, ii. 503 ; Arndt-Bruckmann, Einzelv. 240, 241.
4 The chief difference is in the hair at the sides falling over the ears.
5 The teeth are plain in the Paris head.

F.g. 8q. — Terminal bust in the British Museum.

212 MYRON
of the moustache with its long flowing lines that pass into the beard, and the way in
which the locks of hair are detached from the ground, recall the same bearded Herakles,
the two works probably belonging to the same period of the artist's career.
The hair falls over the middle of the forehead in a rich unstudied mass, which
contributes in no slight degree to the sombre expression. A simple fillet cutting deep
into the locks is the sole adornment. In spite of the severe style the head is perfect in
its expression of character. The custom that prevailed in later days of emphasizing
the under-world divinities x by hair falling low over the forehead is merely a develop
ment of what was begun in these heads.
The drapery of the ' Asklepios ' is arranged in a simple and appropriate manner,
without regard to beautiful or effective motives. The piece brought down over the
shoulder may be compared with the corresponding arrangement in the Riccardi bust.
We cannot enter into details, for the copyist of the statue was in all probability as
inexact in the drapery as in the hair ; he doubtless tried to add richness to the broad,
severe traits of the original by the introduction of lesser intermediate folds ; we must
probably subtract considerably from the number of folds, and conceive the main
features of the drapery to have been treated more simply and severely.2
Thus this statue of a god or hero of healing appears to be a copy after an earlier
work by Myron, in which however he had already unfolded to the full the powers of
characterization which we have learnt to admire in his later figures of gods and
heroes ; among these the Florence Asklepios, in which a definite action is combined
with the turn of the upper torso, deserves a special place.

XIV. The Munich Zeus and the first Argive School.
Another statue of extreme interest must be considered next, as much for its points
of difference as of resemblance. We have seen that Myron, like Pheidias in his
younger days, often employed the canon of Hagelaidas. It is therefore well to become
acquainted at this point with a statue almost contemporary with the Myronian works,
and which not only exhibits this old canon but seems to have been created in Argos
itself. I refer to the familiar statue in Munich (Figs. 90, 91)3 which Brunn and
Kekule refer to Polykleitos, but which differs widely from the undoubted
Polykleitan works, and is plainly considerably earlier. It is quite clear that the
type of Hagelaidas underlies this conception, and has only been so far modified
that the right leg is drawn back in the walking motive. For the rest there
is the same characteristic hardness so familiar in works of the old type, the same
disproportionate width of the breast in comparison with the hips, and the same angular
shoulders well drawn back.4 The attitude of repose with the weight on the left foot,
The type afterwards so frequently used for Sarapis must have come into existence by the end of the fifth
century. A very well preserved and significant statue of the Louvre (Frohner, 33 ; Bouillon, iii. stat. I, 3 ; Clarac,
311, 681)- represents Zeus (eagle and thunderbolt are in part ancient) quite in the style of the fifth century:
the head has the wisps of hair falling over the forehead. It is probably a Zeus Meilichios, and its artist should
perhaps be looked for in the school of Alkamenes.
The Cassel copy of the Lemnia is a clear example (supra, p. 5 seq. ) of the extent to which copyists
remodelled the simple draperies of more ancient statues at times.
3 Brunn, Glypt. 5, No. 160 ; Friederichs- Wolters, Gipsabg. 480; Kekule, Jahrb. d. Inst. iii. 1888, p. 37, PI. I ;
L. v. Urlichs, Belli: z. Gesch. d. Glypt. p. 16; Brunn-Bruckmann, Denkm. No. 122. No replicas are
known to me.
Cf. 50th Berl. ]Vinckelmannst>rogr. pp. 138, 140, 143, 146.

Fig. 90, — Statue in Munich.

214 MYRON
the droop of the left shoulder, the inclination of the head to the same side, the hang-
ino- rio-ht arm, and the left forearm extended horizontally— all these are familiar and
unmistakable traits of the old scheme. The head shows also that the artist was com
pletely under the influence of Hagelaidas. This is especially evident in the profile :
the whole outline of the head, the line of the neck, the fillet that lies flat on the
hair, its shape, and the very way in which the hair meets it ; further, the form of the short
thick curls themselves— all these accord so remarkably with the best copies of the
head of the figure by Stephanos (Fig. 92),1 and on the other hand present such
marked contrasts to the other works of the same period, such as those of Myron, that
there can be no doubt about the tendency of the artist. He must stand in direct
relation to the artist of the Stephanos figure, whom we conjecture to have been
Hagelaidas. As compared however with the style which we must assume for Hage
laidas, the artist of the Munich statue is considerably less hard ; he gives fuller, softer
flesh to the body ; he indicates the veins and sets the eyes more deeply ; he opens
the mouth and makes the lower part of the forehead project ; in a word, his system
of forms shows a development equal to Myron's. Yet one has only to compare the
modelling of the forehead or the treatment of hair and beard to find that he lacks
all that character of individuality, that charm of real life, which stamps the Myronian
works. The inclination to recognize in the Munich statue the manner of Polykleitos
originates in a perception, which I fully share, that all the Argive works of the fifth
century bear an inner relation to one another. In the head of the Munich statue,
for instance, the first aim, as with Polykleitos, is to secure pure bright beauty of
form, not, as with Myron, the expression of individual character. I myself some years
ago conjectured this statue to be an early work of Polykleitos as pupil of Hagelaidas,
and found a support for my theory in the oldest extant basis of Polykleitos, that of
Kyniskos in Olympia, on which the traces of the feet point to a figure standing much
like the Munich statue.2 But the stylistic divergences from Polykleitos are too
pronounced, and the whole system of forms so strongly contradicts this artist's
personal style,3 that I now consider my former theory quite untenable. On the
contrary, the statue must belong to the generation of Argive artists who intervened
between Hagelaidas and Polykleitos.
It is often assumed — on the ground of the familiar passage in Pliny — that it was
Polykleitos who introduced the walking motive. This opinion is however erroneous.
Polykleitos borrowed this motive for his canon, and thereby won for it widespread favour
and popularity ; but he did not invent it. The merit of the invention belongs to the
Argive artists of the generation that preceded him, about the 80th Olympiad. Any
doubts as to this point owing to the Munich statue being the work of a Roman
copyist can be disposed of with the help of an original monument. This is a block of
marble from the dedicatory offering of Smikythos in Olympia ; to judge from the
perfectly plain footmarks, it once supported a statue which stood in just the same
attitude as the Munich statue.4 The period of the offering is the same as that which
1 Specially with the head in the Lateran, No. 356 (Fig. 92), and with the Orestes in the Louvre group : the
last may be studied to advantage in the cast of the Ecole des Beaux-Arts, No. 6757 ; 50th Berl. Winckel
mannsprogr. pp. 136, 138. The Capranesi head in the British Museum (ibid. p. 148, note 82) may also be
compared. The face measurements of the Munich figure come close to those off the Stephanos figure.
Cf. Kalkmann, Gesichtsproport. p. 60.
2 Cf. infra, under Polykleitos, p. 249 seq. 3 For instance, in the treatment of abdomen and pubes.
4 It is the block with the largest fragment of the dedicatory inscription. Lowy, /. G. B. No. 31 a ;
deep depression for the back part of the left foot, which stood firm. Close by, on the same level as the heel
of the left foot, is the hollow for the ball of the right foot. On the Munich statue too the ball of the

.^i

'ARGIVE STATUE 215
must be ascribed to our statue, about the 80th Olympiad (soon after B.C. 466) ; and
its artist, Dionysios of Argos, belongs to the school of art to which the Munich statue
must be ascribed.
The first introduction of the walking motive was clearly a significant departure
on the part of some individual artist. In the period preceding the basis of the work
of Dionysios and the Munich statue there is no trace of it, and after them, even in the
Argive school, the older attitude of repose remained long in favour. In the works of
Myron, and above all in the older Attic works, the walking motive is wholly unknown.
We may, then, conjecture that it was Dionysios of Argos or his contemporary
Glaukos, the artists of the offering of Smikythos, who first introduced this innovation.
And to them or to an artist of the same standing must also be ascribed the original of
the Munich statue. This statue, while still keeping in everything else to the old type
of Hagelaidas, shows what a powerful new impetus was derived from the introduction
of the walking motive ; not but what the hard angular forms, in which this motive
first appears, are separated by a long distance from the perfect rounded harmonies
which Polykleitos imparted to it.
We have still to consider an interesting find from Olympia, which shows that the
statuary type of the master of the Munich figure was employed in this centre of
Peloponnesian art activity, as we can prove to have been the case with the canon of
Hagelaidas.1 A colossal torso from the Olympia excavations 2 displays so close
a resemblance to the Munich statue that it must be referred to the same artist.
Unfortunately it is not an original, but evidently only a late copy, executed in Pentelic
marble, possibly to replace a stolen bronze original, or possibly it was a new dedica
tion, copied from an older work. The torso accords with the Munich statue in all but
trifling points ;3 in it too the shoulders are disproportionately wide,4 yet here again,
as compared with the canon of Hagelaidas, the forms display the same full, round,
fleshy character ; here too the veins on the arms are indicated.
The torso of Olympia by reason of its size is in all probability to be referred to a
statue of Zeus ; indeed, from the place where it was found, one might hazard the con
jecture that it was a Zeus named by Pausanias without mention of the material, which
had no inscription, but which was described as a dedication of Mummius ; 5 a statue
of this kind might very well be copied from an earlier work. The Munich statue
right foot is on a level with the heel of the left foot. The block belonged to the offering of Smikythos erected
on the north side of the temple, assigned by Pausanias (v. 26, 2) to Dionysios. There are ancient footmarks on
two other blocks of this offering ; in this case however both feet were resting with full sole. From these it is
apparent that the figures were considerably less than life-size (length of the footmark 16 cm. ) The Roman
footmarks on the under side of two of these blocks are quite different.
1 Cf. in 50th Berl. Winckelmannsprogr. p. 146 seq.
2 Perfectly nude torso in the museum at Olympia: the neck is altogether wanting, as also the legs fiom
a little above the middle of the thigh ; a considerable part of the left upper arm is preserved ; smaller portion
of right upper arm. From the hollow of the neck to the upper edge of the pubes o'75, to the navel 0-555.
Distance between the nipples 0-40. Greatest breadth in the hips 0-49, in the shoulders 075. The statue
was therefore considerably larger than the Munich one (where the distance between the nipples measures 0-32).
3 The left upper arm of the torso issomewhat drawn back instead of forward, as in the Munich statue. Further,
the left shoulder of the torso droops less : the forms are altogether somewhat less meagre. The Munich statue is
considerably better in technique, and seems a more careful copy than the Olympian torso.
4 The pubes is sharply defined as in archaic work : the locks of hair are treated in the stiff style, but they
are only lightly indicated, and not worked out.
5 The torso was found on the 12th of March 1880, between the western terrace walls and the western Altis
walls, a little south of the southern line of the palaistra, lying free (i.e. not built into any construction) below
the green deposit that came down from the Heraion, and according to all appearances in its original site.
Treu, to whom I am indebted for the above information, is reminded of the Zeus of Mummius, which stood
here near the west wall of the Altis facing towards the west (Paus. v. 24, 8) : this identification seems to me
very probable,

Fig. 91. — Head o. statue in Munich.

ARGIVE STATUE

217

is interpreted by Kekule as Zeus, bearing in his left hand the eagle, in his right the
lightning. This interpretation satisfactorily explains the position of the arms, and
also the whole majestic, kingly personality of the figure, and is now further cor
roborated by the kindred Olympian torso.1 Yet how comes Zeus to have quite close-
cropped hair ? We certainly can only then interpret this figure as Zeus, when this
strange and wholly unwonted mode of treatment is attested for the epoch and school to
which our statue belongs. And this actually appears to be the case. It is true that
among the numerous representations of Zeus on monuments of every kind I can only
point to two examples with the same close-cropped hair, but these are sufficient.
Shortish hair is quite common in representations of Zeus : on Attic vases the type

Fig. 92. — Replica of the head of the statue by Stephanos.

appears first in the period about 403 B.C. (cf. supra, p. 42), and at a later date becomes
very popular ; quite in accordance with these Attic vases is the type on the familiar
Eleian coins of the fourth century.2 But this type of hair, in which the locks though
short are yet flowing, does not help us where the question is of hair close-cropped
like an athlete's ; besides, it belongs to a considerably later period than that of our
statue. On the other hand, an isolated example of the type in question occurs in
archaic vase-painting in the Zeus engaged in combat with the Giants.3 On this
follows a famous coin-type of Elis 4 (PI. VI. 22), which however was only struck
1 It must however be mentioned that a coin of Nicaea (Brit. Mus. Catal. Pontus, PI. 32, 13 ; p. 158) struck
under Commodus represents Hephaistos with a hammer in the right hand and a bar of metal in the left, in a
scheme closely resembling that of the Munich statue ; the original must have been a work of the same school.
2 Head, Hist. Num. p. 355, fig. 234 ; Guide, PI. 23, 33. 3 Overbeck, Kunstmyth. Atlas, PI. 4, 6.
4 Only'two specimens of the coin are at present known : (a) London (Head, Hist. Num. p. 354, fig. 230 ;
Guide, PI. 14, 29 ; P. Gardner, Types, PI. 8, 6 ; Stephani, Compte Rendu for 1876, p. 224, PI. No. 5). (b) Beriin
F F

2l8 MYRON
during quite a short period, and may probably be dated circa B.C. 420 ; for it is the
counterpart to the beautiful type of Hera, which according to a plausible conjecture
of Percy Gardner's is to be referred to the League contracted in that year between
Elis and Argos (PI. VI. 23).1 In the head of Hera it is thought, and probably
correctly, that we have the Polykleitan Hera of Argos ; the head of Zeus too must
have been based on some famous work of art, but not on one of the period of 420,
for the whole type — with the slanting profile, the massive beard, and the ear set rather
high up — seems to me to point pretty clearly to the- fact that we have on the coin a
work of the severer style rendered with the freedom natural to the fifth century ;
the obvious suggestion is that this was an Argive work, dedicated by the newly
formed League, and highly prized by them.
The next and last instance of a short-haired head of Zeus occurs on the coins
of Syracuse 2 and of the Italian Lokri (PI. VI. 20).3 It is merely a type which was
stamped after B.C. 345 4 on a portion of the coinage of these two cities (which were
intimately connected both politically and commercially), after the expulsion of the
tyrant Dionysios from both, on the occasion of the restoration of the Democracy.
This type is, however, evidently the free reproduction of an older work of the fifth
century,5 for it has nothing in common with the types that reproduce the spirit and
taste of the epoch circa B.C. 345. The inscription on the Syracusan specimens, which
characterizes Zeus as 'EXevOepios, leaves no room for doubting what this older work
was : it was the colossal figure of Zeus Eleutherios — of Zeus the Deliverer — which
the Syracusans had set up as an enduring memorial of that earlier expulsion of a
tyrant and democratic revolution in 466.® To this pledge of freedom the people had
recourse on the similar occasion in the fourth century, when the festival and the great
sacrifices in front of the statue were once more revived. Now it is an old conjecture
of Haverkamp, approved by Eckhel,7 that a reproduction of this Syracusan Zeus
Eleutherios has been preserved in a coin-type of the two consuls of the stormy year
49 B.C. — L. Cornelius Lentulus and C. Claudius Marcellus.8 On these coins the type is
associated withadesign that is purely Sicilian — namely, with the triquetra, which appears
on the reverse in its essentially Sicilian form.9 This must evidently have been selected
with reference to the Consul Marcellus and his glorious ancestor the Conqueror ot
Sicily.10 The conjecture has therefore everything in its favour. The figure of Zeus in
(still unpublished) : from a different die to the London specimen, yet agreeing with it almost exactly ; the hair is
somewhat better preserved. The little curls are very like the hair on the Munich statue ; it clearly imitates a
work of this style. The moustache streams on either side over the beard.
1 P. Gardner in the Brit. Mus. Catal. Peloponn. p. 36 sqq. PI. 12, 11.
2 In Syracuse only in bronze : Head, Hist. Num. p. 157 ; Numism. Chron. 1874, PI. 7, 10, II ; 1876, PI. 2,
6 ; 3> 17 > °f- Overbeck, Zeus, p. 213. On the electrum and silver coins of the same period is represented an
ordinary free type of Zeus, corresponding to the taste of the time.
3 In Lokri in both silver and bronze : Head, Hist. Num. p. 86 seq. ; Guide, PI. 25, 21 ; Gardner, Types, PI.
5, 14; Sallet's Numism. Ztsch. xvii. PI. 1, 7 ; Overbeck, Zeus, p. 101. The finest silver specimen appears to
be that in the Gotha collection. 4 For the date cf_ Head i/oc cit j
5 This is also Von Sallet's view, whom I was able to consult on the point.
« Diodor. xi. 72, 2. Cf. Kekule in Jahrb. d. Inst. 1888, p. 43 ; Busolt, Griech. Gesch. ii. 292.
7 Doctrina Num. v. 182.
8 Good reproduction in Head, Guide, PI. 66, 16. Cf. Babelon, Monn. de la Riiubl. Rom. i. 424 ; Jahrb. d.
Inst. 1888, p. 43.
9 We find too as reverse the curly head of a youth, closely related to the Zeus Hellanios of the Syracusan
coins (Overbeck, Zeus, Miinzt. 3, 1, 2 ; p. 196), and therefore possibly connected with it : this is rendered
however less probable by reason of the fact that a similar type— though at the same time less akin to the
Syracusan— appears elsewhere on the coins of Roman families, where it is explained as Vejovis.
w So Eckhel (loc. cit.), who also remarks— and rightly— that these coins need not on this account have been
struck in Sicily, which would be unlikely after the events of the year 49 ; this supposition has been made into a
fact by many of the more recent authorities.

ARGIVE STATUE 219
question — and this is what is of special interest to us — displays a decided likeness
to the Munich statue. The design on the coin, in spite of the careless execution,
reproduces as though intentionally just the same pose, particularly in the drooping
left shoulder and the disproportionate width of breast. The lower part of the right
leg seems to have been conceived as drawn back, though this is not very obvious, by
reason of the front view which has been adopted. The sole divergence in the pose —
the right hand stretched out sideways — may be explained from the impossibility of
representing the thunderbolt from the front. The head is so slightly and carelessly
indicated that we must not lay too much stress on the fact that it seems to wear the
usual crown of hair. Thus it seems permissible to conjecture that the Munich statue
is actually a copy of the Zeus Eleutherios erected in Syracuse in 466. This date
would exactly accord with the style of the statue, and would not conflict with what
we have ascertained about its school, for the Argive artists, Hagelaidas among the
first, seem to have been employed on many occasions for the West, and the artists
who worked for Smikythos might well have received commissions from Syracuse
as well. It is quite in keeping with the spirit of the Argive artists that they should have
wholly divested the king of the gods of his characteristic adornment of hair, and
formed him after the same pattern of manly beauty as the victorious athletes ;
for, in their search after pure beauty of form, they got further away from the power of
characterization .
Still this type could only have a transitory duration. The characteristic creations
which Attic art was producing about this time soon became the prevailing types in the
representation of the gods. We have on several occasions tried to show that in this
Myron played a leading part ; we may recall by way of contrast the Zeus that we
ascribed conjecturally to him ; there we found a deep understanding of the nature
of the god represented ; it was the veritable Zeus, while in the statue we have just
been studying we seem merely to behold a beautiful man.
With this contrast we close for the time our inquiry concerning Myron, and in
the following chapter turn to Polykleitos himself, who follows immediately upon
the stage of art with which we have just become acquainted.

POLYKLEITOS

POLYKLEITOS

I. Historical and Epigraphical Evidence.

T

'HE preceding chapters have already taught us some
thing about Polykleitos, which it is well to begin
by recapitulating. As regards his date, we saw
(p. 130) that there was no reason for doubting the evidence
of Plato, who makes Polykleitos a contemporary of Pheidias.
We next obtained B.C. 440 as the approximate date of his
Amazon, a work which must naturally have been executed
at a time when he was already an artist of acknowledged
position. The famous Doryphoros must belong to a still
earlier period, partly because it anticipates the Amazon in
many points, and partly because of the flat treatment of the
hair. These considerations show that the artistic activity of
Polykleitos began at least as early as B.C. 450, a date at
which the walking motive must have been current in Argive
art, since, as we noted before, the Argive artist Dionysios
employed it as early as about B.C. 460. We saw too (p. 196)
that Polykleitos cannot have been, as ancient tradition had
it, actually the pupil of Hagelaidas, for a whole generation of
artists intervenes between him and the founder of the school.
The next certain date in the life of Polykleitos is B.C. 420 (01. 90), when he
made the gold and ivory Hera for the new temple at Argos, an event by which
his whole career was dated in classical times (Plin. xxxiv. § 49). We can obtain
a fair notion of the artist's style at this period ; for it seems certain that his Hera
was the model for the majestic head of the goddess which makes its appearance
on the coinage of both Argos and of Elis at the very time when the statue of
Polykleitos was set up (Plate VI. 23).1
1 See supra, p. 218. It is scarcely necessary to remind the reader that there is no connexion what
soever between the Hera of Polykleitos and either the Farnese ' Hera ' (Conze, Beitrdge x,. Gesch. d. Grie.
Plastik, p. I, and others) or the girl's head recently discovered during the American excavations at the Heraion
of Argos {see Arch. Stud. H. Brunn dargebr. 1893). As to the Farnese head, the affinities it presents to the
Artemis of the Aktaion metope from the Heraion at Selinos, and the further affinities between these Selinuntian
metopes and the school of Kritios and Nesiotes, make me refer this ' Hera ' — or more correctly this Artemis  to
the latter school.

224 POLYKLEITOS
Nothing is known as to the date of the death of Polykleitos ; he may have lived
to extreme old age, and for all we know he may have been working up to the end
of the fifth century. None of the works, however, known to be undoubtedly by him can
be dated later than B.C. 420. The Zeus Meilichios, which, according to an anecdote
in Pausanias (ii. 20, 1), should be dated after 418, is really not a dated work at all, for
the anecdote is only an explanation of the name Meilichios, and therefore has no
chronological value.1 It is not altogether impossible that the tripod with the figure
of Aphrodite consecrated at Amyklai after the battle of Aigospotamoi (B.C. 405) was
a work of the elder Polykleitos, though it is highly improbable ; for as Polykleitos
passed on to his pupils the work on the more splendid votive offerings, put up in
Delphi after the same battle, he would hardly have undertaken a comparatively
unimportant commission for obscure Amyklai. It is more likely that the tripod was
by the younger Polykleitos.2
A number of bases inscribed with the name Polykleitos and belonging to statues
of athletes have been found in Olympia ; but it is by no means always easy to
decide to which of the two artists each basis should be ascribed. The basis of the
Kyniskos has undoubtedly been correctly assigned to the elder artist (Lowy, / G. B.
50). It is not so much the archaic character of the writing that proves this, for
Arkadia was singularly conservative in this respect,3 as the way in which the words
are distributed around the upper surface of the base. Next comes the basis of the
Pythokles (Lowy, I. G. B. 91) on which the Argive Lambda occurs. Now on the
well-known Polystrata relief from Argos — a piece of sculpture which from its style
should be dated before rather than after the Peloponnesian War 4 — this same Argive
Lambda appears together with the Ionic H. Thus it is safe to date the Pythokles
as early as 430, since the Argive alphabet was falling into disuse at a still earlier
period. The basis of the Xenokles (Lowy, /. G. B. 90) follows closely upon those just
mentioned, for, though the inscription is entirely Ionic, there are still all sorts of archaic
elements about it. It should probably be dated about 420 ; for, as Ionic letters were
already being introduced into Argos even before the Peloponnesian War, it is quite
likely that by 420 the Ionic alphabet was fully established. Both the Pythokles
and the Xenokles may therefore be ascribed to the elder Polykleitos. On the
other hand, the Aristion inscription (/ G. B. 92) has a totally different character ; 5
1 Cf. Robert, Arch. Mdrchen, p. 102.
2 The fact that the signatures of Polykleitos and of Lysippos (Lowy, I. G. B. No. 93. See also Robert, Arch.
Mdrchen, p. 103^?., and E. Preuner in Bonner Studien, p. 217 seq.) appear side by side cut inidentical characters
on one block of stone (found at Thebes) which supported two statues by these two artists, has given rise to the
theory that this younger Polykleitos was a contemporary of Lysippos. But the character of the epigraphy shows
that the Theban basis belongs to the period succeeding 316, when the city was rebuilt after its destruction by
Alexander. It is evident that, as Preuner (loc. cit. p. 220) has already hinted, two statues which were originally
unconnected were now set up anew side by side and inscribed afresh. There are many examples of such renewals
of artists' signatures. The basis is therefore no guide for recovering the date of the younger Polykleitos, and all
the theories lately advanced by Robert (loc. cit. ) become untenable. Rather must we admit that works handed
down with the name Polykleitos and belonging to the end of the fifth century may just as well be by the elder as
by the younger artist. The view propounded above accounts for the Boeotian form £ir6eiirtv that occurs on the
basis. I note that in the latest publication of the inscriptions in C. /. Gr. Septentr. i. 2532, 2533, Dittenberger
dates them on epigraphical grounds after 316. His further improbable conclusion, that the Polykleitos named
here is an unknown artist, the third of the name, need not be discussed after what has been said above.
3 See Kirchhoff, Studien, 4th ed. p. 159.
4 This is also the opinion given in the Beschr. d. ant. Skulpt. in Berlin, No. 682, where however the illustra
tion reproduces the inscription incorrectly with an E instead of the H which is clearly visible on the stone in the
aviBriKe. Kirchhoff (Studien, 4th ed. p. too) assumed that the relief belonged to the period after the war, on the
ground of the style, but it is just this that points to an earlier date.
5 It is also carelessly executed and not deeply graven.

INSCRIBED BASES 225
it belongs evidently to the fourth century, and must be assigned to the younger
artist. Polykleitos was an Argive. Plato, our earliest authority in the matter, calls him
'Apyelos in a dialogue (the Protagoras) written in any case shortly after the death of
the artist. To Plato, Argos was the home of Polykleitos in the same sense that
Athens was that of Pheidias, to whom he applies the adjective 'AOrjvalos immedi
ately after calling Polykleitos an Argive. He is evidently citing the two most
celebrated artists and leaders of the two schools x most in renown at the time in
which the dialogue is supposed to have taken place.
In the Pythokles inscription, which, as we saw, there is good reason for assigning
to the elder artist, Polykleitos expressly calls himself 'Apyelos. It is true, however, that
this epithet only occurs in the restored inscription of the first century B.C., and it is
not quite certain if it was part of the original. The other Polykleitan inscriptions
mention no nationality, probably because the artist was so famous and well known
at Olympia. The younger Polykleitos, to whom the Aristion base must belong, signs
simply Ho\v/c\et,To<; e-n-oirjae, without any epithet. He does this probably with
intention, for it was not to his interest to be distinguished from his famous elder
namesake, who, as the Xenokles basis shows, signed his later works in this simple
manner. Pausanias in various passages calls both elder and younger Polykleitos
simply 'Apyelos. Moreover, Naukydes, whom Pausanias calls the brother of
Polykleitos, meaning of course the younger, is also 'Apyeios, as is proved by Pausanias
(vi. 1, 3) 2 and by his own signature on a basis at Athens. It is only the third brother,
Daidalos, who in an inscription calls himself a Sikyonian (Lowy, /. G. B. 89), as he
is also styled by Pausanias (vi. 2, 8, etc.) He evidently migrated to Sikyon. From
Sikyon also came two of the artists employed on the great Delphic votive offering —
Kanachos, a pupil of Polykleitos, and Alypos, a pupil of Naukydes. Among other
later members of the Polykleitan school we must also reckon Kleon of Sikyon,
who belongs to the fourth century. These facts seem to show that a branch of the
Polykleitan school in Argos was established in Sikyon,3 probably by Daidalos, about
the end of the fifth century. In the fourth century this Sikyonian offshoot gradually
overshadowed the parent school, and Lysippos, its greatest outcome, eventually cast
Argos and her artists completely into the shade. Its school of painting in the fourth
century would in itself have made Sikyon a leading art-centre ; but in the fifth matters
had been very different, for sculpture was then in the hands of Aristokles and his
pupils, who were absolutely insignificant as compared to the Argive school founded
by Hagelaidas and developed by Polykleitos. The assertion, found only in Pliny
(xxxiv. 55), that Polykleitos was a native of Sikyon, must in all probability be due
to that same historian of art whose Lysippian and Sikyonian bias can be so clearly
detected at various points of Pliny's borrowed narrative (cf. supra, p. 171 seq.) He
was guilty of a sort of anachronism in transferring Polykleitos to Sikyon. There is
however some excuse for his error in the later migration of the Polykleitan school
to Sikyon, and in the eventual exclusive leadership exerted by the Sikyonian school
in the Peloponnesos.4
1 This passage does not necessarily imply (as Robert assumes, Arch. Mdrchen, p. 101) that Polykleitos was
resident in Athens.
2 The corrupt MSBavos (Paus. ii. 22, 7) must for the present remain a mystery. Robert's last suggestion,
vfdrepos (Hermes, 1888, 429), is based on his untenable hypothesis concerning the family of Patrokles (Arch.
Mdrchen, p. 1 07).
3 Cf. also Brunn, Bayr. Sitzungsber. 1880, p. 473. Lowy, /. G. B. p. 67, No. 86.
4 It seems to me better to set aside the information of Pliny than to try to combine it with the ascertained fact
G G

226 POLYKLEITOS
Polykleitos was the head of a large school, and in his actual family were many
artists, though his sons, according to Plato,1 cannot have attained to any great
distinction. It is likely, though not actually proved, that the younger Polykleitos
was related to the elder. As he was brother to Naukydes and Daidalos, he must
have been a son of Patrokles. Now the facts, we know, fit together best if we regard
Patrokles as a somewhat younger brother of Polykleitos, so that the younger Poly
kleitos, together with Naukydes and Daidalos, would be nephews of the elder.
However much Polykleitos may have been — as he indubitably was — the domin
ating personality in his circle, we must yet allow for the possibility that Patrokles
may have kept at his side a certain independent style of his own, which he be
queathed to his sons. At any rate Naukydes and Daidalos, who mention their
father in their inscriptions, very probably do so because he had been an artist of
merit. The monuments will throw further light on this point.

II. The Doryphoros.
The Doryphoros of Polykleitos has been longer and better known than any
other statue by this artist. It has, however, not yet been subjected to an exhaustive
analysis based on a careful comparison of all its numerous replicas, nor has its great
significance in the history of art been determined by reference to all other monuments.
Though I am not in a position to undertake this task, a few modest contributions may
yet be offered here.
We have already assigned this work to the period circa 450 — 440 — the very
time when Pheidias produced his Parthenos. Now the rise of Polykleitos in Argos
marked no less an epoch than that of Pheidias in Athens ; the Doryphoros is as
decided an advance on its predecessors as the Lemnia and the Parthenos are on
theirs. The Doryphoros may be compared with two works dating circa B.C. 460, which
belong to the generation of Argive art immediately preceding Polykleitos. The one
is the Munich Zeus already described ; the other, the little bronze athlete from the
Argolis now in the Berlin Museum.2 In the Zeus, the walking motive is already
introduced, so that Polykleitos found it ready to hand ; some attempt is also made to
represent the muscles and to indicate the veins. There is still, however, a wide gulf
between this statue and the Doryphoros. The attitude is stiff and angular, and there
are evidences throughout to show that the old canon of Hagelaidas has not been
superseded : the shoulders are too wide, the hips too narrow, and the stomach flat
and wooden. No attempt is made to show the effect produced on the muscles
that Polykleitos was an Argive, by means of such conjectures as— that Polykleitos was made a citizen of Argos
after his gold and ivory Hera had been set up (Loschcke, Arch. Ztg. 1878, p. 11), a theory on which Robert
(Arch. Mdrchen, p. 101) builds further improbable conjectures.
1 oiSev vpbs rbv iraripa eM. Plato would certainly not have said this, if at that time any sons" of Polykleitos
had been distinguished artists. The son of Polykleitos whom Lowy conjectures in the inscription I.G.B. 89 is
quite problematic. So too is the conjecture of Robert (Arch. Mdrchen, p. 107) that a Patrokles who was
employed B.C. 404 on the Delphic votive offerings was the son of Polykleitos, and to be distinguished
accordingly from the elder. Patrokles. This is quite unnecessary, for even were Patrokles the father of Naukydes
and Daidalos he could quite well have been working after B.C. 404.
2 50th Berl. Winckelmannsprogr. 1890, ' Eine argivische Bronze,' Taf. i. The provenance of this bronze,
Liguno in the neighbourhood of Epidauros, has been certified to me in the most reliable manner. Frdhner's
statement (Coll. Tyskiewicz, text to PI. 13) that it was found at Olympia is quite unwarranted (cf. Berl,
Phil. Woch. 1894, p. 1 140). ¦ •

THE DORYPHOROS 227
between breast and hip by the position of the left leg, and the lowering of the left
shoulder ; a treatment that practically makes no sort of distinction between the two
sides of the trunk when the body is in motion cannot be true to nature. In various
other points we see signs of archaic treatment ; the hair, for instance, lacks variety
and delicacy, and the fillet encircles the head in an archaic, lifeless manner. The
bronze athlete, on the other hand, is much less stiff and angular ; the shoulders are
no longer too broad, and the hips, like those of the Doryphoros, are fuller ; L the
abdomen is rounder and more natural, and the back is really excellently modelled.
The old canon is, however, closely adhered to in the pose of the figure and in the
archaic treatment of the hair.2
These works form the foundation on which arose the harmonious conception of
the Doryphoros. In this marvel of proportion all stiffness disappears, and every
detail of attitude and movement produced by the momentary pause on one leg while
walking is rendered with truth and accuracy. The artist has overcome that crux of
former times, the rendering of the muscles between breast and hip on the side of the
leg that carries the weight of the statue. The powerful muscles are developed
equally all over the body without any undue exaggeration of particular parts, and all
the principal veins are indicated. The formation of the abdomen is not inferior to
that of any other part ; it is in its way quite as true as that of the breast ; the fond
ness for large plane surfaces with clearly defined edges — in a word, the very quality
which, by comparison with nature, produces a highly conventional effect in the breast
and abdomen of the Doryphoros — was universal in Greek art up to the time of
Praxiteles and Skopas, when a rounder and more natural method was introduced.
The carefully balanced proportions of the Doryphoros show at a glance how far it
surpasses its predecessors ; and this superiority can be more accurately estimated by
the help of measurements.3 The head too, with its flat, almost angular contour, is
noticeable as a characteristic innovation of Polykleitos. The hair is no longer in
any way archaic, though it lies close to the skull and is only faintly modelled ;
the attempt at realism in its superimposed layers is particularly noteworthy and
original. The symmetrical parting over the brow is quite in keeping with the
harmony and measured proportion that governs the whole figure ; it may almost
be said that the desire for regularity is excessive. Apart from the walking motive,
the actual pose differs from the old Argive canon in two points. The one is purely
external, and consists in making the figure rest on the right leg, instead of on the
left, as was the case with all the earlier artists, including the immediate predecessors
of Polykleitos. The other difference is more essential, for it involves a complete
reversion of an old arrangement whereby the arm on the side of the supporting leg
was bent at the elbow and stretched out, while the other arm hung loosely down
on the side of the free leg. Thus the body fell into two separate halves, the one
absorbing all the tension and movement, the other being left in complete relaxation.
Now in such cases the introduction of the walking motive only tended, as we
noted in the Munich Zeus, to emphasize disagreeably this lack of balance.
Polykleitos, by simply reversing the position of the arms prescribed by the old
canon, restored the necessary balance between the two halves of the body, and
produced moreover a great effect by means of the crossing lines of the upper and
lower extremities. In his statues therefore the inactive arm hangs down on the side
1 Cf. 50th Berl. Winckelmannsprogr. p. 142 seq.
2 The omission of the pubes is also a sign of the old canon.
3 Cf. 50th Berl. Winckelmannsprogr. p. 142 seq.

228 POLYKLEITOS
of the supporting leg which is in full muscular action, where it harmonizes with the
curving outline infinitely better than it did with the free leg, while the other arm,
which is stretched out and holds some attribute x — in this case a spear — is of agreeable
effect on the same side as the leg which is carelessly drawn back. In this manner
activity and relaxation are equally distributed between the two sides of the body.
The harmony of the whole is further heightened by the position of the head,
which, following the old canon, is turned to the side on which the weight is thrown ;
it is not bent forward, but is set fairly straight.2 The expression of the face is
resolute and determined, but not very animated : it has however none of the heaviness
and dulness that is so universal in the heads of the old type.
The word ' Doryphoros,' as we saw before when discussing the statue of that
name by Kresilas,3 is merely a vague term borrowed from the art jargon of a late
period. Literally the name was given to ' an attendant bearing a lance,' and was
not at all applicable to our statue, which not only does not represent a servant,
but in its original form, as we learn from a copy on a Berlin gem, carried a short
spear or javelin.4 The original Doryphoros is much more likely to have been the statue
of an Olympic victor, probably of a pentathlete with his javelin. It must have been
first set up in Argos, as we find it adopted there for a figure on a heroic relief.5
One proof that it is the statue of an athlete lies in the fact that a marble copy
of the Doryphoros has been discovered in the Altis of Olympia,6 for a statue of this
kind found on this site must also necessarily be that of an Olympic victor. The
material, like that of all later statues at Olympia, is Pentelic marble. It was no
doubt dedicated by an athlete of some later period, who preferred a copy of the
famous figure of the Doryphoros to an original statue. Unfortunately only the torso
survives, and that in a damaged condition. The execution, though rather mechanical,
is on the whole less hard than in most of the Italian replicas. The copy probably
belongs to the first century B.C. or A.D.
There are many other replicas of the torso of the Doryphoros in different
museums.7 Two of these are so superior to the others in execution that they have a
real aesthetic significance. The first, which is in the Uffizi at Florence, is executed in
green basalt, and produces the effect of a bronze covered with an exquisite patina.8
It is a fine bit of careful workmanship.9 The other torso, of equally good workman
ship, is the Pourtales torso at Berlin.10 It is interesting to notice in it the veins on the
abdomen. As they do not occur in other replicas, I cannot feel certain that they
1 In the excellent Florence copy (Mon. d. Inst. a. 50, 2) the right arm is (correctly) stretched horizontally
forward. 2 Michaelis points out (Annali, 1878, p. 9) that in the Naples copy the head is rather too upright, and that
the other replicas are more correct in this respect.
3 Cf. p. 163 seq.
4 On the relief (Ath. Mitth. 1878, Taf. 13) a short javelin appears, not a long $6pv.
5 Ath. Mitth. 1878, Taf. 13. Collignon's view (Hist, de la Sculpt, i. 490), that the Doryphoros was a
decorative statue from one of the gymnasia at Argos, involves an anachronism, for such purely decorative statues
were not in vogue till much later.
6 Still unpublished, and apparently unnoticed.
7 Thus in Mus. Chiaramonti, No. 293 ; ibid, in reduced scale, No. 484 ; both unrestored. Two with
wrong restorations in Pal. Giustiniani, two in Pal. Massimi alle colonne, and others. One in Vienna was lately
published by R. von Schneider, Die Erzstatue vom Helenenberge, 1893, pp. 16, 17 (Jahrb. d. Kunsth. Samml. d.
Kaiserh. Bd. xv. )
8 No. 307. Dtitschke, Uffz. 535 ; Arndt-Bruckmann, Einzelverkauf, Nos. 94, 95. The back is admirably
worked. 9 This is the only replica in which the pubes is really well executed. It shows that Polykleitos also arranged
the curls on the pubes quite symmetrically from the middle outwards towards the two sides.
10 Friederichs- Wolters, Gipsabg. 507.

YOUNG PAN 229
existed in the original. Quite different is it with the veins on the extremities, which
appear on all the replicas, and which were already usual even in the severe style. We
shall see, however, that late Polykleitan art certainly represented the veins of the
abdomen, and perhaps the copyist got the idea in this way and transferred it to
the Doryphoros. There are a considerable number of heads which are copies of the Doryphoros.1
The one which gives the best idea of the original is evidently the bronze bust by
Apollonios at Naples ; 2 the carefulness of the workmanship, into which no incon
gruous elements have been introduced, produces a singular impression of fidelity ;
the hair especially seems to be very faithfully copied. In comparison to this
bronze most other copies appear superficial, and allow themselves every sort of
simplification in the hair ; the proceeding is the same as that noted in the case of the
Myronian Diskobolos. In the main features, however, such as the division of the hair
over the brow, all the copies are unanimous. In addition to the marble heads,
there is a good copy made of the same green basalt as the Florentine torso ; it is
unfortunately much restored.3
The Doryphoros and the ' canon ' of Polykleitos were, according to the exact
wording of our tradition, two distinct works, yet it is probable that the same statue
has been handed down under two names ; 4 both appellations are moreover equally
inappropriate, and belong to a much later period. When Lysippos, as Cicero
{Brutus, 86, 296) tells us, called the Doryphoros his teacher, he must have referred to
the canonical figure of Polykleitos. The monuments are quite in favour of this inter
pretation, for they prove how very frequently the Doryphoros was used as a model.
T shall only discuss here those works that are intimately connected with the
Doryphoros, although to confine oneself to these is to gain only a very partial idea
of the influence exerted by the statue. It is quite certain that a very considerable
influence must have been exercised upon subsequent art by the combination exhibited
in the Doryphoros of perfect dignity and calm with the walking attitude, yet this
influence is mostly matter of conjecture, and can only be definitely pointed to in
a very few cases.
The most notable figure directly derived from the Doryphoros is a bronze statu
ette in Paris 5 representing the young Pan. Though lacking the finishing touches of the
chisel, it is an exquisite original from the immediate circle of Polykleitos. I pointed
out some years ago the strong affinities which this bronze offered to the Doryphoros,
not only in the motive but in the system of bodily forms. The only difference is that
the arms, the left one particularly, are not so close to the body,6 and that the left
arm is more bent owing to the substitution of a short pedum for the lance ; yet the
fingers of this hand correspond exactly with those of the Doryphoros. The head is
1 So in Mus. Chiaramonti, on a statue to which it does not belong. — A good copy in the Vatican, also on a
wrong statue. — In the Pal. Valentini (staircase) on a wrong statue. — Inferior copies in Museo Torlonia, 469, in
the Coll. Barracco, PI. 43, 43 a, good ; PI. 44, indifferent ; in the Palazzo Pitti (Diitschke, ii. 12), in the Uffizi
(downstairs), in the Villa Mattei (Arndt-Bruckmann, Einzelverkauf, Nos. 116, 117), and others. Also, Notizie
degli Scavi, 1879, i. I ; Cavaceppi, Raccolta, ii. 2.
2 Friederichs-Wolters, 505. The clumsy eyeballs are a modern restoration. (Cf. p. 138, note 4.)
3 Guedeonow, Ermitage, Sculpt. Ant. No. 75, called ' Drusus l'ancien.'
1 There is no need to alter Pliny xxxiv. 55. It is undoubtedly correct to put a comma between puerum und
fecit, as Detlefsen does. The comment on the Doryphoros (which belongs to that on the Diadumenos) and the
comment on the canon probably come from quite different sources, which would account for the use of
different names.
5 Ath. Mitth. iii. 1878, Taf. 12 (Furtwangler) ; Babelon, Cabinet des ant. dc la Biblioth. Nat.
6 The illustration in the Ath. Mitth. is from a cast in which the right arm is closer to the body than it
is in the original.

230 POLYKLEITOS
also entirely Polykleitan in character, although the fuller, more plastic treatment of the
hair and the greater softness of mouth and chin point to its being a later work. The
interval between them however need not have been very great, and there is no reason
for dating this bronze later than the fifth century. Though no one is likely to regard
it as an actual work of Polykleitos, it can be assigned almost certainly to one of
his pupils, and it proves that within the school the Doryphoros was treated as a model,
and adapted with the slightest possible alterations to youthful figures of a quite
different order.
No other monument equals this in importance and originality, but the relief
from Argos x may be taken as the next in merit. From this free adaptation of the
Doryphoros to an heroic subject we learn how highly that statue must have been
esteemed in Argos about the middle of the fourth century.2 Passing now to
marble statues of the Roman period, we can recognize in a statue of the Villa
Albani 3 the Doryphoros transformed into a youthful hero, perhaps an Ares, armed
with a sword. In design and bodily structure alike it follows the model exactly,
only the lance being here replaced by the sword. The head is unfortunately lost.
The original probably belonged to the Polykleitan circle. Two bronze statuettes,
the one in the British Museum,4 the other at Geneva,5 are based on the Doryphoros,
and carry a spear in the left hand ; but the head with its Corinthian helmet and the
structure of the body differ from the model in their rounder, softer treatment.
A statue of Hermes in the Boboli Gardens6 at Florence follows the Doryphoros
in the structure of the head and body, and, but for the right arm, also in motive ; the
head, however, is rather more inclined. The right arm supports a small boy, probably
the infant Dionysos, of whom only one tiny foot is antique. The addition of an
ordinary Roman chlamys thrown over the breast,7 and of wings on the head,
considerably disfigures the original Polykleitan type. The workmanship (the
pupils of the eyes are carved on the ball) is that of the second or third century A.D.
As it is only owing to the Roman additions of chlamys and of wings that the
Polykleitan athlete has become a Hermes, and as the child on the arm does not fit
the figure, it may be that the whole statue is merely a Roman composition or
pasticcio. However, it is also possible that the original was a Polykleitan Hermes,
who, of course, would carry the kerykeion as his sole distinguishing attribute.8
There is no doubt that the group of Hermes and the infant Dionysos was not first
introduced into statuary by Praxiteles, for we know the composition from a Roman
bronze found in France, copied from a work that was not only pre-Praxitelean
in style but probably actually Peloponnesian.9
1 Ath. Mitth. iii. 1878, Taf. 13, p. 287 seq.
2 This date (ibid. 289) has recently been confirmed by the basis-relief of Bryaxis (B. C. H. 1892, PI. 3, 7),
on which the horses are modelled in a strikingly similar manner.
3 No. 604. Clarac, 833 C, 2074 A. Cf. Flasch in Bull. deW Inst. 1873, 10 ; Michaelis, Annali, 1878,
9, K. ; Helbig, Fiihrer, 824; Brunn, Bayr. Sitzungsber. 1892, p. 674; Kalkmann, Gesichtsprop. p. 53, note 6.
The middle of the sword is antique. On the right thigh a puntello has been planed away, which, as in the
Doryphoros, was for the support of the forearm. A piece of the neck is restored, and the head is probably foreign.
4 The so-called ' Bunsen Mars.' The right arm is missing. Workmanship latish.
6 Geneva, Musee Fol. 1275. The head with the hair parted in front is nearer to the Polykleitan type
than the London bronze.
6 Dtitschke, Zerstr. Antiken in Florenz, 84 ; Arndt-Bruckmann, Einzelverk. Nos. 103—105. Cast in the
Ecole des Beaux- Arts. The head is certainly genuine. The lower part of the left wing is ancient.
7 As in the Myronian Hermes (see p. 182 seq.)
8 The suggestion of Overbeck (Gesch. d. Plastik. 2nd ed. ii. 7), that the group is derived from Kephisodotos,
seems to be quite unwarranted.
8 Revue Archiol. 1884, vol. iv. PI. 4. The original was probably a work of about 400 B.C. This is shown

THE DORYPHOROS 231
A statue of Hermes of Roman date recently discovered at Troezene resembles
the Doryphoros,1 both in motive and in the bodily forms. The god, who in this
case wears a chlamys of good Greek style, leads a ram by the horn with his right
hand, and in the left he holds the herald's staff. Although the head with its
winged petasos is not purely Polykleitan, the statue must be referred to an
original of the school. This motive, the leading of a ram by the horn, recalls
Pliny's mention of the immolans arietem of Naukydes,2 which was perhaps similarly
composed. A series of Polykleitan heads representing Hermes with the petasos or with
wings will be considered later in connexion with another type of Polykleitan
Hermes which differs in pose.
A list of torsos must however be given here which follow the Doryphoros
more or less exactly in forms and attitude, but which are shown by the chlamys, or
by the end of a cloak thrown over the left shoulder, to be either representations of
Hermes, or perhaps mere portrait-statues.3 As the drapery is by no means the same
in all, and is for the most part decidedly late in its treatment, these torsos are
interesting only as showing that even up to quite late times the Doryphoros did duty
as a model. A well-preserved and interesting statue from Carthage has recently been
acquired by the Louvre ; it repeats the figure and pose of the Doryphoros with the
additions of a chlamys over the left shoulder ; the right hand hung down, holding an
attribute. The head however is totally different, being slightly turned to the left, and
of the ordinary Dioscurus type with the pilos ; a horse's head below serves as attribute.
This association of heterogeneous elements in head and body is essentially Roman.4
Lastly, in a statue at Carlsruhe a later type of head has been combined with the body
of the Doryphoros, though the action of the arms has been reversed.5
Passing now to bronze statuettes of later date, we find that the largest and most
important of all is in the closest possible connexion with the series we have just
considered. This is the well-known Hermes from Fins d'Annecy,6 at present to my
knowledge in the possession of M. Dutuit at Rouen. For all the delicacy of the
workmanship, this charming bronze, which was once gilt all over, must be considered
as a Roman work of the Augustan age. The finely modelled body is evidently
inspired by the Doryphoros ; the arms, however, are too short, and, just as in the
Carlsruhe statue, their action has been reversed.7 Other variations from the model
betray a late and imperfect style. The head is very similar to that of the Dory-
by the unnatural rendering of the child, the simple flat modelling of the principal figure, the attitude,
recalling a Peloponnesian Hermes type which rests on the left leg (to be discussed later), and finally the head
with the parted hair.
1 B. C. H. 1892, PI. 2, 17 ; p. 165 seq. (Legrand). Poor Roman work : not an exact copy, but a free
adaptation of a Polykleitan type.
2 The current identification with the ' Phrixos ' on the Akropolis is quite uncertain.
3 (a) Naples, Mus. Naz. No. 6102 ; measurements = those of the Doryphoros : portrait-head foreign.— (b)
Villa Albani, 596 : the head is foreign, and appears to be a replica of the Ludovisi Hermes ; the bodily forms are
somewhat softer. — (c) Mus. Torlonia, 343 : head, left forearm, with sword and right arm, are modern. — Palazzo
Valentini : restored as Pertinax.— (d) Lateran, No. 836 (Benndorf-Schone, 445) : puntello for the right arm in the
usual place, yet the left seems to have been raised. — (e) Lower torso in the Vatican, Belvedere, No. 8 : puntello
or the right arm in the usual place ; a cloak hangs down behind ; a horn of plenty on the tree-stem ; probably
a portrait in the character of Genius.
4 Louvre, No. 2735. Colossal statue of coarse marble much weathered ; good Roman work.
6 Sketched without the restorations in Arch. Anz. 1890, p. 4, No. 6. The position of the arms is fixed by
the puntelli. 6 Mon. d. Inst. x. 50, 4 ; Annali, 1878, 25 seq. (Michaelis) ; Gazette Archiol. 1876, PI. 18. Cf. B. C. H.
1892, 169. I saw the original in 1881.
7 They are cast separately, as is so frequently the case in larger bronzes.

!32

POLYKLEITOS

phoros, but here and there, espe
cially at the back of the head, the
Roman artist has introduced later
elements into the hair. The
statuette probably represented a
Roman deity, but it is hard to
determine which, as the remains
of the attribute are doubtful.1
The current view of its being a
copy from an original composition
of Polykleitos (Michaelis ranks it
as third along with the Doryphoros
and the Diadumenos) is absolutely
untenable. This figure can only
be correctly appreciated in con
nexion with later adaptations from
the Doryphoros, and it certainly
cannot help us to recover the
Hermes of Polykleitos.
A much rubbed Roman
bronze in the Louvre shows the
Doryphoros type without any
addition of drapery, but marked
as Mercury by the winged feet.2
Other Roman bronzes unite in
a most disturbing manner the
Polykleitan torso with a head of
the ordinary later Hermes type.3
Very common are the Mer
cury bronzes in which the chlamys
is added to the Polykleitan body :
in the best examples it is arranged
after the older fashion, so as to
cover the entire left arm. The
finest specimen comes from Gaul,
and is in the British Museum
(Fig. 93).4 Both head and body
are excellent, though not wholly Polykleitan.5 Still better in style, but not in such
good preservation, is a figure at Copenhagen found at Pompeii.6 A similar but
1 It might possibly be the end of a bow.
2 Louvre, Bronzes, 1031 : the left arm is missing, the right is lowered ; the head still shows the Polykleitan
type. 3 Paris, Cab. d. Medailles, Coll. Janze, 81 : the body reproduces with fair accuracy the forms of the Dory
phoros ; veins on the abdomen. ' The right arm is missing. —Similar but inferior is u bronze in Brunswick, No.
289, with prominent muscles.— British Museum, R.P.K. xlvi. ; others from the Thames Archceologia, vol. xxviii.
PI. 5. — Frohner, Coll. Griau, Bronzes Ant. 933.
4 Specimens of Ancient Sculpture, i. 33, 34; Miiller- Wieseler, Denkm. a. K. ii. 314. Cf. my remarks in
Tahrb. d. Ver. v. Alterth. fr. im Rheinl. vol. xc. p. 58.
5 The pubes less good than the hair. Veins on the abdomeh as in the Pourtales Doryphoros torso. The
left foot less drawn back than in the Doryphoros.
6 Bronzes, No. 14. Pub. by Jul. Lange, Fremstilling af Menneskeskikkelsen (Abh. d. Kopenh. Akdd. 1892).
p. 407, fig. 62. The lowered right hand is empty ; the left evidently held the kerykeion,

Fig. 93. — Bronze statuette in British Museum.

THE DORYPHOROS

233

inferior one belongs to Sir Francis Cook
at Richmond,1 a fourth is in Paris.2
None of these statuettes can claim to
be an actual copy of a Hermes by Poly
kleitos.3 They all vary in detail, and
must be regarded as Roman creations,
founded on the Doryphoros, and borrow
ing the chlamys from some other fifth-
century model.
Among these figures also are some
in which the Polykleitan head is replaced
by that of a later type of Hermes,4 while
others again show the body of the
Doryphoros draped in a chlamys hang
ing from the shoulders in a point in
front.6 We have an interesting work
of this class in a bronze Mercury of
the Oppermann collection in Paris (Fig.
94).6 Here a fine chlamys executed in
the style of the fifth century covers the
whole body to the knee. The parted
hair points to a Polykleitan inspiration,
but the Doryphoros attitude is consider
ably modified by the increased inclina
tion of the head and right shoulder, and
by the extension of the right arm. These
changes give a softer, sweeter effect to
the whole figure. It need not surprise
us to find the Doryphoros so often
adapted to representations of Hermes
in Roman times, for copies of this statue
were placed in the various palaestrae
and gymnasia,7 which were all conse
crated to Hermes.
All this is interesting, as showing the vast influence exercised by one creation of
Polykleitos ; but it has added nothing to our knowledge of the artist himself, since not
one of the works described can be definitely referred to any actual work of his. The
1 Height, C15. The left forearm raised, the left hand held the kerykeion ; remains of a purse in the lowered
right hand. The left foot was slightly drawn back. Polykleitan characteristics in the head much weakened.
Rich sandals ; eyes and nipples of silver.
2 Cab. d. Medailles, 3351. Head much defaced, with traces of Polykleitan type. The lowered right hand
empty, the extended left hand probably held the kerykeion.
3 Cf. Treu, Arch. Anz. 1889, p. 57.
" Paris, Cab. d. Medailles, 3350.— Clarac, 666 D, 1512 F.
5 Sacken, Bronzen in Wien, Taf. 10, 4.
8 Paris, Cab. d. Medailles, Coll. Oppermann. Height, about 0'20. The left hand evidently held the kery
keion ; the right has a hole, possibly for the purse. To judge by the colour of the metal, the figure seems to
. be from Gaul.
7 The Naples copy comes from the Palaestra of Pompeii. Cf. Nissen, Pompej. Studien, p. 166. As the
figure carried a lance, it cannot possibly be a Hermes. But the Ephebe statues with the lance, which according to
Pliny (xxxiv. 18) were called Achilleae, and were set up in the gymnasia, were very likely for the most part copies
of the popular Doryphoros by Polykleitos. v
H H

Fig. 94. — Mercury, Coll. Oppermann (Bibl. Nat., Paris).

234 POLYKLEITOS
assumption that Polykleitos often repeated himself by adapting the Doryphoros type
to other statues is certainly not supported by the monuments themselves.
On the other hand, our study of the Doryphoros will enable us to point among
our copies to other original creations by Polykleitos. There is, for instance, a type of
head (Fig. 95) so like the Doryphoros that it is often confused with it, though it differs

Fig. 95. — Head of Herakles. From Herculaneum (Naples).
in a distinct detail, i.e. in the rolled fillet round the head, and also in size, being only
life-size, while the Doryphoros is rather above. The rolled fillet is a well-known
attribute of Herakles, and as at any rate one replica of the head in question shows an
evident attempt to give the characteristics of that hero 1 (though it must be admitted
1 The head from Herculaneum in Naples, Comparetti de Petra, Villa Ercol. Tav. 21, 3 (our Fig. 95). B.
Graf (Rdm. Mitth. 1889, pp. 215. 202 sqq.) had already conjectured a Polykleitan Herakles. He also mentions

HERAKLES

235

that in so doing the artist abandoned the original and introduced traits belonging
to a later type), it is quite possible that the head actually represents Herakles.
In favour of this interpretation is the fact that almost all the replicas known to me
originally belonged to terminal busts, which explains why so many heads and no
statues belonging to them have been found ; for it was precisely terms of Herakles
that were in great demand for palaestrae and gymnasia. The Skopasian Herakles
likewise is found almost exclusively in the term-shape. Both types appear to have

Fig. 96. — Head of Herakles. From the collection at Broadlands (Hampshire).

been very popular for this purpose, and in both the inclination of the head varies from
right to left,1 though the former is more common ; it is very slight either way, and in
the original the head was probably set straight.
two replicas — viz. Lateran, 896 = Bennd.-Schone, 491, and Museo Chiaramonti, No. 139 (Helbig, Museums, 69).
In the second the neck is antique, the bust modern. The head might have belonged to a statue ; the ends
of the fillet are missing. The work is sketchy and poor. Other replicas are : A term at Broadlands (Fig. 96 ;
Michaelis, Anc. Sc. p. 220, No. 10). The hair is worked in much more detail than in the other copies named
above. — In Berlin, No. 478, mediocre copy ; probably intended for insertion into a term. — The janiform term,
Berlin, No. 477, may be counted among the good copies. Slight turn to the left. Hair very carefully rendered, as in
the Broadlands example, much better than the Chiaramonti copy. — A replica in Dresden, for insertion in a term
(Zug. Verz. 1106) from Rome. The length of face in five copies is 0-182 — 0-185 (in the Doryphoros, O'2oo —
C204), width of face at the cheek-bones o-U3 (in the Doryphoros, 0-156).
1 So Chiaramonti, 139 ; Berlin, 477.

236 POLYKLEITOS
This Polykleitan Herakles may perhaps be copied from the statue in Rome
mentioned by Pliny (xxxiv. 56) l ; unfortunately all the replicas are very mediocre.
The term in Naples is undoubtedly the most characteristic of Herakles, and has least
in common with the Doryphoros. The chin and lips are unusually full and powerful,
and the nostrils are distended. The widely opened eyes appear to project a little, the
hair of the eyebrows is indicated, and two deep furrows rise from the root of the nose.
In the other replicas all these details are lost ; their features assimilate to those of the
Doryphoros to the extent of appearing even calmer and more impassive. The careless
treatment of the hair in the Naples replica shows, however, that it is not accurate in
matters of detail ; the intention of satisfying the common ideal of Herakles by just
giving a look of concentrated strength to the face is so evident that we must be
cautious how far we trust this head in those points where it contradicts the other
replicas, in which the hair at any rate is much more accurately modelled.
The resemblance between the Polykleitan Herakles head and the Doryphoros is,
as remarked before, very great. The stylistic treatment of the hair is the same ; yet
a close examination brings differences also to light. There is less anxious symmetry
in the Herakles ; the hair on the top of the head is treated in broader, looser masses,
and the locks towards the ends are accordingly all the richer. The little wisps on
either side over the middle of the forehead are not symmetrical as in the Doryphoros.2
It is not till we come to the next two larger meshes of hair with their points turned
inwards (these meshes are a chief sign by which our head can be distinguished from
the Doryphoros) that we find approximate symmetry. The little curls in front
of and behind the ears3 are more varied in form and movement than they are
in the Doryphoros. Further, the modelling of the forehead is somewhat more
accentuated, while the lower part of the face produces a more delicate effect.
From all these remarks it may be inferred that the Herakles is later than the
Doryphoros. The rolled fillet in the hair, shown in all the replicas, and which is sometimes
decorated with flowers, must have belonged to the original, but without the long ends
falling on the breast ; these are absent from some copies, and vary in the others,
and seem to have been introduced by the copyists who made the terminal busts.
There are other instances of Herakles wearing the rolled fillet, e.g. in the beautiful
statue of the Museo Chiaramonti {infra, Figs. 146, 147), where he is represented
carrying the infant Telephos on his arm. It may therefore be regarded as an external
means of characterizing the hero. It is no simple victor's fillet ; it seems borrowed
from the symposion, and distinguishes the glorified hero, rejoicing in the heavenly
banquet. Of inner characterization by means of an individual form of head there is
scarcely a trace, if we leave out of account the additions made by the Naples copyist.
The head is of pure youthful beauty, but it might just as well represent any other hero.
How different are those other two Herakles types, which we ascribed to Myron—
the bearded head from the artist's earlier period (p. 179), the beardless from the later
(p. 202), in both of which we can trace something of wild heroic force. The gentle
beauty of the Polykleitan head is most closely paralleled by the Herakles head on some
1 Pliny, xxxiv. 56, Herculem qui Romae. By Detlefsen's punctuation, which is certainly correct, the following
hagetera arma sumentem is a different work. Cf. Benndorf in Festschr. fiir Springer, 1885, Eine Statue Polyklets,
p. 1, note i. ; Roscher's Lexikon, i. 2157.
2 In the Naples copy these are omitted. The Broadlands head (Fig. 96) and the Berlin janiform term
(477) giye them exactly. In the Dresden copy they are broken away.
3 These are carelessly copied on the Naples and Berlin (478) heads. The Broadlands (Fig. 96) and the
Berlin (477) copies (which agree exactly), and still more the Dresden head, give a more exact version,

HERAKLES 237
coins of Kleonai and Stymphalos,1 i.e. from the neighbourhood of Argos, a district
where Polykleitan art must have been a dominant influence.
The Herakles of Polykleitos is by no means a mechanical repetition of his
Doryphoros, and yet the work shows a striving rather towards harmonious repose

Fig. 97. — Polykleitan statue in Coll. Barracco, Rome. (By permission of Messrs." Bruckmann.
and pure abstract beauty than towards characterization. We can understand the
judgment of antiquity as given by Quinctilian, according to which Polykleitos non
explevisse deorum auctoritatem videtur . . nil ausus ultra leves genas.
1 Cf, Roscher's Lexikon, i. 2163, /. 53 seq.

238 POLYKLEITOS
Another creation, which must be assigned to Polykleitos himself, and which has
in common with the Doryphoros the walking motive (the right leg being the support
ing one), the turn of the head to the right, and, above all, the bodily forms, affords
fresh proof of how little this artist simply repeated his ' canon,' and of how fresh
and independent was each of his works. By merely altering the movement, in this
case of the arms, he has created a totally new conception (Fig. 97).
The statue is known to us complete only on a much-reduced scale, but there
are copies of the torso in the original size, which is the same as that of the Dory
phoros.1 The restrained repose of the Doryphoros is here abandoned. The left
hand is laid on the back — it rests behind the left hip on the upper part of the gluteus
— and the right arm, which hung down, appears, from a fragment above the right
hip in one of the statuettes,2 to have held an attribute shaped like a staff. The head
is turned to the right with more animated freedom than in the Doryphoros. Finally,
the left leg is not, as in the Doryphoros, simply drawn after the other in such a way
that the front surfaces of both thighs lie in the same plane, but the left foot is more
turned out so that the thigh too slants outwards. All these changes introduce more
life and freedom ; in fact, the attitude might almost be called bold and unrestrained.
The analogy to the Doryphoros in the formation of head and body, although
striking, does not amount to identity. In our statue all is rounder and less severe.3
It is certainly a later work than the Doryphoros. Who is represented it is difficult
to say, probably not an athlete. The bold presentment is not suited to a human
victor, at least not according to the Doric notions of the time, which always lay
stress in athlete statues on modesty and restraint. The remains of the attribute,
too, would be difficult to explain. I therefore incline to interpret the figure as
Hermes with the herald's staff in his right hand.4 The motive of the left arm would
make the figure a forerunner of the Belvedere Hermes.

III. The Diadumenos.
The Diadumenos is the second undoubted work of Polykleitos which survives to
us in copies. Michaelis has shown, principally by comparison with the Doryphoros,
that the statue from Vaison in the British Museum is a copy of this statue.5 But
the further questions as to the artistic value of this copy and the existence of other
replicas have not yet been sufficiently investigated. Yet this inquiry is unavoidable
in order to form an exact notion of the original.
1 Statuettes : Gall, dei Candelabri, 269 B : head broken, but genuine ; right arm restored ; feet ancient ;
workmanship poor. The Barracco statuette (Coll. Barracco, PI. 45, 45 a - our Fig. 97) is better : clean, sharp
workmanship; no restorations.— Life-size torsos : (a) Palazzo Mattei (cf. Petersen in Bull, della Commiss. Comm.
1890, p. 191) ; (b) formerly in possession of Spiess the sculptor in Rome (photograph in the Berlin Museum), and
is now in the Coll. Jacobsen, Copenhagen ; (c) in the Thermae of Caracalla (Matz-Duhn, 1013).
2 In the Barracco copy. In the Jacobsen torso the arm hung down as in the Dorvphoros (puntello at the
same spot).
3 The insipid Vatican statuette, in which all the forms have a marked roundness, is in this respect quite
inexact : the mouth, for example, has lost all Polykleitan character. The Barracco statuette is much more severe,
and more similar to the Doryphoros in head as in body. In the large torsos the body comes very near the
Doryphoros. 4 The Hermes of Polykleitos mentioned by Pliny was probably in Rome, as Pliny's words— qui fuit
Lysimacheae-show, and must have been copied. For the motive cf. the coin of Chalkedon in the British
Museum ( Catalogue Pontus, PI. 28, 3), where the left arm is set against the body, and the right hand holds the
kerykeion, but the head has a different turn.
5 Ann. d, Inst. 1878, p. 10 seq.; Monum. a. 49.

THE DIADUMENOS 239
E. Petersen lately made an important contribution to this subject by discovering
in a number of torsos copies of the body of the Diadumenos.1 Most of these are better
and more careful pieces of work than the Vaison statue. The following are worthy
of special notice : a torso in Palazzo Giustiniani,2 another in the Louvre,3 to which I may
add an excellent one in Turin 4 (not mentioned by Petersen). A small statuette
torso worked as a cabinet piece should also be reckoned among the exact copies.5
For the study of the body of the original we have a reliable basis in these torsos.
But what of the head ?
The head of the Vaison statue can give us but a very unsatisfactory notion
of the original ; not so much because, like the body, it is a poor and careless
piece of work, as because it is so much defaced and — worse than this — so much
and so arbitrarily worked over. The forehead is completely damaged, the eyes
are injured, the nose is entirely gone, the ears and all the prominent parts of the
hair have been knocked off, and the edges of the fractures rubbed smooth, so that
only the foundation or lowest stratum of the hair-modelling remains. On the top of
this the modern rrstorer has engraved lines according to his own fancy.6 This makes
the hair look as if it lay close to the head and fell away from the crown in long wavy
lines. In order therefore to form an accurate conception of the original we must
begin by disregarding this head with all its misleading interpolations, and look for
copies in better preservation.7
A terra-cotta statuette from Asia Minor is interesting in this connexion.8 It
belongs to a small class of terra-cottas which are real copies from statuary originals : 9
they are naturally somewhat less exact than the larger ones in marble, though they
do not aim at being anything but copies ; further, they belong approximately to the
same period — the first century B.C. — at which the exact copies in marble of older
originals came into vogue. If we compare the body of the terra-cotta Diadumenos with
that of the marble replicas, we find a very close correspondence,10 so that the same is
likely to be the case with the head. Now, from the treatment of the hair and the
modelling of the forehead it is clear, in spite of the small scale, that the head repro
duces a well-known Diadumenos type, known to us through separate heads, in Cassel
and in Dresden, and usually supposed not to be Polykleitan.
1 Bull, della Comm. Arch. Comunale, 1890, 185 seq. 2 Petersen, No. 5.
3 Petersen, p. 189, note 2. The statue is now numbered 2235 (photo. Giraudon).
4 Museum, Turin. Torso without restorations ; no puntelli on the shoulders. Measurements the same as
those of the other replicas.
5 Berlin, Skulpt. No. 513.
6 It was Loschcke who first called my attention to the fact that the Vaison head had been worked over. A. H.
Smith kindly examined the original at my request, and confirms my suspicion. He writes : ' You are right : in
those parts where the hair is especially sharp, the surface has been worked over by a restorer.'
7 Michaelis (Ann. d. Inst. 1878, p. 11, A; tav. d'agg. B.) gives the first place among copies of the Poly
kleitan statue to a small bronze of the former Janze Collection. Fie goes so far as to try to recover the propor
tions of the original from this bronze. This is a mistake, for the technique and style of the bronze prove
it to be a Greek work of the best period, and as such it is excluded from the category of copies. The artist
is of course influenced by the famous work of Polykleitos, but he gives - free imitation of it : no definite
conclusions can be drawn from this bronze as regards the Polykleitan Diadumenos.
J. H S. 1885, PI. 61 (A. S. Murray), now in the possession of Mr. C. Blacker in London. I have held
the original in my hands, and can testify that it is genuine. The head of another terra-cotta belongs to M. Misthos
in Smyrna. These figures survive mostly in fragments. They come from the neighbourhood of Smyrna ; there were
numerous fragments of this kind in the former Greau Collection.
Opinions on this figure have hitherto been founded on wrong premises. E. Petersen, Bull. d. Comm Arch.
1890, 186, 4, calls it an original. Von. Sybel (Weltgesch. d. Kunst, p. 194) assimilates it to the Farnese type,
from which the motive alone would be enough to distinguish it. Sal. Reinach (Gazette Arch. 1887, 281) thought
it might be a copy of a more recent work, perhaps by Praxiteles.

240 POLYKLEITOS
The most beautiful and best preserved of all the Diadumenos copies, the statue
in Madrid (Fig. 98),1 shows however, beyond a doubt, that this type is really Poly
kleitan. The head of the Madrid Diadumenos is still unbroken, and the body
corresponds exactly to the torsos known to be after the Polykleitan Diadumenos.
The head resembles, feature for feature, the Cassel and Dresden heads.
The head of the Vaison statue too was originally similar to these, although it
could never have been a good copy. The strands of hair that have not been worked
over correspond exactly to those heads. The measurements are throughout the
same. As there are so many torsos of the Polykleitan Diadumenos in existence, it would
seem likely that there should be several copies of the head. And this is really the
case. The heads which, from their agreement among themselves in measurement
and in the forms of the hair and face, must be referred to one original, that original
being, as the statues of Madrid and Vaison show, the Diadumenos of Polykleitos, are
the following : —
1. The Dresden head (Plates X. and XI.),2 a very exact and carefully
executed work. The copyist evidently took some trouble to reproduce the original.
He certainly interpolated nothing of his own : his work is dry, but the more reliable
on that account. From the manner in which he indicates the eye-pupil and forms
the upper eyelid, we can scarcely suppose that he worked before the latter half of the
second century.
2. The Cassel head,3 a somewhat earlier work, is less dry and expressionless, but
also less exact in details. The hair is rendered in a looser and more general manner
without the rich and luxuriant variety of No. 1, and yet the agreement of the two
heads down to the most trifling details is so close that we must conclude they are
derived from the same original.4 The Cassel head is the better of the two as a work
of art, but inferior as a copy.
3. Fragment of a head in the Louvre : 5 only the front part of the face — forehead,
eyes, nose, and mouth — is preserved. This is a careful piece of work, but dry and
lifeless. In the endeavour to render truthfully the sharp lines of the bronze
original all refinement of expression has been lost. This fragment corresponds
with the Dresden and Cassel heads even in the peculiar modelling of the
brow. 4. Head in the Barracco collection in Rome. A poor copy, much defaced,
especially in the hair. What is left, especially the hair on the top of the head and the
forehead, corresponds exactly to the other replicas.
5. A head, formerly in the possession of Steinhauser, in bad preservation, and
much restored.6
6. A head which I noted in Rome in 1 892 in the market : a very poor copy, the
hair merely blocked out.7
1 Michaelis, Annali, 1878, p. 11, u. Besides the legs, only the right arm is restored, and that wrongly.
This is evident from the photograph. I have not seen the original.
2 Ann. d. Inst. 1871, Tav. v. p. 281 (Conze). Friederichs-Wolters, 51 1. The nose down to the left nostril
and the right half of the mouth are restored. Coarse-grained Parian marble.
3 Conze, Beitrdge, Taf. 2; Friederichs-Wolters, 510. Only the front half of the nose is restored. The
mouth is in complete preservation. Parian marble. Clear traces of the drill in the hair.
4 Michaelis was mistaken (Ann. 1878, p. 24) in referring the Dresden head to a different original.
s Bouillon, iii. Busies, PI. 3, 'Athlete' 2. Photographed as 'Diadumene de Polyclete' by Giraudon, No. 1283.
6 Matz-Duhn, Zerstr. Bildw. in Rom, 1671. Known to me only through the cast in Strasburg (Michaelis,
592). Most of the face is restored, the hair much battered.
7 Arndt- Bruckmann, Einzelverkauf, Nos. 190, 191 (incorrectly called a replica of the Farnese type).

X.
Head of a Diadumenos {full face
j DRESDEN.

Fig. 98.— Diadumenos in Madrid.

I I

242 POLYKLEITOS
7. A much-defaced head on a statue to which it does not belong, in the court
yard of the Palazzo Giustiniani.1
That these copies reproduce the famous work of Polykleitos seems probable from
two considerations alone. First, they are all referable to the same original, which
must have been a famous one. Second, all other ' Diadumenos ' heads are entirely
different and are certainly un-Polykleitan.2 But this probability becomes a certainty
when we add the evidence of the Vaison and Madrid statues.
Since, then, we possess reliable copies of the head of the Diadumenos of Poly
kleitos, as well as of the body, we have the material before us for a discussion of the
whole work.
A correct appreciation of the great difference which undoubtedly subsists between
the head of the Doryphoros and the Diadumenos heads enumerated above has given
rise to the mistaken opinion that the latter were not the work of Polykleitos, but
were specifically Attic in character. They were supposed to be better and larger
replicas of the Farnese Diadumenos.3 These hasty conclusions were the result of
defective observation. The different arrangement of the fillet, which is tied in a
bow behind, and the treatment of the hair are enough in themselves to exclude
the Farnese type from all relation to these heads.
The softer execution of the Diadumenos shows it to be a later work than the
Doryphoros. The contrast between the two works was summed up epigrammatically
in the passage of Pliny, where the Doryphoros is ascribed as viriliter puer, and the
Diadumenos as molliter juvenis.
The Diadumenos should probably be assigned to the same later period of the
master's career as the Hera. I think I am not mistaken in finding a certain kinship
between the head of Hera on coins of Elis (PI. VI. 23) and the Diadumenos. The
short yet full and plastic locks, the broad diadem sunk deep in the hair, the angular
contour of the powerful skull, the large features and especially the lower part of the
face, are strong points of resemblance made doubly remarkable by the difference of
sex and of subject.
The body of the Diadumenos is scarcely to be distinguished from that of the
Doryphoros. It is difficult to decide whether the small differences that do occur are
due to the copyist or to the original, yet in the best copies of the Diadumenos the
flesh-forms produce an impression of greater spareness and elasticity, and of softer
transitions in modelling. With regard to details, it should be noted that on none of
the Diadumenos copies do veins appear on the abdomen.
In movement and carriage the Diadumenos comes very near the Doryphoros.
He also rests on the right foot, pauses in the act of walking, and inclines his head to
the right. Here too, notwithstanding the different direction, the left arm is the one in
more active motion ; it is freer, and held more away from the body, while the right
upper arm is kept as closely as possible to the body, till the forearm bends sharply
back and upwards. The pose of head differs from that of the Doryphoros in being
much more bent and inclined to the right, although the change is not required by the
action represented. The intention evidently was to give to the figure a softer and
more flowing rhythm. Another small point in which this figure differs from the
Doryphoros is that — as we have already noticed in the figure of a youth laying his
1 Clarac, 872, 2218; Matz-Duhn, 1091.
2 The Farnese type and the Petworth type discussed above (p. 16 seq.)
3 Cf. Botticher, Verzeichniss d. Abgiisse in Berlin, 714, 715; then Conze, Annali d. Inst. 1871, 282;
Michaelis, Annali, 1878, 23; Friederichs-Wolters, 510, 511.

XI.
Head of a Diadumenos {profile).

DIADUMENOS AND DORYPHOROS 243
hand on his back (Fig. 97)— the left foot is a little more turned out, so that the left
thigh is no longer in the same plane as the right, but takes a slanting direction.
This change adds animation to the pose.
The two statues differ to a far greater extent in the head than in either torso
or attitude. It is from a study of the head that we first realize the rapid development
of which Polykleitos was capable. He treats the hair of the Diadumenos in quite a
new manner, gives up the flat clinging strands, and adopts the full raised plastic
curls ; the almost finicking trimness of the Doryphoros gives way to a living variety.
Only at one spot in front does the hair really resemble that of the Doryphoros, just
where the line which rises from the forehead makes, when seen in profile, a sharp
turn almost at a right angle. The way in which the hair is here disposed in
layers is essentially the same in both heads, and is thoroughly characteristic of
the artist. Again, the Diadumenos differs from the Doryphoros in the modelling of the fore
head ; what was begun in the head of the Herakles has now been carried further. While
the forehead of the Doryphoros projects evenly below and falls away in simple planes
to the sides, in the Diadumenos the forehead is slightly hollowed in the middle {i.e. at
the intercilium), and the arches formed by the long framework of the eyebrows are
distinctly emphasized above the inner ends of the eyebrows themselves.
Finally, the lower part of the face is in the Diadumenos decidedly softer and less
prominent than the same part in the Doryphoros ; the cheeks and mouth are modelled
with a richer variety of planes ; the same difference is to be observed in the nose,1
which, instead of a hard angular formation, shows rounder and more lifelike modula
tions. The bend of the head enhances not a little the general effect of softness.
The innovations introduced by Polykleitos into the head of the Diadumenos —
namely, the strongly plastic hair, the softer rendering of the lower part of the face,
and the richer modelling — certainly approach the style which had obtained in Athens
considerably before this time, probably as early as the middle of the century. We
may assume a certain amount of Attic influence in the later period of Polykleitos, and
for this Kresilas may well have been the medium, for we know that he was working
at Argos just at this time. And we must bear in mind that, as I have previously
shown,2 the fragments of sculpture from the Heraion of Argos are worked in a style
in which Attic elements preponderate, and are decidedly akin to the figures on the
Nike balustrade at Athens ; all of which seems to point to the probability that Attic
artists were at work in Argos.
Withal, Polykleitos did not give up one jot of his individual manner in the
Diadumenos ; he only developed it further by bringing to greater perfection the
ideal of abstract beauty which he had in his mind. Looking at the Diadumenos
head we first rightly understand what Quinctilian means when he says that Polykleitos
liumanae formae decorem addidit supra verum? The human form is indeed here
represented with a degree of beauty which surpasses nature. It is not strange that
this statue by Polykleitos should have fetched at an auction the price of one hundred
talents (Plin. xxxiv. 55), an enormous sum in antiquity. Many a smaller, freer imitation,
1 In the Paris copy the nose of the Diadumenos is almost intact. For the Doryphoros our most reliable
evidence is the Naples bronze term.
2 Ath. Mitth. iii. 1878, p. 296 ; Archdol. Studien H. Brunn dargebr. 1893, p. 90. The head lately
ound at the Heraion, which Waldstein considers to be Polykleitan, is more likely Attic. (Cf. p. 223, note I.)
3 Quinct. Inst. Or. xii. 10, 7. Rrunn's translation of the word decor by ' wurdevoller Anstand ' (Gesch. d.
Kiiustler, i. 225) is rather too one-sided. Decor means both 'decency' and 'beauty,' but in this passage stress
should be laid on ' beauty. '

244 POLYKLEITOS
especially on engraved stones,1 testifies to the admiration of the ancients for the head
of the Diadumenos. I may also mention a bronze head of magnificent workmanship
applied to a decorative purpose (De Clerq Coll. in Paris) which is simply a free replica
of the Diadumenos head.2
The one weak point about the statue is the motive. The pause in the act of
walking is not appropriate to the principal action represented. No one walks
along while tying a ribbon round his head. Polykleitos, as we saw in his Amazon
statues, cannot identify himself with his subject sufficiently to create the motive
from the centre outwards. The first consideration for him is the beauty of
rhythmic movement : the meaning of the movement comes second. The result
of this is that the movement is beautiful indeed, but appears unnatural, nay, even
affected. In this respect the Farnese Diadumenos in the British Museum far surpasses
the Diadumenos of Polykleitos by the very simplicity and naturalness of the con
ception. The youth stands still and puts on the fillet just as he would have done in
real life. There is neither stride, nor bend of the head, nor forced attitude of the
right elbow near the body. Nor is there — we must at the same time confess — the
charming effect produced in the other statue by rhythmic motion and beautiful closed
curves.3 The ' Farnese ' Diadumenos is almost universally designated as Attic,4 and a
closer comparison makes it possible to define still more exactly its claim to rank as
such. The head and the bodily forms, the whole appearance and bearing, are nowhere
more closely paralleled than on the frieze of the Parthenon. If we compare it, for
instance, to the standing youth on the west frieze (No. 9), we shall see that the forms
in the abdominal region, the waist, and the lower edge of the ribs have the same
characteristics in both.
Now, since there is a tradition that a statue by Pheidias representing a boy, name
unknown, winding the victor's fillet round his head,5 stood in Olympia, and since the
Farnese figure not only answers to this description but belongs to the very style
which we must assume to have been that of Pheidias in the period when he was
working for Olympia, we are justified in expressing the opinion (held already, though
on insufficient grounds, by Gerhard and Botticher) that the Farnese statue is a copy
of the Pheidian Diadumenos in- Olympia.
The statue fits admirably into the series of Pheidian productions with which we
have become acquainted. The attitude, with one foot set to the side and freed from
the weight of the body, yet resting firmly on the ground (on the inner edge of the
sole, at any rate), recalls the Lemnia and the Anakreon.6 The copyist has treated the
1 E.g. Cades, iy. A, 112, 113 ; further, Berlin, Tdlken, iv. 399 (= Cades, iv. A, 116 ; Jahrb. d. Inst. 1888,
Taf. 3, 16) ; paste in Copenhagen (Ddnische Abdrucks. in Berlin, No. 929) ; an inferior stone in St. Petersburg
(Abdrucks. in Berlin, 23, 2).— For a good copy of the whole statue see Cades, cl. iv. F, 68 = Impronte dell'
Institute, 6, 73. It is an emerald-plasma of the same sort as those with representations of the Doryphoros (Berlin)
and the Amazon (Natter). (Cf. p. 137, note I ; p. 163, note I.) An inferior copy on another gem of the same
kind is published, J. H. S. ii. 352.
2 The neck of an oinchoe rises above the head, and the whole was used as a vase. The eyes are of silver.
The head (called Alexander in the collection) is of extraordinary beauty. It comes from Syria.
3 Cf. Kekule, Idolino, p. 13.
4 Only Brunn identifies it with the Polykleitan Diadumenos type (Annali, 1879, 218). Cf. for a contrary
view Michaelis, Ann. 1883, 154 seq. Kekule (Idolino, p. 12) reckons the Farnese statue among the Myronian
series. • Cf. p. 39.
6 In common with the Anakreon, this statue has the closely curled pubes defined by a horizontal line at the
top, a peculiarity also to be seen in the Dioscuri of Monte Cavallo.

BOXER 245
hair very carelessly ; it seems to have resembled the hair of another Pheidian head
which we conjectured to be an Eros (p. 69).1
It is usually taken for granted that the Farnese type is later than the Polykleitan :
this opinion is founded on the presumption that Polykleitos, because his Diadumenos
enjoyed the most fame, was the creator of the motive. Now it is clear from a con
sideration of form alone that the Farnese type is a pure product of the older style,
as the attitude and the treatment of the hair show most clearly ; further, the natural
simplicity of the motive as it appears in the Farnese statue proves that we have
it here in its original form. Polykleitos borrowed the motive from Pheidias, and gave
it an artistic elaboration which never could have formed part of the earliest con
ception. This order of succession is confirmed by the chronology of the two statues
gained from other sources ; the work of Polykleitos was full fifteen years later
than that of Pheidias.
For the invention of the motive, then, Pheidias is responsible. We can well
believe that he was pleased with his subject, since he repeated it among the figures
on the throne of Zeus.2
We may connect with the Diadumenos on account of the similar motive a statue
of a youthful boxer, also a creation of Polykleitos (Fig. 99).
Unlike the Doryphoros and Diadumenos, which are above life-size, this figure
is rather under life-size. It represents a youth just out of boyhood — that is, with more
undeveloped and less powerful forms than the other two. The motive is again a
pause in the act of walking, the weight being supported on the right leg. The head
is turned to the right. As in the Diadumenos, the arms are both raised and bent,
but there is a more marked difference between the attitude of the two arms : the
right is lower, the left higher. The forearms are wound round with the boxing straps,
the right fist ready for attack, the left for defence.
I know two copies of this work — one in Cassel, with the original head (Fig. 99),
and one without a head in Lansdowne House.3 The second gives the body better
than the first. The bodily forms correspond with those of the Doryphoros and
Diadumenos in all essentials (the abdomen and navel are very characteristic), except
that they are more youthful and boyish.4 The head, which survives only in the
feeble sketchy Cassel copy, has hair arranged as in the Doryphoros.
The motive, in its fulness of dignity and balanced harmony, is incontestably fine,
though for a real and energetic expression of this brutal sport we must turn to those
1 The head, which is wrongly placed on the Penelope of the Vatican (Ant. Denkm.), is apparently, as far
as the hair is concerned, a more accurate replica of the head of the Farnese statue. The principal proportions
(fillet to chin 13^ cm.) correspond, although the eyes of the Penelope head are smaller and flatter, and the
mouth is narrower. In these points the Farnese head is the more correct, but it gives only a sketchy rendering
of the hair, which is more exact in the ' Eros ' head.
2 Cf. p. 39. For a similar motive cf. a goddess on the frieze of the Athena Nike temple (No. 12, called
Amphitrite by Sauer, Aus der Anomia, p. 94 seq. )
3 The Cassel statue (ii. 17) is poorly illustrated in Bouillon, iii. Statues, PI. 17, I. Parian marble, the
same in the head and body. The neck is restored, but the head seems genuine. The turn of the head is known
from the one piece of the neck attached to the torso, and another piece attached to the head. The head was
probably a little more bent. The left arm and the right shoulder are restored, the right arm and hand are ancient;
the legs and basis are modern. In the Lansdowne statue (Michaelis, p. 438, 3 ; Clarac, 851, 2180 A), only the
torso survives ; the head is foreign ; the left shoulder with the beginning of the raised arm is ancient ; the rest of
the arms and the legs from the knees downwards are new. The proportions correspond with those of the Cassel
statue; distance between nipples 21 cm.; neck-pit to navel 31 ; neck-pit to line between nipples 13^ ; from the
latter to navel 175. The head of the Cassel copy is 19 cm. high; face-length 13^, nose and lower face
each si cm.
4 The pubes is accordingly but slightly developed.

246 POLYKLEITOS
violently agitated statues of severer style which we have ascribed to Pythagoras
(p. 172). A bronze statuette found on the Akropolis x repeats the motive without attempt
ing to be an exact copy. The right fist is clenched, but without the straps ; the left

Fig. 99.— Statue ot a boxer (Cassel).
is missing. The severe weathering the work has undergone has effaced most of the
detail, but the back and the glutei recall Polykleitos. The bronze should be dated
about 400 B.C.
1 Akropolis Museum, among the older discoveries. Height 12 cm. Left hand broken. This bronze may
be said to bear the same relation to the Cassel boxer which the Janze statuette does to the Diadumenos.

THE AMAZON 247
A marble torso in Florence, on the scale of the Doryphoros and analogous to it
in modelling, is derived from a work the motive of which was very similar to that of
the boxer. This was probably a more recent creation of the school of Polykleitos.1
Finally, the Roman artists made use of the Diadumenos, as they did of the
Doryphoros, for portrait-statues.2
A bronze statuette in Carlsruhe, with an Etruscan inscription, evidently a
derivative from the Diadumenos of Polykleitos, seems to me to be a forgery.3
In conclusion, and still in connexion with the Diadumenos of Polykleitos, we may
mention a head which, although hitherto almost unknown, is one of the most beautiful
antiques in existence. It is a head of a youth in the Turin Museum. The hair is
full and curly ; a peculiar plaited fillet is imbedded deeply in the hair and tied
behind.4 The head is turned to the right and slightly bent; the mouth is a little
open. The hair is parted in the middle in Polykleitan fashion, but the locks are
much fuller than even in the Diadumenos.5 The ears are covered by a thick mass of
curls, but the curls on the nape of the neck which appear under the fillet are remark
ably like those on the Diadumenos, which the Turin head surpasses however in
sweetness and serenity of expression. Whether it was Polykleitos himself who made
this great advance on his own Diadumenos is a question impossible to decide. In
any case, the original of the Turin head must have been the work of an artist of the
first rank, who derived his inspiration from the Polykleitan statue.6

IV. The Amazon.
The Amazon is the third of those works of Polykleitos which are universally
acknowledged to exist in copies. We have already discussed this figure in connexion
with Kresilas and Pheidias (p. 128 seq.): it now remains to examine its relation to
the other statues of Polykleitos.
In style the Amazon is nearer to the Doryphoros than to the Diadumenos : the
head {supra, Fig. 55) shows this very clearly. The hard and prominent lower part of
the face, the mouth projecting strongly in the middle and receding towards the
sides, the simple flat-lying hair, connect the Amazon closely with the Doryphoros,
and distinguish it from the Diadumenos. Even in attitude the Amazon is more
like the former of the two, for the left forearm is bent and stretched forward hori
zontally, the thighs run parallel, the point of the left foot not being turned outwards.
This conclusion is borne out by what we know from other sources as to the date
1 Uffizi, 67 ; Diitschke, 76; Photo. Alinari, i. 11 79. The head is foreign ; both arms new ; the left was
raised, the right hung down. The palm-stem indicates a copy of a ' victor ' statue.
2 The statue in Naples, Inv. 6271, evidently makes use of the Diadumenos, yet the turn of the head
(now replaced by a restored one) is altered, and the round chlamys is added on the left shoulder. A dolphin
and a polypus on the stem. Plainly a Roman portrait. The statuette, Museo Torlonia 72, is influenced by
the Diadumenos ; the arms wrongly restored ; neck modern ; the head resembling Tiberius is antique, and
probably belongs to the statue.
3 Schumacher, Beschr. d. Ant. Bronzen, 932. From the style, especially of the head, and some external
technical signs, I am inclined to suspect that the figure is not genuine.
4 Diitschke (iv. 52), who describes it as a female head ! Marble Parian. The head is placed on an ugly
nude bust. The nose, part of both lips, and the chin are restored. Face-length 19 cm. ; inner eye-corner to
chin = hair to nostril I2| cm.
6 The full curl with rolled end is one of the chief motives. On the top of the head the hair is only loosely
indicated. 6 In presence of the original, Kresilas occurred to me, but I cannot prove the connexion by comparison, as I
was unable to obtain a photograph.

248

POLYKLEITOS

of the Amazons, falling about 440 B.C. Our Amazon would thus come near to the
Doryphoros, and be separated from the Diadumenos by a considerable period
of time. The Amazon of Polykleitos, as we have seen, expresses in one figure the Doric
KocrpioTrji; and the Doric tcapTepeiv. The wounded heroine is not mastered by pain :
her attitude shows weariness only,1 her appearance and clothing are faultless in
beauty and order. The lines of her form are rounded to a perfect harmony. Here
too the motives are disposed in a crossing scheme : the raised arm is balanced by the
pillar on the other side ; the right leg and the left arm are the carrying, supporting
parts, the right arm and the left leg the parts at rest.
The motive which shows the left arm supported while the right is raised was
a favourite in Praxitelean times. Polykleitos seems to have invented it for his
Amazon,2 though the left arm leans but lightly and the rest of the body is disposed
as in the Doryphoros. We should expect to find other instances in the Polykleitan
circle of this ' leaning ' motive, and there are, in fact, some other traces of it.
A Greek engraved stone of the fifth century, from the period of Polykleitos
himself (Fig. 100),3 represents the youthful Herakles, in an attitude corresponding in all
essentials to a Herakles torso of Polykleitan
character at Dresden.4 The hero is naked
and without the lion-skin ; the weight is
supported on the right leg, while the left
is drawn back in the walking motive ; the
club is shouldered and held by the right
hand, but the left forearm is supported, as
in the Amazon, on a low pillar. The lion
stands beside the hero as an attribute. The
whole design is clearly taken from a statuary
composition of the period — perhaps of the
school — of Polykleitos.
A second gem, this time of Roman date (Fig. 101),5 represents Hippolytosin the
walking motive of the Doryphoros, but with the left forearm, just as in the Amazon,
supported — in this case against a tree ; the lowered right hand holds a hunting-spear ;
below is a dog. The figure is evidently derived from a statue : the heavy forms of
the body, the head with its large skull and close short hair, point to a work of the
Polykleitan circle. The Eros standing on the tree behind and reading the love-letter
on the diptyckon is of course an addition of the gem-cutter, who wished to make clear

Fig. too. — Greek gem.

Fig. ioi. — Carnelian in
St. Petersburg.

1 Wound and support are certainly genuine. Cf. supra, p. 134, notes 5, 6.
2 Cf. Robert, Arch. Mdrchen, p. 109 ; but Polykleitos need not therefore be dated later.
3 Cades, CI. iii. A, 110. From the style, the above date is more likely to be too late than too early.
Remnants of severe style ; a line round the edge. Species and owner unknown.
4 Recent purchase ; till lately in Rome ; Arndt-Bruckmann, Einzelverkauj , No. 184. Preserved till below
the navel. The connexion with the Doryphoros is striking. I had myself imagined it might represent the statue,
to which the type of Herakles' head (Fig. 95, 96) belonged ; the measurements, which are practically the same as
in the Doryphoros, are however too large. It is more probably to be restored on the analogy of the ' Theseus
term (of course no Theseus, but a Herakles) in the Mus. Boncompagni-Ludovisi (Helbig, Museums, 861 ; Mon. d.
Inst. x. 57, 2). The artist of the original had affinities to Kresilas, but was influenced by Polykleitos in the bodily
forms. Breast and abdomen are quite Polykleitan, though the transitions are rounder, the edge of the ribs less
accentuated, and the flesh softer. The right hand shoulders the club, the left hangs down, the head (differing in
this from the Doryphoros) was turned to the side of the free leg, according to the Attic fifth-century scheme.
Measurements : distance between nipples = 0-300 (Doryph. = 0-303) ; neck dimple to centre of navel = 0-455
(Doryph. = 0-450) ; breadth of hips, 0389 (Doryph. 0-393).
5 Carnelian in St, Petersburg (Berl. Abdr. 19, 24 = Cades, i. K, 91).

KYNISKOS 249
that the figure represented Hippolytos, and no common huntsman. The original statue
probably stood in Troezene, for on coins of that city the same type, though roughly
reproduced, occurs, the only difference being that the right arm is raised higher.1
The motive of the supported left forearm was used in the circle of Pheidias
combined — not, of course, with the stepping motive, but with the left foot set slightly
forward and full on the ground. On p. 71 we have tried to prove the existence of an
Aphrodite by Pheidias represented in this motive.

V. The Basis of the Statue of Kyniskos. — Statue oj a Boy placing a Wreath on
his Head, and Kindred Works.
A sure foundation for the knowledge of Polykleitan statues is afforded by the
discovery at Olympia of the inscribed bases that supported some of his works,
inasmuch as the extant footmarks on these bases enable us to recover the attitude
in which the statues stood. Three of them belong, as we saw, in all probability to
the elder Polykleitos, one to the younger.2 What first strikes us on examining the
footmarks on the Polykleitos bases is that in not one of them does the scheme of the
' canon ' occur — i.e. there is no single instance of a right foot bearing the weight of the
figure combined with the left foot drawn back and resting only on the ball. Nor
does this motive appear on the bases of works by Naukydes or by Daidalos ; only
the Zan of Kleon (01. 98) 3 and the following Zanes, so far as the footmarks can be
made out, illustrate this scheme. The obvious conclusion is that Polykleitos himselt
as well as his school made use of other motives besides the one which we associate
with the Doryphoros, the Diadumenos, the Amazon, and other analogous works.
The basis of Kyniskos, the earliest of them all, shows that the left leg bore the
weight of the figure, while the right foot was set back and rested on the ball ; 4 the
statue represented a boy victorious in the boxing match.
Now we actually possess numerous copies of a statue of Polykleitan style repre
senting a boy victor, in an attitude that corresponds closely with the footmarks on the
Kyniskos basis (Fig. 102). To identify the statue with the ' Kyniskos' of Polykleitos
would be no far-fetched theory, and I have long thought it worthy of adherence :
1 Imhoof-Blumer and Gardner, Num. Comm. PI. M. viii.
2 We should have a fourth inscription of the older artist, could we see grounds for accepting Benndorf's con
jecture (Ges. Studien zu Kunstgeschichte, Festschrift f. Springer, 1885) that the basis in the shape of an astragalos,
found in Olympia, formerly supported the nudum talo incessentem of Polykleitos (Pliny, xxxiv. 55) ; there are,
however, weighty reasons against this identification. Pliny's words mean literally ' a naked man advancing with
a knuckle-bone.' Benndorf supposes that Pliny wrongly translated some Greek sentence such as yv/ivbs dtrrpaydXtp
iiriKei/ievos, and made nonsense of it. But even if a Greek sentence bearing the meaning supposed by B. ever
existed (which seems doubtful), and if Pliny did understand iiriKei/ievos in the sense of 'advancing, pursuing,' he
must have connected the word with darpayaXip, and explained the whole phrase as ' advancing towards an astragalos. '
There is no hint of this meaning in the Latin talo incessens ; talo must be the instrument, the weapon, with
which the nude man is advancing. As a knuckle-bone cannot be a weapon, there must be an error somewhere.
If_we read telo for talo (Benndorf's own former suggestion) all difficulty disappears (cf. Ovid, Metam. 14,402,
saevisque parant incessere telis, 13, 566 ; telorum lapidumque incessere iactu coepit). The corruption of the passage
is then explained by the talis ludentes immediately following, from which talo was transferred to take the place
of telo ; the item nudos refers to the preceding nudus. The evidence, then, does not permit us to identify the
bronze statue set up at Olympia on a marble astragal with a work of Polykleitos. The statue may, as Benndorf
suggests, have represented Kairos ; but it was very possibly only the portrait of a human being who, by the shape
he adopted for the basis of his statue, recalled the particular stroke of good fortune which had moved him lo bring
a thank-offering to the divinity. As the attitude of the feet corresponds to the Kyniskos basis, the work belonged
probably to the Polykleitan circle.
3 Ldwy, Inscr. gr. Bildh. 95.
4 See drawing of upper surface of the basis, Arch. Ztg. 1882, 189 ; Lowy, 1. G. B. p. 43, No. 50.
K K

?S0

POLYKLEITOS

lately both Petersen and Colhgnon have expressed
the same opinion.1 Unfortunately, however, abso
lute certainty can scarcely be attained,2 as it is
no longer possible to measure accurately the length
of the foot of the Kyniskos figure ; in any case,
it must have been more than 20 cm. ; 3 the length
of the foot of the statue is 233 mm. ; the distance
between the left heel and the ball of the right
foot is, in the statue,4 16 cm. ; the two correspond
ing holes on the Kyniskos basis are about 175 mm.
apart. These measurements are very nearly alike
in the basis and the statue. The boy Kyniskos,
like the statue, was of life-size, and his feet were
exactly in the same position.
Two replicas of this statue have long been
known. One is the so-called ' Westmacott ' athlete
in the British Museum, the other is in the Barracco
collection.5 The London copy is undoubtedly the
better of the two. The Barracco figure is dull and
slight in workmanship ; the London replica re
produces the body more sharply and finely, and
the head especially with much more exactitude,
The hair in the Barracco copy is carelessly ren
dered by mere indications of superficial effect ; in
the London copy there is plainly a conscientious
imitation of the bronze original.
To these two copies may be added a whole
series of others. These are, it is true, mostly
torsos and separate heads, but the very fact of their
existence in such numbers proves that the original
was famous and by a celebrated artist. The
following are exact replicas : a statue in the garden of the Palazzo del Quirinale in
Rome ; the head has never been broken.6 It is not so good as the London copy, and
1 Petersen, Rdm. Mitth. 1893, P- IQI seq.; Collignon, Hist, de la Sculpt, i. p. 499.
2 Petersen (loc. cit) thinks the identification more certain than I could venture to assert, as there must
have been several Polykleitan statues of boys with the same position of the feet. Further, it is hardly
likely that the original of our favourite and oft-copied statue was still in the Altis in the time of Pausanias, as
the Kyniskos was, on whose basis there are no traces such as occur on the base of the Pythokles to show that the
original had been taken away.
3 Cf. Purgold, Arch. Ztg. 1882, 192. The distance between the two holes for the insertion of the left foot,
probably meant for the toe and heel, is about 20 cm. ; the lower hollow may have been quite close to the point of
the toe. Hence the foot need not have been more than 23 cm. long.
4 In the London copy. Cf. following note.
5 For discussion of these statues see Petersen, Arch. Ztg. 1863, 131 ; Helbig, Bull. d. Inst. 1885, 76;
Winnefeld, Hypnos, p. 30 seq. ; Treu, Arch. Anzeiger, 1889, 57; Kekule, Idolino, p. 13 ; Wolters in Lepsius,
Marmorstudien, p. 83, No. 164; Philios, 'Eip-it^phkpx. 1890, p. 207 seq.; Sauer, ibid. p. 211 ; Petersen, Rdm.
Mitth. 1891, p. 304 ; Collignon, Hist, de la Sculpt, i. 499 ; Helbig, Rendiconti deW Accad. dei Lincei, 18 ;
Die. 1892, p. 790 seq., • sopra un tipo di Narcisso' ; Milchhofer, Arch. Studien H. Brunn darg. 1893, p. 63;
Petersen, Rdm. Mitth. 1893, p. 101.— Photograph of the Londonstatue in Brunn-Bruckmann, Denhndler, No. 46 ;
of the Barracco statue in the Collection Barr. PI. 38, 38 a. Sketches of both in Kekule, Idolino, Taf. iv., and in the
'Ecpnp.. p. 207.— In the London statue the head is unbroken, and even the nose ancient. The left hand is broken
but ancient. The execution is not slavish in detail, but gives the broad general characteristics ; the marble is Parian.
6 Matz-Duhn, 210. On a fountain, drapery round the loins restored. The right arm may, I think, be antique.

Fig. 102. — Statue of a victorious boy placing the
wreath on his head. (Restored.)

WESTMACOTT ATHLETE

251

comes nearer to the Barracco head. The right arm is joined, and perhaps antique.
The hand makes a movement towards the head. There are two torsos in Berlin one
of them specially good,1 another in Dresden,2 of poor and exaggerated workmanship.
The Museo Torlonia possesses a very good torso 3 restored as a Hermes ; the nipples
imitate bronze technique. A poor
copy in the Villa Albani is restored
as a Brutus by the addition of a
portrait-head representing a Roman
of the end of the second century.4
Sir Edgar Vincent possesses a
good copy of the head alone (Fig.
103),5 preferable even to theLondon
one. Mouth and cheeks, particu
larly, have kept a simplicity and a
severity which can only be due to
the original. The hair, though in
dividual enough, is not 'worked with
much detail. The style is distinctly
Polykleitan. This is still more the
case in a copy of the head in the
Hermitage (Fig. 104),6 by far the
best that exists. Here the hair is
carefully imitated from the bronze
original : the way in which it is
disposed in layers, the shape of the
separate strands, especially behind
and in front of the ear, are so
entirely Polykleitan, and corres
pond so closely to the Doryphoros,
that the work must be ascribed
to Polykleitos, and dated not far
from the Doryphoros. Entirely
Polykleitan also are the angular build of skull, the large planes of the cheeks,
the protruding mouth,- the formation of the lower part of the face, and the regular
forehead only slightly projecting at its base. It seems that a good replica of the head

Fig. 103.— Head of boy'iin possession of Sir Edgar Vincent).

1 Berlin, Skulpt. No. 514, is a very well executed, exact copy. A second copy in Berlin has restored head
and limbs.
2 Hettner's Catalogue, No. 254. Cf. Treu, Arch. Anz. 1889, 57.
3 Mus. Torlonia, No. 59.
* Villa Albani, No. 46. Puntello for the left arm in the usual place.
" Formerly in the Van Branteghem Coll. ; from Italy. The right side of the head is less finished than
the left. 6 Guedeonow, No. 28. Bust and neck restored, otherwise in good preservation. Admirable workmanship.
Cited by Conze, Beitrdge, p. 7, as analogous to the Doryphoros. That it is a replica of the Westmacott
athlete is proved by the correspondence in scale, the vigorous turn of the head, and the similarity in the
hair. Even the bad copies, like the Barracco head, have the characteristic wisp of hair in front of the left ear.
The rest of the hair matches the Westmacott copy strand for strand. The proportions of the St. Petersburg
head are : length of head, 19 cm. ; length of face, 148 mm. ; length of eyeball, 26 — 27 mm. ; width of
mouth, 39 ; inner corner of eye to hair, 58 mm. ; length of ear, 54 mm. The face-length is an important
standard measurement for the torso ; from the ensiform process of the breast -bone to the navel, and from the
navel to the insertion of the penis, are each one face-length. The^chest, from the neck -pit to the ensifonn process,
is a little more.

2C2

POLYKLEITOS

stands in the Museo Torlonia.1 I know a mediocre one from a cast in Strasburg,2
and there is another in the Lateran.3
The Polykleitan origin of the statue, placed beyond a doubt by the good copies
of the head, is confirmed by the modelling of the body. The points which distinguish
it from the Doryphoros and the Diadumenos
are only a consequence of the difference
of age in the persons represented ; the
style is absolutely the same. The large
surface planes clearly marked off from
each other are characteristic, and so are
the details of form, such as the flat navel
and its surrounding parts, the central
groove or linea alba below the navel, and
the well-marked hollow at the side of the
gluteus,4 a point on which Michaelis lays
stress as being specially characteristic of the
Doryphoros and the Diadumenos. In both
cases it is striking that the hollow is equally
pronounced on the side of the ' free ' leg
as on that of the ' supporting ' leg. Michaelis
was doubtless right in calling this a personal
trait. Other peculiarities of bodily forma
tion, cited by Michaelis as common to the
Doryphoros and the Diadumenos, are here
reproduced as nearly as they can be con
sidering the more tender age of the subject :
such are the modelling of the knee which
bears the weight of the figure, and the
strongly marked roll of muscle over it.
Again, the back of the figure (Fig. 105) 5 must be studied in order to understand that
its distinguished beauty is in complete accord with the Doryphoros and Diadumenos.
The head proves that this statue must be dated much nearer to the Doryphoros
and the Amazon than to the Diadumenos — that is to say, that it belongs to the period
about 440 B.C. The hard, spare modelling and the flat rendering of the hair differ
as much from the Diadumenos as they resemble the Doryphoros and the Amazon.
It is also worthy of notice that the depression marking the crown of the head is as
shallow as in those two works, and that, just as in the Doryphoros, the hair comes
barely to the edge of the ear. Finally, the walking motive is expressed as simply as
in the Doryphoros and the Amazon, both thighs being in the same plane.
The motive of the statue is allied in meaning to that of the Diadumenos.
The one binds on the victor's fillet, the other presumably places the wreath, also a
token of victory, on his head 6 — at least no more suitable restoration than this could

Fig, 104.— Head of a boy (Hermitage).

1 No. 474. The head is placed leaning too far back.
2 No. 597, Michaelis. I do not know where the original is.
3 Lateran, No. 498. Neck new ; head wrongly set on. Petersen (Rom. Mitth. 1893, p. 101) mentions
two further replicas : (1) a lower part of the face from the Palatine ; (2) a torso at Marinangeli's. I have seen
neither. * Michaelis, Ann. d. Inst. 1878, p. 16.
5 Cf. Michaelis, loc. cit. 17. Fig. 105 is from a photograph of .the London statue.
6 Winnefeld (loc. cit.) first suggested this, then Wolters, Treu, Petersen, Collignon (loc. cit.), and, hesitatingly,
Kekule. Philios, Sauer, Helbig, and Milchhofer disputed it.

Fig. 105. — Back of Westmacott athlete (British Museum).

2c;4 POLYKLEITOS
be found for the Barracco copy,1 where the arm is preserved down to the wrist, and
the hand was not raised but slightly bent ; further, a whole series of monuments which,
as is evident by the correspondence in their attitudes, must be regarded as free
imitations of our statue,2 show the boy in the act of placing a wreath on his head,
and the whole posture corresponds to this motive.3 The bent carriage of the head is
no objection, for it is evidently meant to express the modesty which the artist, in the
true spirit of antiquity, conceived as the fairest ornament of victorious youth. That Poly
kleitos actually did think of the subject in this way seems certain from the evidence of
the Diadumenos, where the act is the same, and the head is bent. This is one of many
coincidences which intimately connect our statue with Polykleitos. The rest of the
body also is in a purely Polykleitan attitude ; besides the walking motive, we should
notice the arm hanging idle with open fingers by the side of the supporting leg.
A new feature, and one not found in any of the Polykleitan statues we have
hitherto examined, is that the head has a decided turn away from the side which
supports the weight of the body. This breaks the central line of the figure, and
makes the rhythm altogether more animated.
The Amazon of Kresilas shows such a striking resemblance in attitude to this
statue4 that we might almost suppose that Polykleitos is here influenced by the Attic
master. The reverse is certainly not the case; for the motive of the boy is slightly
artificial, and does not follow necessarily from the action as does the motive of the
Amazon. The boy might just as well look to the other side ; not so the Amazon, for
all her activity and emotion are centred on the wound. Again, the left arm of the
boy is not included in the activity of the body, but hangs carelessly down, following
the ' canon ' of Polykleitos, while to the corresponding arm of the Amazon a necessary
share in the action is assigned. This is an instance of the contrast between the two
masters. The motive can only have been invented by the artist who makes it appear
a natural consequence of the situation. Polykleitos adapts part of it merely for the
love of beautiful form.
One figure of the Parthenon frieze 5 is very like the Polykleitan statue, only that
the movement is more animated, the sides are reversed, and the one arm does not
hang down, but leads a horse by the bridle ; the left hand was arranging the wreath ;
a direct connexion between this figure and the Polykleitan statue, which was the
earlier work, is not at all impossible.
1 The bend of the hands is shown by the folds in the skin underneath the remaining portion ; the hand must
have held something over the head. The wreath was probably of bronze, even in the marble copies.
2 I mean the monuments which correspond with our statue in having left supporting leg, the walking
motive, the hanging left arm, and boyish forms : e.g. the Berlin relief, 948 ; the Vatican relief, Mus. Pio Clem.
v. 36 ; the Zodiac relief, Le Bas, Mon. Fig. 21 ; Campana, Opere in Plastica, 94 ; an onyx gem in St. Petersburg
(cast in Berlin, 26, 56), motive well rendered ; another gem, where the motive is adapted for an Eros victorious, in
the Palestra (Cades, iv. F, 18). In all cases the boy is putting on a wreath with the right hand, the left occa
sionally holds a palm-branch. For modifications in the ' crowning ' motive, cf. p. 255.
3 The recent suggestion that the boy is holding a strigil in his raised right hand (Sauer and Philios, loc. cit.)
is certainly wrong : the strigil would come just over the forehead and in front of the hair ; looked at from below, it
would seem held even higher. Now a strigil was never used for cleansing the hair. The Pompeian painting (Rdm.
Mitth. 1888, p. 199) on which the suggestion is based represents the athlete holding the strigil to his forehead ;
moreover, the weight is on the right leg, the forms are more adult ; the whole attitude different. Still less happy
is Helbig's most recent suggestion, that the hand is raised to shade the eyes from the sun. The lowered head
speaks against this view. The motive of shading the eyes with the hand is only natural, and as far as I know occurs
in ancient art only when the eyes are looking towards the light, i.e. towards the horizon or the sky, and never
when they are looking down to the ground as here. Nor is there any foundation for Helbig's interpretation of
the figure as Narkissos. The left hand is complete in the London statue : it was empty, and could never have
held a spear as Helbig suggests.
4 Kekule (Idolino, p. 14) noticed this. 5 North frieze, 131 (Michaelis).

BOY CROWNING HIMSELF 255
A statue found within the sanctuary of Eleusis x seems to be an approximately
faithful rendering of the body of the original, although on a somewhat smaller scale ;
but the head, which seems less inclined, is rendered much more freely, in the Attic
marble style of the good period.2 The left arm may have been in a different position,
as the puntello at the usual place is missing. The statue could hardly represent a
victorious athlete ; the provenance and the soft expression of the face suggest rather
some divine boy, perhaps Triptolemos. A torso in the Museo Torlonia3 would be an
exact copy, except for a large round fracture-surface on the right gluteus, indicating
that some object either touched the body or was connected with it by means of a
support. A torso of the Villa Albani 4 is another ordinary copy ; but an antique piece
of a boar's head on the tree-trunk beside the left leg shows that the type was used for
a youthful hunter — Adonis more probably than Meleager. The type was also adapted
to Dionysos, as we learn from a torso of the Museo Torlonia ; this is an exact copy,
with the addition of a nebris reaching from the left hip to the right shoulder, the
ends of which probably fell over the arms.5 The statue has been made into a
Hermes. A torso in the Terme resembles the usual copies in all respects, except that
a chlamys of good Greek fashion is laid on the left shoulder and falls down over the
arm, while the left hand holds the kerykeion.6 Unfortunately in all these torsos only
enough is left of the right arm to show that it was raised. We do not know how the
motive was accounted for ; the placing of a wreath would be appropriate only for a
victor, and cannot be assumed for other subjects.
One variant of the ' crowning ' motive is to be seen on a coin of Commodus 7
representing the youthful Herakles in the same position as our figure, but holding a
club and a skin in the left hand. A coin of the Brettii 8 uses the motive for a young
hero who carries lance and chlamys in his left hand.
The extant modifications in sculpture — for Triptolemos (?), Adonis (?), Dionysos,
and Hermes — may be mostly referred to the good period of Greek art. Specially
interesting is a coin of Troezene,9 which evidently reproduces a statue of the city. It
supplies a much-needed restoration for the Hermes torso in Rome mentioned above.
The motive is that of the victorious boy, only that chlamys and kerykeion are held in
the left hand ; the right is brought close to the head, and is empty. The Eleusinian
figure does not, then, stand alone as witness to the fact that the graceful motive of the
victorious boy was adapted for representations of a religious character to be set up
in Greek sanctuaries. Finally some gems show that the essential features of this
beautiful motive reappear in other representations of the athletic circle.10
1 'Etpri/x. apx- 1890, Taf. 10, 11. (Philios, p. 207 seq.) The body was found on the procession street between
the Propylaia and the Telesterion, the head a little farther away.
2 The statue is no Roman copy ; judging from the workmanship, it appeared to me to belong to the fourth
century B.C. It may come from the Praxitelean circle, from the period in which the master was an enthusiastic
follower of Polykleitos (cf. infra, p. 317).
3 Mus. Torlonia, 37. The torso unrestored : the puntello for the left arm is at the usual place.
4 Villa Albani, 222, ' Meleagro. ' Head foreign ; arms and neck restored.
6 Mus. Torlonia, 22. Only the torso (including the right thigh) is ancient : the nebris is only indicated in
low relief on the back.
6 Museo delle Terme, in the cloister garden. Proportions of torso correspond with the other copies. Torso
unrestored, and the greater portion of the thighs remains. The attitude of the head and the right arm as usual.
7 Frohner, Midaillons Rom. p. 123.
8 Garrucci, Le Monete, Tav. 124, 13, 14. The youth is crowning himself. Garrucci's explanation as Pan is
wrong. a Imhoof-Blumer and Gardner, Num. Comm. PI. GG. xi. p. 162. Bronze coin of Septimius Severus.
10 Fasten in Berlin (pp. 2377 and 2378) : athlete holding discus in his lowered left hand, the right hand,
empty, moved towards the head. Attitude of head and legs as in the statue. — Carnelian, Brit, Mus. (Cat. of

256

POLYKLEITOS

A famous painting by Eupompos,1 of Sikyon, representing a victor certamine
gymnico palmam tencns, seems to have been directly derived from the Polykleitan
statue. The youth holds a palm-branch in his
left hand, and the head is not bent, but in an
upright attitude more appropriate to a painting.
Eupompos seems to have followed the same ten
dency in painting as Polykleitos in plastic art.
When Polykleitos was at the end of his career,
and his disciples were beginning to migrate to
Sikyon, Eupompos founded in that city a school
of painting on much the same lines as the school
of bronze statuary founded by Polykleitos. It is
therefore not surprising that Eupompos should
have followed Polykleitos very closely. Invention
was in any case not his strong point.
This beautiful Polykleitan motive seems to
have been adopted even by a modern painter:
the Christ rising from the tomb by Fra Barto-
lomeo in the Pitti Palace 2 so strikingly recalls it
that it is probable the artist had seen one of
the copies of the statue.

The same position of the feet visible on the
Kyniskos basis, and which we found to be that
of the 'boy crowning himself preserved in the
Westmacott statue and its replicas, occurs on
two other bases at Olympia, which probably sup
ported statues of the Polykleitan school. These
are the basis in the form of an astragalos (p. 249,
note 2), and the basis of Hellanikos of Lepreon,3
who as a boy was victorious in the boxing contest
(OI. 89 = 424 B.C.) The footmarks are very similar
to those of the Kyniskos ; here also the boy was
represented life-size : the left foot, firmly placed
on the ground, was 22 cm. long;4 the right was
set back and rested on the ball ; the stride was
longer than the stride of the Kyniskos and of the

t'ltj. 106. — Figure from Ildefonso group (Madrid).

boy crowning himself,' and the right foot was more to the side (the distance from

Gems, 1833) ; the youth holds the wreath in the lowered left hand, and with the right he leads a horse by the
bridle ; position of head and legs as in the statue ; a victor with his horse. — Gem, Cades, CI. iv. F, 46 ; the left
hands holds a long palm-branch, the right is raised and placed on an urn over a fountain.
1 Milchhdfer, in Arch. Studien H. Brunn dargebr. 1893, p. 62 seq., maintained the opinion that the athlete
putting a crown on his head with the right hand and holding a palm in the left, who is repeated frequently and
with many variations in relief and painting, is to be referred to the work of Eupompos. This is no doubt correct.
But the essential features of this composition as seen in their simplest form in some of the replicas (cf. p. 250,
note 5) are those of this Polykleitan statue, which therefore must have afforded the model for the painter's work.
2 Bruckmann, Klassischer Bilderschatz, No. 373.
3 Arch. Ztg. 1878, p. 88, No. 138. Treu has rightly observed that the inscription only survives in its restored
condition, Arch. Ztg. 1879, p. 208.
4 The length of the foot can be exactly calculated, as beyond the depression for the main part of the fool
Ihe outline of the toes is still visible.

ATHLETE AT PETWORTH 257
the left heel to the ball of the right foot being about 245 mm.) In exterior the
basis is very like those of Polykleitos and his circle ; the artist is not named.
Among surviving statues there is no doubt that one figure of the Ildefonso L
group, that of a boy wearing a wreath (Fig. 106), should be reckoned as a work akin
to the statue under discussion. In the position of the legs it corresponds exactly,
and in the attitude of the head approximately, to the Westmacott statue, but the
arms are different ; the right hand hung down, the left was raised. The head is of
purely Polykleitan style, very like the Doryphoros in the hair and the modelling of
the forehead, while in expression and formation of the lower part of the face it stands
very near the best copies of the ' boy crowning himself.' The eyes look modestly
to the ground. In the original, the raised left hand may have touched the wreath
which it had just placed on the head, and the right may have held a fillet. On the
other hand, the body is much softer and feebler than we have any right to presuppose
was ever the case in Polykleitan works. We may conjecturally charge the copyist
with the alteration, and assume that the original was by Polykleitos. I know no
other replica of this work.
We possess still another interesting Polykleitan statue of an athlete with the same
pose of the feet as that shown by the Kyniskos basis, i.e. with the weight on the left
leg and the right foot drawn back ; but the head is not turned, as in the other statues
which fit the basis, to the side of the free leg, but according to the ordinary norm, to
the side which supports the weight. The work survives only in one copy, the ' Oil-
pourer ' at Petworth House in Sussex (Fig. 107, after a photograph from the original).2
The statue is remarkable more for excellent preservation than for good workmanship.
The head and both arms, even the right hand with the globular aryballos, are
ancient ; the legs only are restored from below the knee.3 The tree-stem, with the
two objects (probably boxing pads) hanging on it, is also ancient. The right leg
from below the knee, reckoning by the surviving knee-joint, was drawn farther back
than the restorer has assumed, and was in the full 'walking' position. The whole
figure ought to be tilted a little more forward. The athlete is past boyhood ; he '
is an ephebe, the forms of the body being distinctly more powerful than in the
' boy crowning himself 4
Polykleitan style is unmistakable, especially in the head.5 Above the forehead
there is the characteristic symmetrical parting of the hair just as in the Doryphoros.
The forehead is divided horizontally, and modelled like the forehead of the Doryphoros ;
the eyes are purely Polykleitan. The lower part of the face, with the slightly open
mouth, is most like the Dresden athlete which we are presently to discuss (Plate XII.)
In bodily formation, too, our statue resembles this athlete, though the forms are
more powerful to correspond to the advance in age. On the other hand, a great
interval separates it from the ripe adult forms of the Doryphoros and Diadumenos.
1 For literature cf. Friederichs-Wolters, 1665 ; Museo Espagnol, ix. p. 217 seq. with plate ; Hauser, Neuattische
Reliefs, 184 ; E. Bethe in Arch. Anz. 1893, p. 8. The raised left arm with the torch is certainly a restoration—
cf Bull. d. Inst. 1877, 154.
2 Described by Michaelis, Anc. Marb. in Gt. Brit. p. 601, No. 9. The statue had been published before
in Bracci, Memorie degli Incisori, i. PI. 26 to p. 293, • Atleta in Londra.'
3 The only other restorations are the end of the nose with right nostril and the projecting finger of the left
hand. The head has not been worked over.
4 The pubes is beginning to grow : it is formed of two symmetrical pairs of curls on either side. The
undeveloped pubes here only characterizes the youthful time of life, and is totally different from the small pubes
which appears on adult figures of the fourth century.
0 Michaelis, loc. cit.
L L

Fig. 107.- Athletein the collection at Petworth House (Sussex).

ATHLETES POURING OIL 259
The abdomen with the flat navel, the chest modelled in large planes, the hollowed
gluteus — all are true Polykleitan traits. In the bronze the nipples would be inlaid ; in
imitation of this technique the copyist has surrounded them with a ring. The work
manship is poor on the whole, the hair is carelessly done, without detail. We may
hope that a better copy will some day come to light.
In order to appreciate the intention of the motive we must first look at the
other extant statues representing athletes pouring oil. Hitherto they have been all
incorrectly classed together as ' replicas ' or ' copies ' of the same original,1 although
some of them have merely the essential features in common — features which are due
more to the nature of the subject than to the invention of any one artist. Several
red-figured vase-paintings of severe style,2 earlier than any of the statues, show from
their realistic representations of scenes in the palaestra that it was a well-established
custom to hold the round aryballos high in the right hand and pour the anointing
oil from it into the left hand, which was placed horizontally across the body. With
regard to the attitude, accordingly, the statues, naturally enough, coincide. The
action necessitates the weight of the figure being thrown on the left leg, for the left
arm must be held as still as possible to catch the oil, and the head must be turned
towards the left. Now all these common features simply grow out of the subject
chosen. Given the subject, however, we may distinguish the following different
renderings. First the Munich statue,3 of which there are three other replicas.4 The
conception, though full of life, is expressed by a somewhat hard and angular
movement. The upper part of the body leans back too straightly and stiffly, the
definite forward bend of the head produces a harsh effect. The bodily forms are
simple, spare, and dry. Brunn noticed the analogy to the Marsyas of Myron,5
and this analogy undoubtedly exists not only in the conception of the swift,
momentary action but in the separate details of the attitude. As the head-type, too,
may be considered to be a development from Myron, we are justified in suggesting
that Lykios, the son and pupil of Myron, was the author of this work. We know that
he was working about 440 B.C.,6 and this is probably the date of the statue.
An entirely different creation, not at all, as has been asserted, a remodelled copy
of the preceding,7 has survived in the Dresden torso.8 It has several replicas, some of
them better and with the head preserved.9 Only those features characteristic of the
1 e.g. Wolters, Gipsabg. Nos. 462 — 464.
2 Cf. those cited by Bloch, Rdm. Mitth. 1892, p. 88, and especially Arch. Ztg. 1879, Taf. 4; Hartwig,
Meisterschalen, p. 570.
3 Friederichs-Wolters, Gipsabg. 462; Brunn-Bruckmann, Denkm. Nos. 132, 134, 135.
4 A restored torso in the cortile of the Palazzo Mattei (Matz-Duhn, 1025 ; Clarac, 940 B, 2398 A). The
spareness of the forms is well reproduced, but not so well as in the Munich copy. A torso has lately come from
Rome to Dresden which, although in bad preservation, is certainly an exact copy of the Munich statue — Arch.
Anz. 1894, p. 26, 6. A third very much restored replica in the Pal. Pitti (Diitschke, 25).
5 Ann. d. Inst. 1879, 204.
6 Klein's suggestion (Oesterr. Mitth. 1891, 6 seq.) that the encriuomenos of Alkamenes (Plin. xxxiv. § 72)
was an enehriomenos (iyxp'ip-evos), and similar to the Munich Oil-pourer (it would have been better to say to the
Dresden), is not tenable. Cf. Overbeck, Gesch. d. Plastik, 4th ed. i. 386. I cannot agree with Sauer (Festschrift
fiir Overbeck, p. 28) in seeing a resemblance between the head of the Aphrodite of Alkamenes and that of the
Munich Oil-pourer.
7 Brunn's assumption (loc. cit. 217 seq.) rests on a false notion of the character of the surviving copies. (Cf.
above, on the Diadumenoi. )
8 Friederichs-Wolters, 463; Brunn-Bruckmann, Denkm. Nos. 133, 134.
9 The Museo Torlonia contains a good torso (Plate 122, No. 476), with bad restorations ; the right leg is
completely modern. There is a replica in the Palazzo Pitti (Diitschke, ii. 22), with unbroken head, although
the whole figure is poor, and has been worked over. Both legs are preserved. In Rome, in front of the Villa

260 POLYKLEITOS
subject are the same as in the Munich type ; everything else is different. The hard effect
has disappeared. The attitude is quieter and more natural, but lacks the fire and
energy that distinguish the other composition. The upper part of the body is
inclined slightly forward, the head is less bent than in the Munich statue ; the legs
are closer together and more in repose ; the weight of the body rests — as the subject
requires — on the left leg,1 the right leg is placed to the side and slightly advanced,
the heel being somewhat raised.2 The attitude thus becomes lighter and more
graceful. The bodily forms are powerful and of adult growth, a great contrast to
the spareness of the other work. The modulations of the whole muscular system
are indicated by a number of finer details, altogether absent in the other statue.
The head is covered with short curls, the flesh of the face is full and firm, the
impression of power being heightened by the indication of whiskers and by the very
prominent forehead, from which the hair is brushed upwards. Unfortunately there
is no good copy of the head ; its pure Attic character is however undoubted ; in spite
of the coarser workmanship, it is allied to the standing Diskobolos ascribed to
Alkamenes (p. 90). The full plump forms of the body also recall this Diskobolos,
but the differences are such as to suggest another artist. Certain details like the
formation of the chest and the navel, as well as the attitude, make me think that
this work is by the artist of the Florentine athlete which we are shortly to
discuss. This artist must be sought for among Attic masters of the end of the fifth
or beginning of the fourth century.
There are some few statues under life-size which are akin in motive and,
generally, in style to the work last described,3 without being copies of it. They
are to be referred to different unimportant originals derived from the one prototype.
Other repetitions of the motive on a small scale are to be found in terra-cotta
statuettes from Myrina,4 on Greek reliefs,5 and on gems.6 These do not aim at
reproducing any particular statue ; they are simply instances of the general influence
exerted by a widely known and famous plastic motive. As they are most nearly
related to the Dresden type, we may conclude that this was the most popular.
In contrast, then, to the Dresden and Munich statues, which are both derivatives
of Attic art, we have a third new and significant creation of the Polykleitan school in
the ' Oil-pourer ' of Petworth. Unlike the Attic statues, the Polykleitan conception
Borghese, on the right, is a replica with an unbroken antique head in very bad preservation. The chin and the
upper part of the head with the pointed cap are new ; there is a breakage through the face.
1 In the Dresden copy the back part of the left foot is ancient, but it is plain from the original that the left
heel ought to be not raised but on the ground. (The remains of the old plinth are planed away under the sole of
the foot.) 2 Thus in the Pal. Pitti copy, where the foot is old. In the Borghese and Dresden copies it is wrongly
restored flat on the ground.
3 Statue in Turin, Friederichs-Wolters, 464 ; the head broken but antique. It differs from the Munich
statue and from the other types ; it is a poor, insignificant work. Further, statue in Cassel (ii. 1) ; the head
belongs to a replica of the Polykleitan ' Narkissos ' (cf. infra) ; no pubes ; rounded forms ; left hand ancient.
Finally, two of the small athletes in the Braccio Nuovo (Helbig, Museums, 41—45 ; Rdm. Mitth. vii. 93). The
head of No. 99 is ancient, but belongs to a reduced replica of the Florentine athlete ; No. 103 has a modern plaster
head. These two statues might be reduced inexact copies of the Dresden type.
4 B. C. H 1886, PI. 12 ; Pottier and Reinach, Nicrop. de Myrina, PI. 41, 3 ; p. 450 seq. A second copy
(with reversed sides) in the Coll. Misthos. in the National Museum at Athens.
5 Sybe], Catalogue, 534 (Friederichs-Wolters, Gipsabg. 1798), grave-relief of the third to second century. The
raised right hand ls empty ; it was left to the spectator to supply an aryballos in imagination. The same motive
occurs on.a fragment of relief from the south slope of the Akropolis (only the upper part preserved; the raised
right hand empty ; the left lies fiat on the stomach).
" Stone of Cneius, Jahrb. d. Inst. 1888, Taf. 10, 12, p. 315. Also Tolken, vi. 107, and Cades, iv. F, 73.
and a stone in Dr. Dressel's possession. I no longer hold the opinion (Jahrb., loc. cit.) that the gems go back to
the Polykleitan Petworth type. They correspond most nearly to the Dresden type

APOXYOMENOS

261

begins with the walking motive, although this is as little appropriate to the chief
motive of the figure as it is to the representation of an athlete binding a fillet about
his head or putting on a wreath. For pouring oil a firm steady position is
indispensable ; to do it while walking along is not natural — though it is just what we
should expect from Polykleitos. The youth is not giving his mind to what he is
doing ; he bends his head gracefully and modestly, thus forming a harmonious flow
of line as in other statues of Polykleitan style ; but he looks out into the distance
instead of fixing his eyes on the hand into which the oil is dropping. The right arm
is only raised halfway, much less than in the two other statues. We have seen that
when Polykleitos does raise the arm high, as in the Amazon and in the ' boy crowning
himself,' he makes the bowed head fill the empty corner formed by the arm ; this
could not be done here, because the head had to look to the left, hence the arm must
not be raised so high. It is true that this change takes away from the animation
of the figure, but it was the only way to obtain a graceful flow of line. From the
right elbow the line rises nearly straight to the crown of the head. Again, the whole
arm is so placed that it falls as nearly as possible in one plane with the body, thus
avoiding entirely the projection towards the front which the arm must have formed
in the other statues. We noticed a similar intention in the Diadumenos of
Polykleitos. Now pouring oil is an action far too energetic and requiring far too close
attention to be appropriately rendered in the Polykleitan manner, and Polykleitos — if he
and not one of his pupils is the author — seems to me to be less happy than usual in
this figure. The material content of the motive is not satisfying, and the formal
beauty is not, as in other statues, carried far enough to help us to forget the deficiency.
Polykleitos is here entering a region familiar to Attic art, but completely strange
to him. We may presume that he knew the older of the two Attic creations, and
that he was perhaps attracted through it to the theme. The two statues we have
assigned conjecturally to Lykios and Polykleitos must, however, have come close
together in time ; the ' oil-pourer ' of Polykleitos is at any rate older
than his Diadumenos.
In speaking of the second Attic, i.e. the so-called Dresden type of
' oil-pourer,' we had occasion to cite the Florence athlete as being akin
in style. Before closing the present inquiry a few words must be said
about this figure, as it has of late been interpreted as an athlete
pouring oil from a lekythos held in his right hand into an aryballos fig. ro8.— app-
held in his left.1 Such an action is however unparalleled, unclassical, and, a gem.
above all, absurd for an athlete to be engaged in. The left forearm lay
close to the thigh, and the right crossed the abdomen diagonally in the direction of the
left hand ;2 now a gem reproduced in Fig. 108 3 explains this attitude, and corresponds
to it so closely that we may conclude it is an imitation of the original of the statue.
Moreover, the gem belongs to that class of convex stones which has already afforded us
so many exact copies of statues.4 The figure on the gem holds a strigil in both hands,
1 Rdm. Mitth. 1892 (vii.), p. 81 seq. (Bloch). The only possible evidence for calling the statue an 'oil-pourer'
at all is the existence of a statue in Paris (loc. cit. p. 87 ; cf. Arch. Anz. 1891, p. 140) under life-size, in which
the left hand, which is ancient, makes a gesture as if to catch oil dropping into it. But this small statue goes
back to an original similar to, yet quite distinct from, the Florence athlete, so that even if it did represent an ' oil-
pourer ' this would prove nothing with regard to the Florence athlete.
2 The right arm is restored with the shoulder, and perhaps had quite another movement.
3 Impression, Cades, iv. F, 66 ; unquestionably genuine. Beside the athlete stands the usual hydria with
a palm-branch and a wreath over it. 4 Cf. supra, p. 137, note 1 ; p. 163, note I.

262 POLYKLEITOS
and is in the act of removing the dust from his left thigh ; the right hand holds the handle
and the left the blade of the strigil, in order to scrape the skin with more energy.
The tension in the muscles of the upper arm is appropriate to the action. The youth
is quite absorbed in what he is doing. The leg released from the weight of the figure
is placed well to the side with raised heel. This attitude— a preparatory stage to
the Lysippian scheme — is, as I have already pointed out, very near akin to, the
type of the Dresden ' oil-pourer,' and so is the full and muscular formation of
the body with all its characteristic details ; we may therefore
assume that both works are by the same artist.
The Florence statue, then, is not an ' oil-pourer,' but an
apoxyomenos, a destringens se — a magnificent work, famous even
in antiquity, as the remains of several copies prove.1
Polykleitos too was, according to Pliny, the author of a
destringens se. I can trace no copy of this work in statuary,
but I think there is an echo of it in a type very frequent on
gems (Fig. 109). If this be so, the original must have represented
a youth leaning his weight on the left leg with the right drawn
Y ig. log. — Apoxyomenos J ° ° 00
on a gem (Hermitage). back in the ' walking ' motive. The right forearm is stretched
forward, the left holds the strigil, and the intention is to cleanse
the under side of the right arm. The whole scheme is conceived in the Polykleitan
manner ; the act of scraping is subordinate and rules the composition only to a
limited extent ; the pleasant effect of the ' walking ' attitude is evidently the principal
thing in the artist's eyes, and the arms are very little removed from the plane of the
body. The number and remarkable similarity of the gem designs 2 prove that they
are all derived from one definite and well-known original.

VI. The Basis of the Statue of Pythokles. — Statue of a Boy in Dresden: its
Adaptations and Derivatives.
In the former section we took the Kyniskos basis as our starting-point.
Another series of statues may be discussed in connexion with the Pythokles basis
(/. G. B. 91).
This basis, whose upper surface is reproduced in Fig. no,3 is remarkable, not
only in having in addition to the older first inscription a second inscription which is a
1 A copy of the head in Museo Torlonia (Taf. 22, 86), worked over and much restored ; neck new, bust
not belonging. A better copy in the Hermitage, No. 65 ; the forehead modelled more fully and carefully than
in the Florentine example. A reduced copy is No. 105 in Braccio Nuovo (cf. supra, p. 260, note 3 ; Rom,
Mitth. vii. 92). The head is genuine but wrongly put on (the neck new) ; slight workmanship, but a real
copy. Head, No. 99, Braccio Nuovo. Cf. supra, p. 260, note 3.
2 The stone reproduced from a cast in Fig. 109 is specially good. It is an emerald-plasma in St. Petersburg.
A modern paste from it is in the Stosch collection, wrongly described as ancient by Tdlken, vi. 100. Several
smaller ancient replicas in the Petersburg collection (Berlin impressions, 22, 37, emerald-plasma ; 26, 34, 65 ; 29,
21 5 34i 65 ; 44, 10, all of the so-called achatonyx or nicolo). Other replicas in Berlin, Tdlken, vi. 99, 101,
102 ; in 103 the figure is seen more from the back ; also p. 2404 (convex emerald-plasma). From other
collections cf. Impronte dell' Inst. 7, 73 ; Cades, cl. iv. F, 65 ; in Paris (impressions in Berlin, 640), with the
modern inscription rcafov ; another stone, also ancient, with the same modern inscription, Bracci, Mem. d.
Incis. i. 52 (cf. Jahrb. d. Inst. iii. 317) ; stone of the Dressel collection ; paste of the Bergau collection ; and others.
The motives of all these gem designs correspond. The figure also occurs in a few rare instances turned to the
left, when the sides are also reversed (thus Tdlken, vi. 106 ; Cades, cl. iv. F, 67 ; Impr. dell' Inst. 6, 72 ; Paste
Bergau). 3 Drawn by Liibke from my plan of the block.

PYTHOKLES

263

restoration of the earlier one, and may be dated in the first century B.C., but in having
supported two different statues in succession. The feet of the earlier statue were
riveted into circular holes. Behind its right foot, on the upper surface of the basis,
was inscribed the name of the artist, on the front surface the name of the victor.
This statue must have been carefully detached and taken away. It was then
replaced by another with a quite different standing motive, and the feet of which
were fastened into hollows in the stone filled with molten lead. This second statue
appears to have looked in an opposite direction
to the first. The renewal of the inscriptions pro
bably took place at the same time that the new
figure was set up, for the new inscriptions record
ing the names of the artist and of the victor
start in a line with its right foot. Subsequently
this later statue was roughly broken away, prob
ably at the destruction of Olympia. From these
facts we gather that the genuine statue by Poly
kleitos was taken away from Olympia as early as
the first century B.C., and that it was replaced —
not by a copy — but by an entirely different figure,
which was then inscribed as a work by the famous
artist. Now there has been lately found in Rome,
and assigned to the period of Hadrian or the
Antonines, a basis actually inscribed with the names
of Pythokles the Olympian victor and of Poly
kleitos the artist.1 It is true that the footmarks on this basis fit neither of the
statues that once stood on the ' Pythokles ' basis at Olympia, but Petersen has
shown that these footmarks are due to some earlier use of the basis, and have nothing
to do with Pythokles. This Pythokles statue in Rome may possibly have been identical
with the original statue by Polykleitos which disappeared from Olympia in the first
century B.C., and which was doubtless brought to Rome. It is, however, more likely
that the existing basis with its statue was only a copy of the famous original by
Polykleitos, and that this original was placed to more advantage in some other part of
Rome ; for the basis with its statue was set up by a club of athletes, whose ambition
was to ornament their curia with figures of victors in gymnastic contests, but who
probably could not afford to buy originals.
The genuine Pythokles of Polykleitos stood firmly on the right foot. The
length of the foot, calculated from the circular holes, is about 24 cm. ; the left
foot was set back and — herein is the distinctive trait — well turned out ; the left
knee accordingly must also have been turned quite obliquely outwards.2 This posi
tion is natural only when the whole attention of the figure is directed to the left
side, i.e. when the head is turned to the left, and the left hand is occupied with
something. Among the works represented by copies which could come into question there
is one, surviving in two replicas in Rome and Munich, which best reproduces all the

Fig. no. — Basis of the statue of Pythokles in
Olympia.

1 Bull, della Commiss. Comun. di Roma, 1891, p. 280 seq., Tav. a. I ; Petersen in Rdm. Mitth. 1891, p. 304 seq.
2 On renewed examination of the basis in the summer of 1894, it seemed to me that, since the left foot also
had a hole for the ball, it was more probable the statue stood firm on both soles ; in this case however the left
must have been the supporting leg.

264 POLYKLEITOS
conditions of the Pythokles basis (Fig. in).1 It is a youthful athlete, who stands in
the position required by the basis, and whose feet correspond in scale to the foot
marks,2 The head is turned to the left, the left forearm is stretched forward, while

Fig. ut. — Athlete in the Braccio Nuovo (Vatican).
the hand holds a globular aryballos. The right arm hangs simply down, and blends
with the contour of the body ; the hand perhaps held a fillet.3
As regards the time of life represented, this youth is older than the ' boy crowning
1 A. In Rome, Vatican, Braccio Nuovo, No. 101 (Fig. in) ; cf. Helbig, Museums, on Nos. 41—45 — 'reveals
a close kinship with the Doryphoros of Polykleitos. ' Photograph in German Inst, at Rome. Legs and basis
ancient ; arms and neck restored ; the head is ancient and belongs . to the statue, only it is placed leaning a little
too far back.— B. In Munich Glypt. No. 303 ; Clarac, 858, 2175 ; black marble ; left leg restored, and the foot
is wrongly placed flat on the ground to the side, instead of being drawn back. The head, both arms, and the left
hand with the oil-flask are ancient (A is restored from B).
2 On A I measured 24 cm.
3 The right hand which survives in B held some light object, probably a fillet ; in A the hand is wrongly
restored ; it simply hung straight down.

STATUE OF A BOY 265
himself,' and younger than the Doryphoros ; he is very near in age to the ' oil-pourer.' l
The comparison gives us some notion of how nicely the gradations from boyhood
to adolescence were shaded off in the Polykleitan school.
This work stands very near the Doryphoros, not only in the stage of growth
represented but in the style and mode of expression. The hair is treated in the
same flat, spare, and restrained manner, divided into large strands which cling close
to the head ; over the middle of the forehead is the same symmetric parting.2 The
hair leaves the ears free. The forehead too, with its horizontal division, and the whole
expression of the head, which is not bent but freely turned to the side, show a special
likeness to the Doryphoros. The same may be said of the body, except that the
forms are less powerful. We must assign the work to a period not far removed
from the time of the Doryphoros.
The motive of this statue, full of energy and character, reaches its most life
like development in the Diomede of Kresilas. The action of the Diomede is fully
accounted for down to the smallest particulars by the nature of the subject. The
like cannot be maintained of the athlete holding the aryballos. Hence we may
suppose that Polykleitos, in this instance also, was under the influence of a creation of
Kresilas. He saw an attractive motive (the Diomede, we have shown, was made for
Argos), and adapted it to the figure of an athlete without troubling himself to account
for the attitude on internal grounds. I have already suggested that he may have used
the Kresilaian Amazon in like manner (p. 254). Nor were the motives of Diadumenos
and 'oil-pourer' his own. It was certainly not in invention that lay the special strength
of our artist.
A very close analogy to this Roman copy of a Polykleitan statue is afforded by
an admirable small original work from Greece, a bronze statuette now in Athens.3 It
cannot be referred to the master himself, for it differs in too many respects from the
works we know to be his, and is besides too insignificant in character, but it is an
excellent specimen of school work from his earlier period. The hair is more uniform
and restrained than in the statue we last discussed ; the head is peculiarly heavy ; the
time of life represented is that of the Doryphoros, but the forms are drier. The motive
is similar to the preceding, but shows more repose and simplicity. The head is not
turned, but only bent to the side, and the left knee is not directed outwards. The left
arm seems to have been raised, and was perhaps supported ; the right arm hung down.
A small marble figure of very poor late workmanship from Athens, in the British
Museum,4 markedly resembles this bronze.
On the other hand, there is another admirable work which we may trace to
Polykleitos himself, and which must have been one of his most beautiful creations.
There is a well-preserved copy of it in Dresden (Plate XII. and Fig. 112).5 A replica,
1 The pubes is not indicated. This is plain in A in spite of the plaster covering ; in B the membrum is
preserved intact.
2 The head of A is much better and more carefully worked than that of B, but they are copies from the
same original. A well reproduces the sharpness of the bronze.
3 Mon. d. Inst. viii. 53; Annali, 1868, 316 sqq. (Kekule) ; Brunn-Bruckmann, Denkm. No. 280 a; presumably
from Sikyon. The arms, now missing, were put on separately.— Lange (Fremstilling af Menneskesk. 419) is, I
think, wrong in supposing the figure to be earlier than Polykleitos.
4 A. H. Smith, Catal. of Sculpt, i. 502, 'Doryphoros.'
5 Becker, Augusteum, Taf. 88 ; [Clarac, 948, 2437; Hettner, Catalogue, 4th ed. No. 90, 'Doryphoros.'
Mentioned by Michaelis, Ann. 1878, 8, among the ' variazioni ' of the Doryphoros; by -Kekule, Ann. 1868,
316, as a replica of the Athens bronze discussed above ; by Treu in Arch. Anz. 1889, p. 57. — The restored left
arm is now taken away. Everything else is ancient except the left upper lip, the nose, and the membrum. The
M M

:0C>

POLYKLEITOS

formerly belonging to Cavaceppi, is probably somewhere in England.1 Besides these
there are two torsos in Rome,2 and three replicas of the head alone (in Berlin, the
Vatican, and the Coll. Barracco).3
Once more it represents a youthful victor with the feet placed as on the Pythokles
basis, except that the left foot is not drawn so far back, and the toe is not so much
turned out. In accordance with this modification the head is but gently turned to the
left and inclined. The right arm again hung straight down beside the body ; 4 the left
upper arm is directed somewhat to the side, the left shoulder is a little raised and
advanced, so that the front of the upper trunk is not all in the same plane ; this
produces an animated crossing rhythm. The right shoulder is not drawn back, but
held in a comfortable natural position ; from the attitude of the left upper arm and
from the absence of any puntello on the left side we conclude that the left forearm
was advanced. The youth was holding some object on which he was looking down.5
This was doubtless nothing but a victor's fillet or some simple instrument of the
palaestra.6 The modest reserve expressed by the attitude of the head seems to speak
for the fillet. For just this modesty and reserve on the part of the victorious athlete
were demanded by ancient custom, that the envy of gods and daemons might be
appeased. The time of life represented is a new variant between boy and ephebe. The
' Dresden athlete ' — so we may call the type— is older than the ' boy crowning him
self The formation of the body is stronger and the chest is fuller, bearing the same
proportion to the lower part of the body as it does in the Doryphoros. On the other
hand, the Dresden athlete is more youthful and delicately formed than the figure, pre
sumably the Pythokles, which we discussed above, or than the ' oil-pourer.' It is a
great mistake to imagine that the victor statues of Polykleitos were all made upon one
pattern, and merely repeated the proportions and modelling of the Doryphoros ; the
latter evidently gave the ' canon ' for the full-grown male figure only.
The ' Dresden boy ' bears such unmistakable marks of the personal style of
Polykleitos that we may certainly ascribe it to the master himself, and it probably
belongs to his later period.7 It stands midway between the Herakles (discussed
above, p. 234) and the Diadumenos, and is closely akin to both works. The hair
resembles that of the Herakles, but is more developed in style ; the asymmetria of the
head was never broken. Height of head 198 mm., length of face 156, length of foot 253, distance between
nipples 226.
1 Cavaceppi, Raccolta, iii. 5 ; Clarac, 866, 2204. According to Cavaceppi, in the possession of ' Giuseppe '
Nollekens in England ; the arms and the armour as support are probably restorations ; in other respects the figure
appears to be a replica.
2 u. in Museo Torlonia, No. 18. Legs and arms wrongly restored, head antique but not belonging ; the
torso is a good replica of the Dresden statue. — b. in the Vatican, Museo Lapidario, 124. Torso.
3 a. in Berlin, Skulpt. 546. A very good copy. The hair corresponds exactly with the Dresden replica.
The whole upper part of the head is restored. The sharpness of the bronze original is imitated ; the lips are
slightly open, but do not show the teeth. — b. Vatican, Museo Chiaramonti, 507 (Helbig, Fiihrer, i. No. 100). A
good careful copy ; nose and neck restored ; edge of lips sharp. Mentioned by Flasch, Verh. d. Philologenvers.
1874, p. 163 ; by me in Annali, 1877, 203. — c. Coll. Barracco (PI. 46). A good copy.
4 Remains of puntello on the right thigh of the Dresden statue.
5 Hettner assumed a spear, because formerly all Polykleitan figures were supposed to be variants of the
' Doryphoros.' 6 On a Panathenaic prize amphora of the fourth century an athlete occurs very similar to the Dresden type,
He carries a discus in his left hand (Mon. d. Inst. x. 48 g, 10).
7 Kalkmann (Proportional d. Ges. p. 55), in endeavouring to extract results from his rows of figures, places
these and other statues to be subsequently discussed (such as the Pan, the 'Narkissos,' the 'Idolino') in the
pre-Polykleitan period ; this is only one of the many anachronisms into which Kalkmann has been induced
by his figures.

XII.
Statue of a Boy.

fiwled byA.Frach,BerlinW(Germany)

268 POLYKLEITOS
short wisps over the forehead, which occurs first in the Herakles, is here more pro
nounced. In the Herakles the hair already covered the upper edge of the ear ; here
the motive is accentuated, and the ends of the small curls veil with their rich variety
the whole upper part of the ear. The fashion, here so charmingly elaborated, of
separating the hair into numerous narrow strands with intertwining ends also made
its first appearance in the Herakles.1 There is no trace of it in the Doryphoros.
The ' Dresden boy ' is associated with the Diadumenos by the fuller plastic treat
ment of the hair. In both there is a depression on the crown of the head, while the
bunches of hair radiating from the crown — three in each case — are so similar as to
be almost identical ; a sure confirmation of our theory that Polykleitos is the
master and that the ' Dresden boy ' comes near the Diadumenos. It is however
certainly the earlier of the two, for the hair still lies quite close to the skull, and
the step which divides the Diadumenos from all earlier Polykleitan works has not
yet been taken.
Further, the expression of the head, and especially the formation of the lower
half of the face (Fig. 112), indicate a stage preparatory to the Diadumenos. The
modelling here is softer, rounder, and more developed than in the earlier works of
Polykleitos ; the full lips, the folds near the corner of the mouth and nostrils, the some
what receding chin, and the expression of sweetness, satiety, almost of melancholy,
mark a decided approach to the head of the Diadumenos. We can better understand
the place of the latter statue in the development of Polykleitos now that we have found
a connecting link leading up to it.
In studying the Diadumenos we recognized the influence of Attic style on
Polykleitos, and suggested Kresilas as its source (p. 243). In the ' Dresden boy ' this
influence is beginning to be felt, for it exactly explains the variations from the older
Polykleitan type. Not only the formation and expression of the lower part of the
face, but the way in which the ends of" hair tangle together, are clearly inspired by
works from the Myron-Kresilas circle, and more especially by the Diomede. But
Polykleitos has not altered the fundamental characteristics of his hair technique ; he
entirely ignores the short, full crisp curls of Attic work ; in the ' Dresden boy ' he has
arranged the hair on the upper part of the head practically in the same smooth
layers as in the Doryphoros ; only in the Diadumenos does he break away to some
degree from his old manner.
The body of the ' Dresden boy ' is rendered in the style customary to Polykleitos :
e.g., it has the characteristic flat navel with the deep groove below it and the hollow
at the side of each gluteus. Compared with an earlier work, such as the ' boy crowning
himself,' the transitions are softer and more rounded ; this is more than usually
noticeable in the divisions effected by the straight and oblique abdominal muscles
which are still so harsh in the former statue.
The ' Dresden boy' may be classed among the happiest efforts of Polykleitos. The
attitude has more of nature and truth than his compositions usually manifest, without
being the less beautiful or harmonious. The head has a charm of its own which makes
the earlier works of the artist seem cold and formal, and which is unsurpassed except
by the Diadumenos.
With this work two others are nearly connected. One of these survives, as far
as I know, only in one copy, now in St. Petersburg (Fig. 1 1 3) ; 2 it is a figure very like
1 Especially behind the left ear in the Dresden copy of Herakles.
2 Guedeonow, No. 304, ' Mercure ' ; also apparently cited as Hermes by Treu (Arch. Anz. 1889, 57). Legs

YOUTHFUL ATHLETE

269

the Dresden boy. The position of the feet and the turn of the head are the same, and
here too the right arm hangs straight down, while the left forearm is extended. Yet
the shoulders are well set back
in the usual way ; the body-
forms are somewhat more youth
ful and delicate. The head — to
which the restorer has arbi
trarily added wings — is akin to
the Dresden figure, but the hair
is fuller and more curling ; the
personal style of Polykleitos is
missing ; the hair is not parted
over the middle of the forehead,
nor is it arranged in layers. The
ears are not covered by the hair.
The figure undoubtedly repre
sents a boy victor ; in his lowered
right hand is a fragment of an
attribute. The restoration as a
purse is quite ridiculous ; the
fragment is a short straight piece
not round in section but rect
angular ; it might very well be
the handle of a strigil.
I am inclined to assign this
work not to the master but to
one of his pupils, who used the
Dresden boy as his prototype ;
not only the head but the body
shows a lack of the real personal
manner of Polykleitos, and of
all the finer shades of individu
alization which we have learned
to look for in his authenticated
works. The second work survives
in a great number of copies, and
was clearly a favourite and
widely known. Being only two
thirds life-size, it was suited for
the decoration of private houses.
One copy shows distinct marks
of having been used as a figure
represents a divine boy, Pan in

Fig. 113. — Statue of a youth (Hermitage).

for a fountain in a Roman house. The statue
almost human form, a subject we have already
met with in an original of the Polykleitan circle (p. 229). I know four complete

and feet, with the moulded basis, are mostly ancient ; left forearm is restored ; right hand with the remains of
an attribute (wrongly restored as a purse) is ancient. The neck is mostly restored, the head genuine and set on
almost in the right position ; the wings are a later addition placed on the hair, which is complete without them
and carefully worked. The workmanship is not older than the time of Hadrian.

270 POLYKLEITOS
copies, and six replicas of the head.1 The statue in best preservation is the one in
Leyden (Fig. 114).
The feet are placed exactly as in the two works just discussed, and the head is
inclined slightly to the left ; the left shoulder is rather more advanced than in the
Dresden boy, and this produces a slight turn in the upper part of the body. The right
hand again hangs simply down holding the short knotted stick (kaywf36\ov) which
was commonly used in hunting the hare,2 and which constantly appears as an attribute
of Pan, the zealous huntsman. The left forearm, as we assumed in the case of the
Dresden boy, is stretched forward, and holds the other chief attribute of Pan, the
syrinx,3 on which his lowered gaze seems to rest. The motive so suitable to the
modest victor is here simply transferred to Pan, apparently without a thought whether
it is appropriate to him or not.
The head shows the closest kinship to the Dresden boy,4 as well in the form of
the skull with the hollow in the nape of the neck, as in the face and its expression.
The hair is rendered in the same narrow intertwining strands ; separate parts, such as
the lock in front of the right ear, are strikingly similar. Yet the whole mass is
fuller and looser than in the Dresden boy, and more like the Diadumenos. The body5
also corresponds stylistically with the former, i.e. the transitions are softer than in the
' boy crowning himself,' while the forms throughout are Polykleitan.
Nevertheless, the artist has tried to characterize Pan ; the hair in front does not,
as in its prototype, the boy victor of St. Petersburg, fall over the forehead, but grows
upwards as in figures of Satyrs ; in order that it may not break the beautiful
contour of the skull, it must of course lie close to the head ; the two horns lie equally
close, their roots being skilfully hidden in the hair ; the long pointed ears, almost
like those of a beast, are set flat against the sides of the head.
We have here most probably the production of a pupil who worked in immediate
association with Polykleitos, and founded the conception of his own statue on that
' Dresden type' created by the master himself for the representation of a boy victor. His
work was certainly not far removed in time from this type and from the Diadumenos.
We can point to two more instances in which the beautiful Polykleitan creation
of the Dresden athlete has been used for a mythological figure.
One of these works exists, so far as I know at present, in one copy only. It is a
statue of a youth, resembling the Dresden boy in all particulars, even in the principal
1 Whole statues with head, (a) in Leyden, Janssen, Catalogue, i. 62. Excellently preserved; the legs, the
basis, the right arm with the pedum, and the left arm are genuine ; the syrinx in the left hand is at least partly'
ancient. On the head only the end of the nose is new. In 1883 I examined the statues (b) and (c) — the two copies
worked by Cossutius Cerdo, Brit. Mus. (Guide Graeco-Rom. Sculpt. Nos. 188, 190 ; Anc. Marbles, ii. 33, 43 ;
Brunn-Bruckmann, Denkm. No. 47). The arms are wrongly restored ; in one the nose is complete and the hair
carefully copied ; in (a) the body is better. — (d) Vatican, Gall, dei Candelabri, 246 ; Helbig, Museums, i. No. 389.
The poorest of all. The left arm is new ; the right arm is ancient and hangs down holding an oinchoe, which
is supported on a pillar and used as the spout of a fountain : hence the Pan must have been used as the decoration
of a fountain. The head is very sketchy, the hair almost without detail. — Separate heads : (e) Good copy in the
Conservatori Pal. (Bull. Comunale, 1887, Tav. iv. ; Helbig, Museums, i. No. 606). Here the polykleitan style is
very exactly reproduced ; (/) a good but not well-preserved copy in the Terme Mus. (photographs in German
Inst.) ; (g) less good, in Vienna, Hof. Museum ; the features are softened and inexact ; (h) in Palermo, in the
second court of the Museum ; nose and chin new, hair fairly faithful ; (i) Hanover, Kestner Museum, poor
copy ; (k) Lateran, 524; Helbig, Museums, i. No. 666, poor. Lateran, 288 (Helbig, Fiihrer, No. 630), is a free
modification of the type, not a copy. — Cf. Annali, 1877, 202 seq. ; Satyr von Pergamon, p. 29.
2 For the straight form of the pedum cf. Ann. d. Inst. 1877, 212 seq.
3 To be restored with pipes of equal length Cf. Annali, 1877, 214 seq.
4 The head of the Leyden statue here published is not exact in the details of the head, b, c, e, and / are
more true to the original.
5 The body in the Leyden statue is more faithful to the style of the original than the copies by Cossutius
Cerdo, which are too soft and plump.

PAN 271
proportions, except that a sword-belt crosses the body from the right shoulder to the
left hip.1 Probably he held the sword or some other weapon in his outstretched left

Fig. 114.— Youthful Pan (Leyden).
hand, and we may call him Ares. The copyist was one of that numerous class who
reproduce Polykleitan originals in a soft and feeble manner. Such copies are apt to
1 Rome, Conserv. Pal. Bull. Comunale, 1886, Tav. i. ii. ; p. 54 seq. (Benndorf, who, mistaking the details of
the statue, the technique, and the style, sees in it the Thespian Eros of Praxiteles) ; Helbig, Museums, No. 568 ;
Robert, Arch. Mdrchen, 160 seq. The upper part of the head and the back were pieced on. The figure can never
have had wings.

272 POLYKLEITOS
deceive an unpractised eye. Yet sufficient Polykleitan traits answering to the
Dresden type are left to make it apparent that the original was closely allied to
that type. The hair round the ears — covering in this figure also the upper part of
the ear — is treated in the same way ; the modelling of the body, the characteristic
navel, the hollow at the side of each gluteus, even on the side of the leg at rest, and
other details, are all purely Polykleitan, though they are rendered in a weak manner
that lacks all energy. The original must have owed its existence to the immediate
circle of Polykleitos.
There are several replicas of a statue of Dionysos,1 which, like the Pan, is two
thirds life-size. The motive is again essentially the same as that of the Dresden boy
victor ; here too the right hand hung down, and the left was stretched forward.
The forms of the body most nearly resemble the Pan, and are quite in Polykleitan
style, without any mixture of the softness appropriate to Dionysos.2 Unfortunately
there is only one authentic fragment of the head on one of the replicas.3 The
god is characterized by long curls falling on the shoulders, by a bunch of hair on
the nape of the neck, and by a gracefully disposed nebris fastened on his right
shoulder. For the source of the work we must certainly look to the school of
Polykleitos. Another statuary composition, also two thirds life-size, belongs to this same
series (Fig. 115).4 Here the typical motive has become more pronounced, and
expresses quiet and pleasant repose. It is true that the essential features — the
attitude of the legs, the turn and inclination of the head, the boyish forms — are
unaltered ; but in this instance the left hand rests on a pillar, and the weight of the
body is partly transferred from the right leg to the left arm. The right hand, in
1 (a) in Dresden, Hettner (4th ed.), No. 163 ; Becker, Augusteum, ii. 74; Fried. -Wolters, Gipsabg. 1493. The
lower part of the r. leg and the 1. leg from the middle of the thigh are new. Remains of a puntello in the right
thigh show that the right arm was lowered. The neck and parts of the head — viz. right ear, piece of cheek, and
hair above the ear — are ancient and unbroken. At the back, above the battered remains of a tail of hair of the
simple, severe style, the restorer has placed a knot of hair of the usual late fashion. Everything else about the
head is modern, (b) Berlin, Sculpt. 89. Torso, badly restored, (c) Clarac, 684, 1603 A, Ince Blundell Hall ;
Michaelis, Sculpt, in Gr. Brit. p. 347, No. 32.
2 Specially characteristic are the abdomen with the flat navel, the linea alba, and the sharply marked off
pudendum. 3 In a. In c the whole head might from the style possibly be ancient. I have not seen the statue.
4 The so-called Narkissos. Winnefeld (Hypnos, p. 30) has collected a list of replicas, among which some few
variants that really do not belong to the series have crept in. I described several replicas in Bull, dell' Inst. 1877,
158. I know the following in the original : Statueswith head: (a) Berlin, Beschr. d. Skulpt. 223 : stylistically faithful
copy, not very detailed in the hair ; upper part given in Winnefeld, Taf. ii. = our Fig. 115. _(b) Mus. Chiaramonti,
526 : only the upper part preserved, very good, (c ) Pal. Rospigliosi, Friederichs-Wolters, Gipsabg. 525 : poor.
(d) Villa Borghese : upper part only. Cf. Bull. , loc. cit. 159, 6. (e) Mantua, 65, Diitschke, iv. 650 : holds an apple ;
very poor, inexact, and rough copy. (/) Louvre, Coll. Campana, phot. ed. Giraudon, 1295. (g) Carlsruhe, Winne
feld, Taf. i. : style inexact, much softened down ; small wings added to the head ; workmanship of Hadrian's period.
Kalkmann (Gesichtsprop. p. 53) is wrong in thinking the head does not belong.— Torsos: (/^Mantua, 26: right hand
empty, (i) Genoa, Pal. Reale : an apple in the right hand, which is half antique, (k) Rome, Palazzo Colonna : head
does not belong. (/) Pal. Doria, Matz-Duhn, 223. (m) Pal. Barberini, Matz-Duhn, 975 : apple in the right hand.
(«) Berlin, Skulpt. 224. (0) Naples, casually mentioned by Friederichs, Arch. Anzeiger, 1862, 309, and by others.
Height without basis o'6i, basis antique, lower part of legs restored to the ankles, head missing. Both hands
preserved : the right on the back holds an apple, the left is supported by a pillar on which lies drapery. In front
of the pillar a female term, draped above ; the feet come out below ; free style, long shoulder curls, full face.
—Heads : (p) Berlin, Skulpt. 263 ; much defaced, but gives the hair faithfully, (q) Cassel, ii. 1 : a very good copy
placed on the torso of an oil-pourer. (r) Copenhagen, Coll. Jacobsen, No. 1094: poor, hair very sketchy;
wrongly called ' Doryphoros ' in the catalogue, (s) Ibid. No. 1075 : free copy, but unaltered in proportions ; hair,
eyes, and mouth formed in later style, as in c and g. To see how copies can differ, cf. r and s. [To these eighteen
replicas may now be added (t) statue recently acquired by the Louvre ; head and torso in exquisite preservation ;
arms and legs broken but belonging ; admirably published PI. xvii. of Monuments et Mimoires (Fond. E. Piot),
vol i. ; ibid. p. 115 icj. (E. Michon), where also further particulars on the Campana statue (/) are given.— E. S.]

ADONIS 273
order not to disturb the repose of the body, must have been laid on the back- The
head was bent down close to the raised shoulder, and the whole composition was
rounded off in a highly harmonious and pleasing manner. The bodily forms corre-

Fig. 115. — The 'Narkissos' (Berlin).
spond to those of the Pan, and so does the stylistic treatment of the head and the
magnificent Polykleitan build of skull. The hair, although full and plastic, clings
for the most part closely to the head ; the upper part of the ear is covered ; there
is no parting over the forehead. None of the copies show fine separate intertwining
N N

274 POLYKLEITOS
locks of hair as in the Pan ; this however may be accidental ; certainly the latter work
affords the nearest stylistic parallel as regards the hair, and there is great resemblance
in the facial forms. Both works must be referred to the same circle of the master's
pupils. In the Amazon, Polykleitos had already created a figure at rest, leaning slightly
against a support ; but it was a great step from this to a figure completely supported
on one side like the statue we are now considering. The motive strongly recalls
the Herakles, leaning with weary bowed head on the club which props him under
the left armpit, while his right hand is laid on his back or set against his side. This
Herakles type was developed as early as the fifth century,1 and seems to have taken
shape not only in drawings and reliefs, but in a work of statuary which must have
come into being in the Attic artistic circle of about the end of the same century.2
It is quite legitimate to ask whether one of these two creations, the boy or the
Herakles, was not influenced by the other. Since the motive as applied to Herakles
has an eventful previous history extending back to the severe Attic style of
vase-painting, and is closely bound up with the very nature of the hero of so many
toils and troubles, and since the circle of Polykleitos is very deficient in inventive
power, it seems reasonable to suppose that this is another instance of the transference
of a magnificent statuary conception from Attic to Argive art. In the present case
the motive was adapted to the boyish type of figure resembling Pan.
We still have to consider the meaning of the figure, which copyists of Imperial
times adapted to various interpretations. One copyist made it into a Hermes (hardly
a Hypnos) by adding small wings to the head.3 In some replicas there is an apple
in the right hand ; 4 the intention of this addition is doubtful, but it may be the
usual Aphrodisian attribute.5 A figure in relief with an apple on the pillar support
belonging to one copy {o) seems to corroborate this notion. It is a veiled goddess
ending in a term, and doubtless to be interpreted as Aphrodite, who as Ourania
was represented in Athens in the form of a term.6 Our figure, then, must have
some special relation to Aphrodite. Another replica 7 has the arm supported on a
wine-skin, hence it was probably intended to decorate a fountain. The gems too,
which show frequent reproductions of our statue, are of great importance. On one
fine piece a female figure almost nude, probably Aphrodite, is placed on the right
of the boy.8 Another time he is characterized as a huntsman, and holds two spears
in the left hand, which is propped on a rock ; dogs accompany him, and on the rock
are a boar's head and stag's antlers supporting an image of Artemis.9 Both gems
render the statue in a very exact and characteristic manner, and are certainly
1 Cf. Roscher's Lex. i. 2173.
2 Life-size copy in Dresden. Style, end of fifth century. The right hand is set against the side, not laid on
the back. According to Herrmann (Arch. Anz. 1894, p. 26), there is a replica in the Villa Borghese.
3 g in the above list. Winnefeld supposes Hypnos ; but the wings are like those given to Hermes, and
Hermes often occurs in similar motives.
4 e, 1, m, 0. An apple, not a pomegranate. Kalkmann (Gesichtsproport. p. 53) is wrong in supposing it to
be an aryballos.
6 The common supposition that the apple shows the figure to be a death daemon is untenable and quite
unproven (Satyr von Pergamon, p. 30, note).
6 Paus. 1. 19, 2. The manner of draping is specially common to terms, which render the body down to the
hips Figures like Herakles and Pan, which are usually nude, are likewise draped when adapted to terms.
^ I only know it through Laborde, Voyage en Espagne, PI. 99 E. Cf. Bull. d. Inst. 1877, 159.
8 Cades, cl. i. K, 86 : probably a paste ; large, fine style ; drapery added on the left shoulder. Evidently an
imitation of the statue.
" Cades, cl. iii. B, 105 : drapery on the left shoulder, style of the statue well reproduced. Admirable
workmanship.

ADONIS 275
derived from it. Other gems give variants ; the figure is turned to the left and the
sides reversed. A stone of this kind makes the figure into a Ganymede by the
addition of a Phrygian cap and of an eagle below.1 Another makes it into
Narkissos by an indication of water in the background and of a figure leaping into
it.2 A freer variation is one in which not the left hand but the forearm is
supported on a pillar or pedestal : this type is once characterized as a huntsman and
also as Hermes or Bonus Eventus ; 3 with reversed sides and looking to the left
he again appears as a huntsman.4 An Attic grave-relief of the fourth century repeats
the type, adding a small hare which rests on the boy's left hand.5
Nowhere do we find any indication that the type had a sepulchral signification.6
On the other hand, it is important to note that the best gems represent the youth as a
huntsman ; those gems which make him into a Hermes, a Ganymede, or a Narkissos
are few in number, of inferior workmanship, and evidently free modifications of the
original design, like that statuary copy in which wings are added. Further, a near
connexion with Aphrodite is assured, not only by one of the best gems, but also by
one of the statues, and by the apple which occurs in several of the replicas.
The figure must have a mythological meaning ; it is not the statue of a victorious
athlete, but belongs to the same series as the Pan and the Dionysos of the Polykleitan
school. Narkissos is not a familiar figure in this school, nor is there anything definite
to suggest Hyakinthos, an interpretation I was formerly inclined to.7 But if we
call the boy Adonis, I think we shall find that everything fits admirably to the name.
The cultus of Adonis is known to have been practised in Argos,8 and it cannot have
been more recent there than in Athens, where we find it established in the time of
Alkibiades.9 If there was an Adonis of the Polykleitan school at all, it would certainly
be a figure without a trace of Oriental costume, and we could hardly think of it as a
different type from our statue.
On an Attic aryballos from Kertsch, with reliefs of wonderful beauty carried out
in a style akin to the Parthenon frieze, Adonis (inscribed) appears as a slender youth
with no hint of the Oriental. He is closely grouped with Aphrodite and Eros, and is
represented resting, with his arm over his head.10
1 Formerly Coll. Schaaffhausen, Catalogue, 354 (impressions, 14) : chlamys added ; a lance leans on the rock.
Poor workmanship.
2 Carnelian ; Coll. Thorwaldsen, Cades, iii. B, 93 ; Impronte dell' Inst. i. 73 ; Wieseler, Narkissos, Taf. No.
8 ; cf. p. 20. Late bad workmanship.
3 The gem of Koinos, pub. Jahrb. d. Inst. 1888, Taf. 10, 20 ; discussed ibid. 1889, p. 51. Huntsman with
spear and dog. Cades, cl. i. L, 26. Kerykeion in the supported left hand, therefore = Hermes. Poor workman
ship. St. Petersburg (impression in Berlin, I, 34) : nicolo, from the ear of corn in the left hand, evidently
a Bonus Eventus ; very poor.
4 St. Petersburg (impression in Berlin, 19, 31), carnelian; the r. hand holds a hare out towards an eagle ;
a dog beside it ; two spears in 1. hand on the back.
"5 Stele of Telesias. Friederichs-Wolters, Gipsabg. 1014 ; the left forearm is supported on the pillar ; like the
gems in note 1.
6 As I formerly maintained in Bull, loc. cit., this opinion was retracted in the Satyr von Pergamon, p. 29 :
it is still adhered to by Winnefeld, loc. cit. 7 Satyr von Pergamon, p. 29.
8 Paus. ii. 20, 6. The wailing of women was, here as at Athens, the principal feature of the cultus.
9 Aristoph. Lysistr. 389^.; Plut. Alkib. 18 ; Nik. 13.
10 Hermitage, newly acquired : the reliefs are of great beauty and interest. The scene is laid in Cyprus.
To the left the hero Tevxpos (inscr.), the founder of the Kyprian Salamis ; opposite him sits Tekmessa, who was
given over to him by his dead brother (Soph. Aias, 563, 972 sqq.), and the boy Etpvo-mv* (inscr.) ; then Adonis
fAoawios) seated, with Eros standing on his lap ; he holds with his left hand neiflw standing, who lays one hand
on his lap ; in her thin chiton she resembles the Venus Genetrix (of Alkamenes). Behind her, opposite to
Adonis, sits 'AippoSirr,, almost exactly like the Hegeso of the well-known grave-relief. The inscriptions are in
Ionian characters; the style dates the vase in the time of the Parthenon frieze. Delicate painting and
gilding.

276

POLYKLEITOS

It was no out-of-the-way idea to represent the beautiful huntsman who died
young in the attitude of our statue, i.e. in a posture expressing weariness. If the
work enjoyed such a measure of favour in later times, it was probably owing to the
charm of the motive and the ease with which the figure, like the Pan, could be adapted
for decoration. The small scale (two thirds life-size) of the works we have just discussed is

Fig. 116. — Bronze statuette. (Bibl. Nationale, Paris.)
one of their characteristics ; they were commissions of minor importance which
fell to the share of the great master's pupils towards the end of his career or after
his death. A more recent school work, allied in motive to the ' Narkissos ' discussed above,
and to be classified in this series on account of the attitude of the legs and the droop
of the head, can be studied in the admirable little bronze in Paris1 (Fig. 11 6, a and b)

1 Cabinet des Medailles, Coll. Janze, No. 50. Height c. C25. Beautiful pale-green patina. The pupils were
inserted. The left hand, now missing, was made separately. Both feet are restored in wax. I discussed
statuette and gem in the Archdol. Gesellschaft, Arch. Anz. 1891. p. 141.

ARISTAIOS 277
and a gem (Fig. 117) exactly corresponding to the statuette and evidently derived
from the same original.1 According to the evidence of the gem, the figure leaned
with the now missing left hand on a staff round which a snake was twined.
Not only this attribute but the whole type is borrowed from Asklepios, for whom
it had been already elaborated in Attic art at a period not much subsequent to the
middle of the fifth century, as an interesting statue in Berlin shows.2 Later, in the
circle of Skopas and Praxiteles, the type was carefully fostered and
intensified,3 an effective upward turn being now given to the head.
Our figure goes back to the old formula of the type, in which the head
is only turned slightly to the left, and the right hip only curved slightly
outwards. A variation is introduced in the snake-wound staff, which
is not propped under the left armpit, but supported between the forearm
and the body and held by the hand. This brings the position nearer fig. n7.— Gem
„,..., (Cades, ii. D,
the Polykleitan scheme. 16).
The modelling of chest and abdomen is clearly Polykleitan, while
the head also offers conspicuously Polykleitan characteristics — in its large surface
planes, in the formation of nose, eyes, and forehead, and in the symmetrical parting
of the hair in front. The noble simplicity of the drapery corresponds to the style of
the rest. We may conclude, then, that the original is a work of the school of Polykleitos
belonging — if we judge from the strongly plastic freedom of the hair — to the beginning
of the fourth century, and based on an Attic Asklepios type of the fifth century.
But whom does the statue represent ? An unbearded Asklepios would be nothing
out of the way, especially in the Peloponnesos.4 There is, for instance, a statuette
from Epidauros representing an unbearded Asklepios whose forms indicate Polykleitan
influence.5 But the mural crown on our figure points to another interpretation ;
the person pourtrayed must be a protecting divinity, a hero allied in functions to
Asklepios, and reverenced in a peculiar degree as guardian or founder of a city.
The type of Asklepios was frequently transferred without alteration to heroes of
the healing craft, witness the examples of Amphiaraos in Oropos and of Trophonios
in Lebadeia ; even the statue by Timotheos in the temenos of Hippolytos in Troezene
represented the hero according to a scheme borrowed from Asklepios, but certainly
unbearded.6 More definite evidence for naming our statue is supplied by two marble
replicas of it found in Kyrene. One of them, in the British Museum,7 is in excellent
1 Cades, cl. ii. D, 16: without mention of owner or of species; probably an emerald-plasma. The gem
belongs to the series which give exact copies of statues. The agreement with the bronze is striking.
2 Skulpt. 68. Head Pheidian, like the Dresden Asklepios (p. 55, note 6). The original to be dated about
440 B.C., earlier than the Asklepios of the more recent Pheidian school (p. 89, note 6).
3 Wolters, Ath. Mitth. xvii. p. 1 seq. Taf. ii. — iv. Among the fragments of reliefs from the Asklepieion in
Athens I noticed in 1877 three examples of the type.
4 Cf. Wieseler, Gdttinger Nachrichten, 1888, p. 143^?.
' Marble statuette, Central Mus., Kabbadias, Cat. No. 270; copy of an older work ; weight on left leg,
left arm supported, mantle leaving right breast free, beautiful youthful head with falling curls of moderate length ;
the head seems to belong to the statue, although the preservation of the surface differs in both ; the rolled fillet
shows at any rate that the head is that of a beardless Asklepios. The snake is at the side, as we assumed for
the Florentine Asklepios (p. 208).
6 Paus. ii. 32, 4, who calls it simply an Asklepios, but adds that the Troezenians called it Hippolytos. The
statue of Hippolytos with spear and dog, which occurs on coins of Troezene (Imhoof-Blumer and Gardner,
Num. Comm. PI. M. viii.), is certainly not, as Wieseler (loc. cit. 146) thinks, the Asklepios of Pausanias,
but another Hippolytos statue.
7 Smith and Porcher, Hist of Disc, at Cyrene, p. 103, No. 74 ; for the discovery, p. 77. Newton, Guide
Graeco-Rom. Sc. ii. No. 114. /. H. S. iv. p. 46 seq. (Wroth) with illustration. The statue was found in the same

278 POLYKLEITOS
preservation, except that the upper part of the mural crown is missing, whence the
existence of the crown has hitherto passed unnoticed. The other replica, in Edinburgh,
is known to me from illustration and description j1 it is slightly modified in drapery
and hair. Both figures are inferior productions of the later Roman period. The one
in the British Museum certainly does not preserve much of that Polykleitan style
which the bronze reproduces so well. Traces of this style are more visible in the
body of the marble replica than in the head, for, though the hair is fairly exactly
copied, the mouth and the expression appear to be softened down. The drapery
answers, fold for fold, to the bronze. The discoverers suggested Aristaios as an
interpretation, even without noticing the mural crown, and the existence of the crown
only confirms their hypothesis. Aristaios, a divinity closely akin to Asklepios, guardian
of Kyrene, institutor of the culture of the silphium, and by its means founder of the
power and wealth of the city, son of the eponymous goddess Kyrene, satisfies all the
conditions attached to the interpretation of our statue. Aristaios, son of Apollo in
Kyrene, was also called son of Paion,2 an epithet which designates him as a divinity
of the healing craft like Asklepios. He was brought up by Cheiron the physician,
and in Kyrene, as Pindar's allusion proves,3 was considered to be a divinity allied in
nature not only to Apollo but to Zeus. This may account for the head type of our
statue ; the youthful age recalls Apollo, but the strong regal locks correspond to the
type of Zeus, which was already fixed in the period to which we have assigned the
statue. We suppose, then, that the original was made for Kyrene by an Argive artist.
Afterwards the motive was repeated elsewhere. A late echo of it is given by a bronze
bust found in Gaul, a faithful enough though rough reproduction of the youth wearing
a mural crown.4
Among other works belonging to the more extended circle of the Polykleitan
school may be mentioned a beautiful bronze statuette of a youthful athlete in the
Louvre ; it is in place here because its motive is allied to that of the Dresden ' boy
victor.' The attitude of the feet, the turn and bend of the head, correspond ; the
eyes, again, look down towards the outstretched left hand ; but the right arm is raised
horizontally ; the youth appears to be holding the fillet in his right hand and letting
it glide over the open palm of the left hand. The rendering of form is in its ground
work Polykleitan, though later realistic traits have also crept in. I think it probable
that work like this was done by remote followers of Polykleitos in the fourth
century, and among them by Kleon of Sikyon. The bronze is apparently a Greek
original.5
small temple at Kyrene as the well-known 'Kyrene and Libya' relief. Cecil Smith confirms my opinion
that remains of a mural crown are to be seen on the head.
1 Michaelis, J.H. S. v. 157. Cecil Smith had the kindness to obtain for me more exact particulars from
Dr. Anderson : it seems that the broken portion on the head ' might very well have been a mural crown.' Under
the crown is added the rolled fillet which belongs to the Asklepios type. Good illustration in the Archaeologia
Scotua, Transactions of the Society of Antiquaries of Scotland, vol. iv. Taf. 16, p. 337. The head is more
softened down than in the British Museum copy, the curls are longer and fuller at the sides, but the symmetric
parting over the forehead is the same. The drapery is stately ; a three-cornered piece falling over is indicated.
The head is rather more raised. With these exceptions it agrees with the other replicas.
2 Pherekydes, frag. 10 (Miiller).
3 Pind. Pyth. 9, 63 : e^aovrai re viv dtidvarov Zijra Ka\ ayvhv 'A*6xXuva.
4 Frohner, Coll. Griau, Bronzes Ant. PI. 43, No. 1 108.
« Formerly in Coll. Pourtales, then Greau ; well illustrated by Frohner, Colt. Griau, Bronzes Ant. PI. 32,
No. 964. Frohner s assumption that the figure is pouring oil is inconsistent with the position of the hands.
For the above motive with the fillet cf. the Nike on a coin of Elis of severe style, Gardner, Types PI 3 42

XENOKLES

279

VII. The Basis of the Statue of Xenokles. — The Idolino.
The third Polykleitos basis at Olympia, which supported the statue of Xenokles,
a boy victorious in the wrestling contest, is of great significance, for by its means
we are able to connect with Polykleitos, in a more definite way than we could otherwise
have done, a series of extant figures.
The basis (Fig. 118)1 shows both footmarks in complete preservation ; round the
deep cavity for insertion the outline of the foot itself is distinctly visible. Each
foot was 23 cm. long ; hence the scale of the figure was the same as that of the
'boy crowning himself The attitude was one of
repose ; the weight of the body was on the right
leg ; the left leg was slightly advanced and set to
one side, with the foot flat on the ground.
The same motive in which Xenokles was repre
sented was also applied to another Polykleitan figure
of a youth at a more advanced stage of adolescence.
This is proved by an admirable bronze statuette of
the Louvre, which, as is plainly to be seen from
external marks,2 is an undoubted genuine Greek
original.3 (PI. XIII. and Fig. 119.) This work must
belong to the immediate circle of the master's
pupils, and have been directly inspired by the master
himself, for it manifests his style in all its purity,
and the execution is of marvellous perfection.4 The
forms correspond throughout to those of the Doryphoros and Diadumenos : they are
strong and virile, and distinct from the forms that pertain to adolescent boyhood. We
find here all the characteristics of the Polykleitan body: there is, in addition, an indication
of veins on the lower part of the abdomen — a trait which we have frequently met with
before, and which already occurs on some copies of the Doryphoros. The head is an
excellent piece of work, the hair finely incised in the usual Polykleitan style.5 The
feet are placed flat on the ground side by side, the left foot, free from the weight of
the body, being set slightly forward and to one side. One arm again follows the
contour of the supporting leg, according to a favourite Polykleitan scheme. The
position of the right hand, held with the thumb drawn in, corresponds to that of the
Idolino, and is best explained if we imagine it to have held a kylix. The left hand, a
little advanced, carried an attribute formed like a staff. The head is bent, and looks
towards the supporting side. The bronze can hardly represent an athlete, as in that
case the act of pouring libation which seems to be here represented would not be
appropriate.6 Some heroic or divine person whom we have no means of naming
must be represented here. It is precisely in the Polykleitan circle that athletic types
would be likely to influence even the types of divinities.

Fig. 118. — Basis of Xenokles in Olympia.

1 Drawn by Liibke from my plan of the basis.
2 Longperier, Notice des Bronzes Antiques, 214, ' Mercure aptere.' The eyes of silver, inserted, the nipples
of copper. Height o'2l. Cf. Jahrb. d. Vereins d. Alterthumsfr. im Rheinl. Heft 90, p. 53.
3 For instance, each foot is pierced by a bronze nail which fastens it to the basis, a process only found in
earlier Greek originals. The colour of the bronze, too, a splendid dark green, is peculiar to Greek, and more
especially to Argive works (cf. 50th Winckelmannsprogr. p. 127).
4 The back is specially fine. 5 The pubes is less carefully rendered.
fi I at least know no instance of an athlete pouring libation.

Fig. 119.— Bronze statuette in the Louvre.

Printed by GlesecKe & Devrient, Leipzig (Germany}.

282 POLYKLEITOS
belongs to the statue, but has been broken and put on again, and a small piece of the
neck is restored. The head is in bad preservation, chin, nose, and almost all the
mouth being new. The right arm is restored from the middle of the upper arm, the
left from the middle of the forearm.1 The left thigh has been incorrectly turned
outwards (a piece between it and the trunk being new), and the shin has been wrongly
restored. In another replica, without head, in the Vatican, both the feet are ancient.2
Besides these two the Vatican possesses two replicas which are only torsos.3 Two more
torsos are in the Museo Torlonia and in the Palazzo Colonna respectively.4 A replica in
Naples corresponds perfectly, except that a chlamys is added on the left shoulder. In
this case the original type was evidently used for a Roman portrait-statue.5
The head alone is preserved to us in good replicas in the Museo Torlonia 6 and
in the Hermitage," and in a much-defaced copy in Dresden.8 With the help of this
material we can form a judgment on the lost original. A work of Polykleitos which
exists in so many copies must have been famous ; probably it only represented an
athlete.9 The bodily forms manifest the manner of Polykleitos in its full purity, with
new exquisite distinctions indicative of the period of life. The transitions, especially
at the edge of the ribs, are less hard than in the Doryphoros, and the head seems to
indicate that we have to do with a somewhat later work of the master. The figure is
closely analogous to the Herakles. Small locks of hair fall round the forehead :
these are parted in the middle, and are not symmetrical, but are arranged differently
on each side.10 The hair does not yet cover the ears, but just touches their edge,
without falling over ; it grows low down on the nape of the neck, ending in a curved,
not in a straight line. The hair is altogether richer and more individual than in the
Doryphoros, but is treated as a whole in the same flat, spare, and closely clinging
manner.11 The forehead projects decidedly at its base, and shows more variety of
modelling than in the latter statue ; the mouth is slightly open. The bend of the
head alone gives it an expression different from that of the Doryphoros, an expression
of mild and serious character, with a touch of melancholy, that harmonizes with the
tranquil attitude. The old ideal of Argive sculpture, a resigned, self-sustained repose,
has in this work become once again a living reality.
1 Ancient puntelli in the thighs show the position of the arms.
¦= Gall, delle Statue, 392, ' Septimius Severus.' The emperor's head does not belong to the figure. Palm-
tree stem as support. Both forearms new ; one can still recognize that originally the right was slightly advanced
and the left hung down. Both feet are ancient, the soles placed as on the Xenokles basis. Good workmanship.
Sala a croce greca, 590. A wrong protrait-head has been placed on the torso, the whole neck restored,
the left thigh incorrectly joined on (piece between restored). Pubes worked over. In the vestibule of the public
entrance to the Museum, torso restored as Meleager ; head does not belong.
Museo Torlonia, 76. Restored as a Diskophoros ; the head is a modern imitation of the Diskophoros of the
Vatican. The right forearm to the wrist antique ; the left arm with the diskos new. Palazzo Colonna, Matz-
Duhn, 16. Head not genuine ; torso good; right forearm extended forwards. The puntello is higher than in
the fiist Vatican copy. ~
5 Naples, Inv. 6055 ; Clarac, 925, 2351 ; Bernouilli, Rdm. Ikonogr. ii. 1, p. 172, 15. The Roman portrait-
head appears not to be genuine. The feet are restored, but the left leg is ancient nearly to the ankle, which gives
tne atitude Bodily forms and pubes are of purely Polykleitan style; the round chlamys on the shoulder is
evidently a Roman addition.
6 Museo Tori. 469.
< Guedeonow, 83. Conze, Beitrdge, p. 7, note 5. Good copy ; only the point of the nose restored. Hair
on tne top of the head carelessly rendered except at the ends. The lips have incised edges in imitation of
Bronze. .Length of face 182 mm., chin to nose = nose to eyebrow = two eye-lengths (65 mm.); width of
moutn 40 mm.
8 From Rome, lately presented by P. Arndt.
,1 J.he l3a,m-tree as support on one copy is evidence for this.
u I, C°pieS match very exact)y in the details of the hair.
The pubes is rendered as in the Doryphoros, only the curls are still daintier and richer in detail.

BRONZE HEAD 283
A bronze head from the same villa at Herculaneum (Fig. 121, «and b)1 which gave us
the beautiful bronze copy of the Doryphoros head by Apollonios so strongly resembles
the type we have just been discussing— which we shall call the 'Vatican athlete
standing at rest ' — and corresponds to it so closely in scale, that one would be tempted
to regard it as a copy from the same original, did not small differences, which cannot
be due merely to the copyist, bear witness that its original is to be sought for in a
distinct though somewhat similar work. The locks falling over the forehead are fuller,
and the arrangement is more symmetrical, while the treatment of the hair round
the ears, as well as the full plastic rendering of the whole mass, show great similarity
to the Dresden boy ; on the nape of the neck the hair ends in a straight line. All these
are points in which the Herculaneum head differs from the Vatican type. Owing to
a lack of skill on the part of the copyist — for the face is slightly askew — the expres
sion is one of intense melancholy, amounting almost to grief and distress. But the
forms are essentially the same as in the Vatican type, the forehead again projecting
strongly at its base. The original must have been later in date than the Doryphoros.
At this point we may mention two more Roman bronze statuettes. One, repre
senting a Mercury, is in the Museum at Spires ; 2 it follows the type of the ' Vatican
athlete standing at rest,' except that the head is not bent, but turned straight round
to the right. Face and hair are clearly derived from Polykleitan types ; a purse is held
in Roman fashion in the right hand, a kerykeion in the left. This is not a work
based on a Hermes by Polykleitos, it is only another adaptation of a Polykleitan type
to a Mercury. The same is true of a Hercules statuette,3 the head of which, however,
has already lost its Polykleitan character, and shows that type of youthful Herakles
which sprang from an Attic source and became current at a later period ; the body
is a repetition of the Polykleitan figure standing at rest ; the lowered right hand held
the club, the left has the apple.
It is here that we must place the Idolino, the famous bronze statue of Florence,4
for its motive connects it immediately with the series under consideration. The
principal characteristics, the position of the feet, the attitude of the head and arms, are in
essence the same, yet important differences are to be observed. The left leg, free from
the weight of the figure, is placed much more to the side and outwards than in the
Xenokles and kindred Polykleitan works. Again, the way in which the left hand
hangs down, so that the arm forms an angle towards the leg, gives to the figure a
special individual charm,5 and is without parallel in any of the works we have
assigned to the master himself. In this, as in the position of the legs, expression
is given to a rhythmic feeling other than that which we found in the work
of Polykleitos: a comparison with the Paris bronze, for instance (Plate XIII.), which
is in other respects so similar, will make this clear.
In order to appreciate justly the bodily forms, we must decide the age of the boy.
He is at a stage of growth midway between the 'boy crowning himself and the
Dresden boy. The chest is broader and more developed than in the first, sparer than
1 Inv. 5610; Bronzi d' Ercolano, i. Tav. 53, 54, P- 187 ; Comparetti de Petra, Villa Ercol. 7, 4. The
measurements correspond to the St. Petersburg head above. Technique and workmanship are as in the Dory
phoros of Apollonios ; the eyebrows are incised in the same manner ; the two works may be dated about the
same time.
2 (Harster) Katalog d. Histor. Abt. d. Museums in Speier, 1888, p. 25.
3 Known to me only from a cast in Dresden.
* Best illustrations in Brunn-Bruckmann, Denkmdler, Nos. 274 — 277. Cf. Kekule, Ueber die Bronzestatue
des sog. Idolino.
5 Kekule' has remarked (p. 8) that this is a trait which, in the Renaissance, Donatello was the first to employ.

THE IDOLINO

285

in the second. The same result is arrived at from the measurements : the figure is of
life-size, like the two others ; the distance between the nipples and the length of the
foot come exactly midway between the corresponding measurements of the other two
figures of boys.1 On the whole the forms harmonize with Polykleitan style ; the hips,
the navel with the depression of the linea alba continued below it, the decided hollow
at the side of the gluteus of the leg at rest — all are Polykleitan features. And yet in
this instance too there are differences of no slight importance. Above all, the abdomen
runs down to a much more acute angle than it ever does in the work of Polykleitos.2

Fig. 122. — Head of the Idolino (Mus. Naz., Florence).

The modelling of the Idolino is still under the influence of old Argive tradition, such
as would be handed on by the bronze of Ligurio and the Stephanos figure.3 Again,
the Idolino lacks the lifelike and refined surface details such as we find them in Poly
kleitos ; the serrated muscle, for instance, is not indicated. The forms as a whole are
1 Distance between the nipples, 215 (in the boy crowning himself, 205 ; in the Dresden boy, 225) ; length
of foot, 243 (in other two statues, ,233 and 253 respectively).
2 It is drawn in towards the membrum, while in Polykleitan modelling the abdomen is always convex till it
reaches the sudden division made by the line of the pudendum or the pubes.
3 Cf. 50th Berl. Winckelmannsprogr. p. 137 seq. The Pompeian bronze Apollo has the same form of abdomen,
which in other copies is remodelled after the later fashion.

286 POLYKLEITOS
wirier and less full than in the figures of boys by Polykleitos, and more intentional
stress is laid on boyish spareness of flesh.
If these facts forbid us to derive the work from the master himself, yet we may
look for its source at no great distance from him. This is made very plain by the
head (Fig. 122), which bears the most unmistakable marks of the circle of Polykleitos —
more particularly of the period just preceding the Diadumenos. We see the customary
skull, elongated and angular, the characteristic hair, inclining to straightness, the various
layers of short locks arranged one over another on the upper part of the head, the
symmetric parting of the small pointed strands over the forehead, the drawing of the
hair on the nape of the neck as in the Doryphoros and elsewhere in Polykleitos, the
position and shape of the narrow ear with its long lobe attached to the cheek, the same
shape of nose and flat nostrils — finally, the wonderful limpid purity and tranquillity, so
proper to Polykleitos. The distinction which marks off this head from those assign
able to the master consists chiefly, as it does in the body, in a more abstract treatment,
in a still more restrained and flat rendering of form.
The Idolino is, from technique and execution, evidently a Greek original ; only
the round, moulded basis on which it stands must be of Roman date, since that sort
of basis was not given to bronze statues at an earlier period. It is still plainly to be
seen that the feet are hollow below, therefore they were formerly fastened with melted
lead to a stone basis in the usual Greek fashion. Although the execution is of a
kind characteristic of original work, yet we miss the degree of refinement which we
have a right to expect in a master of the first rank. Our studies have shown that
this statue cannot be by Polykleitos himself, but at most by an artist of his circle,
who, however, seems to have followed a path of his own.1 We may venture to
remember in this connexion that side by side with Polykleitos the artist Patrokles,
probably his younger brother, worked and taught (cf. p. 226) ; we conjectured that he
had attained a certain independence of action within the general tendencies of style
prescribed by Polykleitos. All that we can, however, positively assert about the
Idolino is that it belongs approximately to the period of the Doryphoros, and is the
later rather than the earlier of the two.
The inner surface of the outstretched right hand is left rough, evidently because
some object lay upon it ; the fingers are in the same position 2 as in the Paris bronze,3
a position which can only be satisfactorily explained by restoring the statue with a
kylix in this hand. The left hand is smooth inside, and was clearly empty. The boy
was therefore represented pouring a libation. Now this does not fit in with the
current interpretation of the statue as a victorious athlete, since, as we noticed above,
the motive never, so far as we know, occurs in statues of athletes. If not an athlete,
then what is the nature of the boy represented here ? For although boys serve
the libation — in fact, this is one of their special functions — yet I know of no single
instance in all ancient art of a mortal boy performing the ritual act of pouring
it.4 In the same way they do not sacrifice, but only assist the celebrant. If
then a boy be represented pouring a libation, he cannot be a mortal, he must be
divine. Thus the youthful river god on coins of Selinos pours a libation over an
altar. We have naturally no means of naming the god or daemon represented
1 Cf. my remarks in Jahrb. d. Ver. v. Alterthumsfr. im Rheml. Heft 90, p. 53.
2 They are not in the slightest bent, as Kekule (loc. cit.) wrongly assumes.
3 Cf. the bronze published by me in Bonner Jahrb. Heft 90, Taf. ii. and ibid. p. 53.
4 The material has been collected in Stephani's treatise on representations of libation in Compte Rendu,
1873. Cf. also Fritze, de libatione veterum Graecorum, Berliner Dissert. 1893.

ARISTION 287
in our statue. That he should appear in the image of boy-victor is a genuine
Argive trait.
An entirely different view to the one put forward here concerning the style and
origin of the statue is held by Kekule,1 who classes it with the Munich Oil-pourer and
the standing Diskobolos, and pretends to detect in it the same Myronian qualities
which caused Brunn to bring the other two statues into connexion with Myron. But
there is no trace in the Idolino of just those very characteristics which Brunn found
in the two Attic statues. The concentration of the whole attention on one act which
governs by its own inner necessity every movement of the body, while beautifully
illustrated in the Oil-pourer and Diskos-bearer,2 is entirely absent from the Idolino.
His attitude is quite independent of his act — -the libation ; he might hold any other
object instead of a kylix in his right hand, and the effect of the whole would be
unaltered. The principal features of the Idolino motive were fixed creations long
before they were used for a boy making libation. We have seen that the motive was
current in the cycle of Polykleitos ; we may now add that it was developed directly from
the old Argive canon, and differs from it only in one important point — namely, in the
reversal of the sides. The motives of the two Attic statues, on the other hand, are
new creations, invented for the actions they represent. In the case of Argos the
form is ready to hand, and significant content is given to it at the will of the
artist ; in the case of Attica the significant content brings the form into being.3
A marble statue of the Palazzo Barberini in Rome (placed unfavourably
high) 4 must be a copy of an original similar to the Idolino and belonging to
the same artistic circle ; in fact, it might almost be a copy of the Idolino itself. The
left foot is here placed rather more back, so that the heel is raised ; the forms of
the body are harder and more restrained ; the head is very similar. The arms, so
far as they are preserved, correspond.

VIII. The Basis of the Statue of Aristion.— The later Polykleitan School. — The
Beneventum Head.
There is also evidence that the old canonical position of the feet employed
by Hagelaidas and the Argive school 5 was used in the cycle of Polykleitos : this is
proved by the basis in Olympia which once supported the Aristion, a work of the
younger Polykleitos (cf. p. 224, and Fig. 123).6 The footmarks are here intact,7 and show
that the figure stood on the left foot, while the right was placed flat on the ground to
1 loc. cit. p. 10.
2 And also e.g. in the boy pulling out a thorn, which in an early essay I accordingly referred to Myron (Dei
Dornauszieher und der Knabe mit der Gans, 1875, p. 79). Kekule (Arch. Ztg. 1883, p. 238, note 26) remarks on
this point : ' It is quite impossible to refer to Myron all ancient works of art which represent a figure absorbed
in a momentary occupation.' Doubtless correct, but the boy pulling out a thorn is at least one of the works in
which the absorption in one act makes the whole motive, while the Idolino is not. Again, the boy pulling out a
thorn belongs to Myron's period, the Idolino is later.
3 It is still more difficult to follow Kekule in his analysis of the forms of the Idolino. From what has been
said in a previous chapter on Myronian form, it will be seen that within the fifth century there could hardly be
found a sharper contrast than between the Diskobolos and other Myronian works on the one hand and the
Idolino on the other.
* Matz-Duhn, III I. Right arm and left forearm restored ; palm-stem beside right leg.
5 Cf. pp. 52 sqq., 190 sqq.
6 Drawing by Lubke from my plan. The drawing in Arch. Ztg. 1879, 207, is inexact.
7 Each foot was 29 cm. long.

POLYKLEITOS

"i Centimetres.

Fig. 123. — Basis of Aristion in Olympia.

the side.1 This is the scheme of the old Argive school, already so fully discussed.
To understand exactly how it was used by artists of the Polykleitan circle, we must
study existing statues, and especially a Hermes type which is repeated frequently and
in many variants.
The copy in Lansdowne House is at once the best preserved and reproduces
most accurately the Polykleitan style (Fig. 124).2 The god stands as did the Aristion
by Polykleitos the younger : the right arm hangs simply down, as in the statues from
the school of Hagelaidas — even the hand
corresponding exactly to that of the so-
called Stephanos athlete ; the left forearm
is stretched forward horizontally holding
an object, doubtless a kerykeion ; the left
shoulder is lowered, and the head turned
to the left. On the left shoulder lies a
chlamys arranged with great refinement in
fine natural folds. There are no wings
either on head or feet. We know from
the figure of Pan how carefully the Poly
kleitan school when representing divinities
suppresses all peculiarities not purely
human. Chlamys and kerykeion were
enough to characterize the god. The
right hand was empty.
The head repeats a Polykleitan type
similar to the. Dresden boy and the Petworth Oil-pourer. The hair is parted
symmetrically in the middle ; the forehead projects decidedly at the base as in the
' Vatican athlete standing at rest,' to whom the bodily forms are also analogous.
This figure can certainly not be referred to the elder Polykleitos. The forms of
the body are lacking in fulness and vigour ; they are spare and dry to a greater degree
than is usual in the works of Polykleitos, and thus approach more nearly to the manner
of the Idolino. We found in the modelling of the latter a survival of old Argive
tradition. The same tradition has evidently influenced the attitude and carriage of
this figure. The rhythm of the composition as a whole corresponds to the old Argive
canon, and is foreign to the manner of Polykleitos, who always knows how to round
off his work so as to produce a more harmonious and satisfying impression. But
the figure was undoubtedly, as we see, especially from the head,3 made under the
immediate influence of Polykleitos, to whose later works it presents marked affinities,
so that it is among the younger members of his circle that we must look for the artist.
We suggested the name of Patrokles in connexion with the Idolino ; here we might
think of the sons of Patrokles, one of whom, Polykleitos the younger, as the Aristion
basis proves, actually employed the motive we are at present discussing. His brother
and teacher, Naukydes, was the author of a Mercurius, according to Pliny. We
may hazard the conjecture that this his most famous work was the original of
our statue.
1 The statue of the boxer Eukles by Naukydes (Lowy, /. G. B. 86) stood in a similar attitude. The foot was
about 33 cm. long, hence the figure was rather over life-size.
2 Drawn from » photograph ; badly reproduced in Clarac, 946, 2436 A ; described by Michaelis, Anc.
Sculpt, p. 446, 35. Well preserved. No restorations, except possibly the left hand ; head unbroken. Distance
between the nipples 0-23 ; neckdiollowto navel 0-365, to pubes 0-53 ; workmanship ordinary.
3 Cf. the forehead, which is modelled throughout.

THE LANSDOWNE HERMES

We thus obtain the interesting and not improbable result that Argive artists, most
likely Patrokles and his sons, though working side by side with Polykleitos and
dominated in many ways by his influence, did
not cease for all that to foster old Argive
traditions. There is a replica (headless) of the
Lansdowne Hermes in Berlin ; *¦ the wiry
spareness of the forms is here characteristi
cally reproduced. Wings are added to the
feet. In a third replica in the Palazzo Pitti 2
the drapery has ceased to be an exact
copy, while the head, which seems to belong
to the statue although the neck is restored,
has been entirely transformed in character ;
it is adorned with wings, and shows only
feeble remains of Polykleitan modelling.
The head is still more altered in a second
Berlin statue,3 which reproduces the drapery
in a much simplified form without interest
or beauty. The formation of the body in
these two statues is somewhat softer and
feebler, but otherwise riot essentially different.
Another replica from Aegion,4 an inferior
work of Roman period, retains the charac
teristic spare body, and has at least an echo
of Polykleitan manner in the head ; more
over, the wings are absent ; but the drapery
is completely altered, being wound round
the forearm in the familiar later fashion. A
statue in Naples,5 with a head foreign to
it, corresponds closely with the figure from
Aegion ; the forms of the body are charac
teristic notwithstanding the alteration in the
drapery, and are faithfully copied from the
original. The same is true of a copy in Palazzo
Colonna, the head of which is also foreign.6
Finally, the figure was completely
transformed to suit coarse Graeco-Roman
taste ; it appears with full realistic muscles,
while the attitude became more animated and more nearly allied to the Lysippian,
through the altered position of the right foot, which was now placed more to the
side. The statue by Antiphanes of Paros in Berlin 7 and the ' Mercure Richelieu '

Fig. 124. — Hermes in Lansdowne House.

1 Skulpt. 196. Part of the kerykeion lying on the arm is ancient.
2 Diitschke, ii. p. 9, No. 16. Cast in Dresden. The right leg is wrongly restored as drawn back. The end
of the kerykeion on the left arm seems to be old. 3 Skulpt. 199. Wings on the feet.
4 Central Mus., Athens, Kabbadias, Cat. No. 241 ; Ath. Mitth. iii. Taf. 5. The illustration is inexact,
especially in the face and hair. Remains of kerykeion in the left hand.
5 Inv. 6073 ; Clarac, 942, 241 1 (wrongly given without pubes).
6 Head of Trajan put on in a wrong attitude. Neck restored.
7 Skulpt. 200. Found in Melos ; head new. P P

290

POLYKLEITOS

in the Louvre x are instances of this transformation of an original whose principal
features however are still evident through all the changes. The same hardly holds
true of a Roman statue from Atalanti,2 with the turn of the head reversed ; the
statue has thus acquired quite a Lysippian air ; moreover, the head is modelled
on Lysippian prototypes. Thus the original ended by becoming entirely transformed :
instead of charmed repose, we now have impatient restless energy.
We have spoken of these variants in the order of their deviation from the original,
but this is not the chronological order. The Atalanti statue was probably derived
from an original of the circle of Lysippos himself — an original which really represented
an attempt to modernize an old type. From this Lysippian original later and inferior
artists borrowed separate details, such as the arrangement of the chlamys, and com
bined them with the bodily forms and attitude of the old original ; at any rate we
may imagine the different variants to have arisen in some such way.
At this point may be cited some few Hermes heads which manifest Polykleitan
character in a greater or less degree (cf. p. 231). The familiar Lansdowne head with
the petasos 3 is a sketchy, flat, incorrect piece of work, behind which however seems
to lurk the same original that is reproduced in the Lansdowne statue. A head in St.
Petersburg,4 showing holes for the insertion of wings, is a more faithful mirror of
Polykleitan style ; there is no definite known original for this work ; the artist
satisfied himself with a general adherence to Polykleitan types, and added the head-
wings, which are quite inappropriate. Besides these, I know of two heads of less than
life-size, Polykleitan in type and related to the Lansdowne statue.5 They have small
antique wings. There are two more, also less than life-size, wearing the round
petasos, and in style not far removed from Polykleitos.6
We close here the series of works immediately connected with Polykleitos.
Of more distantly related creations we may now discuss at least the most
important. And first the bronze figure of the ' Praying Youth ' from Carinthia.7 „ In my view
this is not a Greek original dedicated by Romans of a later period, but, to judge by
technical indications,8 a Roman work of the same period as the inscription.9 It seems
to be a free imitation of an Argive statue of a victor. The style is not Polykleitan ;
the hard and spare forms are more in the manner of the Idolino, i.e. they point to
Patrokles as the artist.
On the other hand, a bronze head from Beneventum in the Louvre (Plate XIV.)
is a splendid Greek original, and belongs to a statue of a victorious athlete which was
1 Frohner, Notice, 177 ; Clarac, 316, 1542 ; photo, edit. Giraudon, 1196. Neck new, head genuine.
2 Central Mus., Athens, Kabbadias, No. 240 ; Gazette Arch. ii. Taf. 22, 23 ; Korte, Ath. Mitth. iii. p. 98, B.
Photos of the bust at German Institute in Athens.
3 Michaelis, Anc. Sculp, in Gr. Brit. p. 467, 88 ; Miiller- Wieseler, D. a. K. ii. 304.
4 Guedeonow, No. 179; Treu, Arch. Anz. 1889, p. 57, wrongly ascribed to Polykleitos.
5 (a) In the Louvre, Salle des Caryatides, No. 1487 ; Frohner, Notice, 197 ; photo, edit. Giraudon, 1273.
Length of face 14 cm. Neck new. Turn to the left. Bad work, (b) Vatican, Sala Geografica : similar.
b (a) Set on an insignificant torso, Museo Chiaramonti, 589. Turned slightly to the right. Polykleitan
manner evident in the hair. Round petasos as on the Lansdowne head, (b) Palermo. Placed on a torso to which
it does not belong ; height 12 cm. ; sketchy work.
7 R. v. Schneider, Die Erzstatue vom Helenenberge, Wien 1893 (Jahrb. d. Samml. d. Oesterr. Kaiserh.
Bd. xv.) 8 The metal has the yellowish colour of the good Gallo-Roman bronzes of the first century a.d. ; the casting
very thick, no fine chiselling. The eyes are not inserted, as in all large Greek bronzes, but cast with the head ;
all these are indications of non-Greek origin.
9 Inscription was originally only lightly cut ; it was worked over in the sixteenth century.

XIV.
Bronze Head of a Boy.

LOUVRE.

BRONZE HEADS 291
taken from Greece.1 A wreath made of two sprays is worn in the hair : as these
appear to be kotinos twigs, the statue probably once stood in Olympia.2 The
essential elements in the structure of the face, which projects towards the middle
and recedes in large surfaces to the sides, the mouth and nose with its
oblique slope and flat nostrils, are characteristically Polykleitan ; yet there is at
the same time something alien to Polykleitos and akin to Pheidias and the Lemnia
in the infinite charm of this head. The hair is at the stage of the Diadumenos. A
peculiarity is the asymmetria of the locks of hair over the middle of the forehead ; a
similar treatment is to be seen in the Polykleitan Pan ; only at the sides does the
arrangement become symmetrical. Clearly the artist, while deriving his inspiration
from Polykleitos, was open to Attic influence.
The no less magnificent bronze head of a boy wearing the victor's fillet in
Munich is a kindred work, also a Greek original.3 Here too the face, especially the
nose and mouth, and the shape of the head, are closely analogous to the works of
Polykleitos,4 but at the same time we can trace, although less definitely, a certain
foreign element probably due to Attic influence. A marble head in the Vatican, the
production of an inferior copyist, from its great similarity must, I think, be derived
either from the Munich head or from an analogous work by the same artist.5 A basalt
statue of a boy from the Palatine 6 shows the same analogy in the head ; the body is
still under the influence of the old Argive canon, but a slight turn of the head to the
right (the side free from the weight of the figure), and a forward movement of the
right shoulder, give a somewhat easier rhythm to the composition. The master to
whose hand these works are due is interesting as being nearly related to the Argive
school, without however belonging to it.
In conclusion, we must mention a third bronze head, again an admirable original
work — in Naples 7 — a boy wearing a wreath or a fillet in his curly hair, and therefore
probably a victor. Face, mouth, nose, forehead, and shape of head again show clear
marks of Polykleitan influence ; but the hair and the expression are different. The
same holds true of a marble head in Berlin.8
Thus the influence of Polykleitos spread to wider circles. In his own immediate
surroundings, however, he held almost absolute sway. Although we have succeeded in
separating to a certain extent the activity of master and pupils, we have found in this
1 Cf. p. 18, note 1. Frohner obtained the head from Beneventum. Casts are to be had. The original
shows that it was part of a statue. The technique is Greek, but the chiselling not very detailed : well published
in Brunn-Bruckmann, Denkmdler ; cf. Monum. et Mimoires, fondation Piot. PI. 10, 11, also good vignette in
Kalkmann, Prop, des Gesichts, p. 27. 2 Remains of small leaves in front.
3 Glypt. 302 ; Friederichs-Wolters, 216. Brunn supposes it a later casting, but the technique shows that the
work is a Greek original.
4 Polykleitan character has been recognized by Flasch (Philologenvers. Innsbruck, p. 162) and by Brunn
(Glypt. 5th ed. ; Bayr. Sitzungsber. 1892, p. 658). Kekule sees here a further development of the style of the
Olympia sculptures and of the Spinario (Arch. Ztg. 1883, p. 246). I cannot see any analogy to these works, which
bear the impress of a totally different individuality.
5 Museo Chiaramonti, 475. The scale seems to be about the same. The hair differs slightly, the ear being
left freer. 6 Now in the Terme. Photo in German Institute at Rome. Weight of the figure on left leg. The right
hand hung down, the left held something shaped like a staff, the mark of which is visible on the breast.
Twig in hair, hence a copy of a victor's statue.
7 Naples, Inv. 5633 ; Bronzi d' Ercol. i. 73, 74 ; Rayet, Monum. de V Art Antique, ii. 67 ; Comparetti de
Petra, Villa Ercol. xi. I. The line of the wreath or fillet is quite visible. The technique seems to show that
the work is a Greek original. The head has no connexion whatsoever with the ' Dancers ' of Herculaneum, which
are only copies.
8 Skulpt. 479, where it is wrongly described as a replica of the Naples head discussed above,

292 POLYKLEITOS
inner group of workers comparatively little personal individuality. Only Patrokles
and his kin seem to have shown some independence, although still in a very limited
degree. In the circle of Pheidias we found much greater originality. But the severity
of school tradition has ever been one of the essential conditions under which Argive art
developed ; as indeed there was more that was capable of being learned in the Argive
school than in the school of Pheidias. Polykleitos did not excel in invention or wealth
of meaning ; in fact, he found his inspiration for a whole series of statues, as we have
seen, in foreign Attic creations, such as the Diadumenos of Pheidias, the Oil-pourer
of Lykios, the Amazon and the Diomede of Kresilas.
We, like the ancients, know no creations of Polykleitos except youthful, beardless
figures. This springtime of life is chosen for divinities as well as for human beings
whenever it is at all congruous — as in the case of Dionysos, Hermes, Ares, Herakles,
Pan ; the artist seems to have represented the Dioskouroi as boys playing with
knuckle-bones.1 But within these limits perhaps no artist has gone so far in repre
senting the more delicate shades of distinction between boyhood and adolescence.
Nor is his choice of motives nearly so uniform as has hitherto been believed.
Again, within the series of works of the artist himself we found a development
leading gradually from the Doryphoros to the Diadumenos. It is true that the
harmony of reserve and repose characterizes all his creations. Yet the Doryphoros,
the Amazon, and kindred statues have more freshness and energy, combined with
reasonable restraint, while the later works are penetrated by a softer spirit and a
greater degree of sentiment. In this second period the master even sometimes
abandons the walking motive for the standing position of complete repose — a change
for which Attic influence is no doubt partly responsible.
We have already had occasion to verify many instances of the influence of
Polykleitos on later art, and to trace many survivals of his style. But the most
important researches in this part of the subject are still to come.
1 That the pueri astragalizontes, which would be inexplicable as a genre group by Polykleitos, represent
the Dioskouroi seems probable from the Greek gem published in Roscher's Lexikon, i. 1 174, even if the gem is not
dependent on the work of Polykleitos.

SKOPAS PRAXITELES EUPHRANOR

SKOPAS PRAXITELES EUPHRANOR

T

'HE succeeding ages of antiquity subsisted off the
immense achievements of the art of the fifth
century. Our increasing knowledge points with
ever-growing certainty to the conclusion that the greater
number of the most important types, both of men and of
gods, had already been elaborated in the fifth century. In
our previous inquiries we have been able to give some
proofs of this, but it is in the fourth century more par
ticularly that the influence of the preceding period may be
traced for a long time more clearly than is generally sup
posed, even as we can trace the current and the colour
of a mighty river long after it has mingled with other
waters. The Eirene of Kephisodotos affords a clear and well-
known instance of this fact. This work by no means marks
a normal transition between the style of two epochs; nor
is Kephisodotos, as he is generally said to be, a ' forerunner '
of the great fourth-century masters ; he is one of them
himself, and may very well have been the elder brother — though scarcely the
father— of Praxiteles. The few extant dates relating to him belong to a period
when Praxiteles must already have been at work,1 and there is nothing to
justify the assumption that these dates coincide with the close of the career
of Kephisodotos ; the contrary is far more likely.2 A serious difficulty in the
commonly received genealogy of the three successive generations — Praxiteles I.,
Kephisodotos, Praxiteles II. — is caused by the gap of half a century which inter-
1 The Eirene of Kephisodotos should probably be dated after 375 to correspond with the institution of the
annual offering to Eirene consequent on the victories of Timotheos (Isokr. xv. 109 ; Corn. Nepos, Timoth. 2 ;
cf. Wachsmuth, Stadt Athen, i. 585, ii. 433 ; Curtius, Stadtgesch. v. Athen, p. 206 ; Wolters in Jahrb. d. Inst.
1893, p. 178). The statue was vowed in 374, and executed soon after, i.e. probablyin 371 — 370, on the occasion of
the festal Peace Congress in Athens (Von Stem, Gesch. d. spartan, u. theban. Hegemonic, p. 149). The date thus
found would tally with that assigned by Pliny to Kephisodotos (01. 102). The large group in the temenos of Zeus
Soter in Megalopolis seems from the late character of the architecture lo have been a work of the younger
Kephisodotos (Dorpfeld in Ath. Mitth. 1893, 218). On the other hand, the groups of Praxiteles in Mantineia
fall soon after the rebuilding of the town (370). Pliny's date for Praxiteles (01. 104) is probably fixed by the battle
of Mantineia (B.C. 362) (cf. Ldwy, Unters. *. Kiinstler Gesch. p. 64).
2 Kephisodotos's sister was the first wife of Phokion. Kephisodotos was therefore a contemporary of
Phokion and of Praxiteles. He loo made a 'Hermes with the child Dionysos.'

296 SKOPAS
venes between the dates of the elder Praxiteles (cf. supra, p. 102) and those of
the two younger artists. Probably an intermediate member of the family — father
of the two younger artists — has been lost to tradition.
The Eirene of Kephisodotos,1 though belonging to the years B.C. 375 — 370, follows
in the treatment of the drapery the manner which was in fashion some sixty years
earlier : it goes directly back to Pheidias, and forms a strong contrast to the whole
intervening development, i.e. to the style of drapery that obtained in the time of the
Peloponnesian War, such as we see it in the ' Caryatids ' of the Erechtheion, the Nike
temple, and the works of Alkamenes, Agorakritos, and Kallimachos ; it abandons
the thin, clinging, semi-transparent tissues, and turns to the earlier forms ; once
more the folds fall perpendicularly from the knee of the ' free ' leg, and this leg is
represented, not in the ' walking attitude,' but only, as in the Parthenos, set slightly
to the side and very slightly drawn back. On the other hand, the infant Ploutos 2 is so
like the little Dionysos by Praxiteles, and the whole motive and spirit of the group so
akin to Praxitelean works, that we at once feel that we are no longer on the thres
hold of the new epoch, but completely within it, and that the artist is intentionally
striving after an earlier style. In a monument of such political import as the Eirene
this was probably not a mere accident. The statue was set up in the early years of
the mew maritime confederacy which (so the Athenians hoped) was to revive the^
ancient glories of their city. Thus the sculptor in his image of Eirene — of the
Peace for which the naval alliance was to be the pledge — reverted to the style of
the brilliant period of the old confederacy under Perikles.
I. Skopas. — Lansdowne Herakles. — Hermes from the Palatine. — Statues with Foot
raised. — Ares Ludovisi. — The Meleager. — Athena.
The influence of the Argive artists of the fifth century upon the Attic artists of
the fourth was especially strong7~ I shall first let the monuments tell their own tale.
The beautiful and excellently preserved Herakles in Lansdowne House (Fig. 125) 3
reproduces without a doubt an important original of the fourth century. The type of
head is thoroughly Attic ; its predecessors are the Munich ' oil-pourer ' (Lykios ? p. 289),
the standing Diskobolos (Alkamenes ? pp. 90 and 260), and the Florence athlete (p. 26),
while closely akin to it are the youth on the relief from the Ilissos,4 two bronze heads
from Herculaneum,5 and other analogous works.6 Now it is jvident that these heads,"
and that of the Lansdowne statue more especially, form the basis for that type of
Herakles which has been assigned to Skopas on good grounds,7 and which in reality
is only the earlier type slightly intensified by an admixture of pathos. The motive
1 Friederichs-Wolters, 1210, 1211 ; Kabbadias, 46v. jnova. 175. A replica of the Eirene, restored as Niobe,
in the Mus. Torlonia, 290 ; a good replica of the child Ploutos in Dresden, head antique.
2 On the Athens and Dresden copies the head survives, and strongly resembles the child Dionysos of
Olympia, as do the body and the drapery. The little head from Ikaria is also very similar (Americ. Journ. of
Arch. 1889, p. 474).
3 Spec, of Anc. Sc. i. 40 ; Clarac, 788, 1973 ; Michaelis, Anc. Sculpt, p. 451, 61 ; Kalkmann, Prop. d. Gesichts,
p. 61. From Hadrian's villa. Very well preserved. A replica of the torso among the restored statues of the
Palazzo Massimi alle colonne. Palazzo Pitti, Diitschke, 35, is a later modification, and so is the bronze Colossus
of the Vatican, Roscher's Lex. i. 2179.
4 Kabbadias, liv. flow. No. 869 ; Wolters in Ath. Mitth. xviii. p. 6 ; Von Sybel, Weltgeschichte d. Kunst,
Fig. 204 ; Zeitschr. fiir bild. Kunst, N. F. ii. 293.
5 (a) Comparetti de Petra, Villa Ercol. Tav. vii. 3 ; Friederichs-Wolters, 1302. (/;) Comparetti de Petra,
loc. cit. Tav. x. 2. Probably a youthful Herakles with twisted fillet, olive-leaves, and fruit ; looking to the left
like the Lansdowne Herakles. (a) is somewhat more severe in the hair, and goes back to an older original.
0 e.g. A good head of an athlete in the Museum of the Peiraieus ; one in the Terme Museum, Rome (Rdm.
Mitth. 1891, p. 304, 2, etc.) 7 By B. Griifj KSlllm Mitth_ l88Q) p> l99 seq_

THE LANSDOWNE HERAKLES 297
of the Lansdowne statue is also genuinely_AiticJn the freedom and boldness of its
attitude, with the left leg placed flat on the ground to one side, and the head turned in

Fig. 125.— Herakles in Lansdowne House.1
the same direction as the free leg. The right arm hangs down, and the left forearm
is extended, carrying some object. In this form the motive is purely Attic, and was
1 [From the cast in Dresden. A good illustration from a photograph of the original is given by Kalkmann,
Prop, des Gesichts, p. 61, — E. S.] 0 <J

298

SKOPAS

elaborated in the fifth century.
walking motive, and by giving a
forhis

Fig. 126. — Heroic statue (British Museum).

from it in the bodily forms,

Kresilas altered it by the introduction of the
more energetic turn to the head, and used it
his Diomede. Later, Polykleitos followed in
steps. 1
As an example of the original Attic composi
tion I may cite the interesting statue of a young
hero (Fig. 126),1 which must belong approximately
to 440 B.C. The motive is precisely that of the
Lansdowne Herakles. A chlamys, falling in firm
vigorous folds, is flung over the hero's left shoulder.
The stylistic forms of the beautiful lifelike body are
obviously derived from the Apollo of the Omphalos
(Kalamis ? p. 81), even to so small a detail as the little
furrow above the navel. The head also is connected
with the series, which I have already assigned (p. 81 seq.)
to the school of Kalamis, but the treatment is some
what freer : the hair has a strongly plastic quality,
the eyes are heavy-lidded, the mouth full and open,
resembling in expression and shape the mouth of the
Herakles in Fig. 32. We therefore have here another
work of that school of Kalamis which flourished
alongside that of Pheidias. It jw_as^fjxuxL-statues__of
this type that the motive of the Lansdowne Herakles
was developed.
Far simpler had been the conception of Herakles
in the Pheidian circle, to judge from a copy of a
Herakles which so closely resembles the Farnese
Diadumenos (p. 244) in the bodily forms, in attitude,
and in bearing that it can only be ascribed to
Pheidias, or to an artist influenced by him.2 The
hero, with his lion-skin over his head, his left foot
at ease and placed with full sole to one side, looks
straight out to the right to the side of the support
ing leg.
Though the Lansdowne Herakles is closely con
nected in conception with all this Attic work, it differs
hich seem influenced much rather by Polykleitan models 3

1 Copy in Brit. Mus., Newton, Guide Graeco-Rom. Sc. i. No. 134, from the Pal. Farnese ; Fig. 126 drawn
from a photograph. Head broken but genuine, the lower part of legs modern, but the position is shown by
what remains. Puntello for the lowered right arm. A still better copy of the torso in the Louvre (photo.
Giraudon, 1289). A replica of the head in the Lateran, No. 891, broken under the chin, wrongly turned to the
right. Both copies of the head have swollen ears ; cf. the Riccardi bust, Fig. 66, and the Diomede, in which
the chlamys may also be compared.
2 Statue under life-size in cortile of Pal. Borghese (Matz-Duhn, 90 ; Winckelmann, Mon. Ined. 78). Head
unbroken, upper half restored. Phot, in German Inst, at Rome.— Similar statue in Museo Lapidario of Vatican,
No. 132. Attic-Pheidian of a rather severer type; except for the bent head, the motive is the same as in the
Borghese statue.— A statue of Herakles in the Louvre corresponds so closely to the Ares Borghese that it must
be either a later Roman copy or an adaptation from the workshop of Alkamenes of the Ares to. the Herakles
type. The head does not belong to the statue, which is identified as a Herakles by the lion-skin on the trunk,
which is ancient (see D'Escamps, Marbres Campana, PI. 40; Zeitschr. f. bild. Kunst, N.F. ii. p. 24, where Von
Sybel wrongly considers the head to be authentic, and connects it with Skopas).
3 Cf. also Kalkmann, Prop. d. Gesichts, p. 60, note 3.

THE LANSDOWNE HERAKLES

299

(as a comparison with the Diadumenos and the Doryphoros clearly shows) : the
undulations of surface are, it is true, softer and rounder, but the forms of breast,
stomach, navel, etc., are those characteristic of Polykleitan figures.
We have already become acquainted (p. 248, n. 4) with a direct predecessor of
the Lansdowne Herakles, in a terminal statue of the Museo Boncompagni-Ludovisi,
and a torso at Dresden, in which we likewise detected a Polykleitan structure of body
combined with an older, Attic type of head, approximating in character to Kresilas.
The motive — the shouldering of the club — is the same, but the position of the arms
is reversed. It is obviously more natural to shoulder the club with the right hand,
but the position of the arms in the Lansdowne statue is more rhythmical, and
moreover follows the Polykleitan model — the Doryphoros — more closely.
The Lansdowne Herakles does not stand alone. Among analogous works may
be cited a torso from the Villa Borghese,1 representing — as the long fillet-ends on the
shoulder prove — either a Herakles or a victorious athlete, and another torso in the
Museo Torlonia.2 Of still greater interest is a splendid bronze Herakles in the
Jacobsen collection ; it repeats the motive of the
Lansdowne figure, except that the left arm holds
the bow and arrow instead of the club ; 3 the head
shows the same fourth-century Attic type, and the
body is again Polykleitan in its forms. Nor must
we forget another beautiful bronze Plerakles from
Dalmatia, now in Vienna,4 carrying the club on the
right shoulder ; here again the slightly drooping
head is thoroughly Attic, and recalls the Florentine
athlete (p. 261), while the body shows the influence
of Polykleitos. This same combination appears in
a superb bronze statuette of Zeus at Munich, while
in a bronze Zeus in the British Museum (Fig. 127)5
the Polykleitan influence extends from the bodily
forms to the whole attitude, which is closely
copied from the Doryphoros ; the head, however,
with the short locks brushed up from the fore
head, is the exact counterpart of the Lansdowne
Herakles, thoroughly Attic in its Skopasio-Praxi-
telean character. Further, a noble bronze statuette of Zeus in Munich 6 manifests,
although less clearly, Polykleitan influence in the body, combined with an Attic
position of the legs, and an Attic type of head.
The Lansdowne Herakles, owing to likeness of the head to the later Skopasianj,
1 Torso in front porch of the villa ; the forms allied to the Doryphoros, but rounder ; veins on the
abdomen. The left leg was somewhat advanced as in the Lansdowne statue, and the head was turned to the
left. Good workmanship.
2 Museo Torlonia, 5 ; allied to the Doryphoros, but rounder ; head turned to the left.
3 Ny Carlsberg Glypto. nyt tyllaeg 1892, No. 1039 a. The type is that of Herakles ; remains of the quiver
strap on the breast ; the position of the fingers of the left hand is characteristic for holding the bow and arrow
(cf. the statuette mentioned in the following note, the ' Paris ' of the Aegina marbles, and Berlin Skulpt. No. 51).
The statue is an admirable work, but the form and ornamentation of the basis prove that it is not an original
of the fourth century, but a more recent repetition of such an original. The bronze statuette from Thessaly
(Oesterr. Mitth. iii. Taf. 7, 8), with Polykleitan forms, is an analogous work.
4 R. v. Schneider in Jahrb. d. Kunsts. d. Oesterr. Kaiserh. ix. p. 135 seq. Taf. I, 2.
5 Payne-Knight, xlvi. 16 ; Clarac, 802 G, 1917 D. Good execution, but from the colour of the metal
evidently not a Greek original. Right hand lowered and holding something, left hand pierced. Copper nipples
inlaid. 6 Friederichs-Wolters, 1750.

Fig. 127. — Statuette of Zeus (British
Museum).

300

SKOPAS

type of Herakles, may not impossibly be copied from an early work of Skopas
himself. A statue of the youthful Asklepios, which must have been celebrated in its day,
has been preserved in a copy on a medallion of M. Aurelius,1 and in a beautiful bronze
statuette of the Roman period (Fig. 128),2 in which
we have to supply the serpent-wreathed staff which
supported the right hand. In this bronze the bodily
forms here again betray Polykleitan influences ;
further, the ' walking ' motive and the sharp inclina
tion of the head towards the side of the free leg
recall creations of the Polykleitan school, such as
the Pan (p. 269 seq) and the works allied to it, while
the position of the hands — the one leaning on a
support, the other resting on the hip — very closely
resembles that of the so-called Narkissos (p. 273).
The statuette, however, avoids all sharp transitions,
and has a grace of movement and a suppleness of
modelling such as we seek for in vain in those
heavier Argive works. The head, with its charming
short locks brushed up from the brow, is genuinely
Attic. A beardless Asklepios is actually known to
us from tradition by Skopas. Pausanias saw it in
H;he god's temple at Gortys.3
There once stood on the Palatine a beautiful
and interesting statue of Hermes (Fig. 129),4 which
has now been removed to the Terme Museum. The
alert grace of the god's pose already suggests the
-Apoxyomenos of Lysippos ; just as in the latter
statue the right leg, which is free, is placed well to
the side and slightly drawn back ; and the youth
balances himself on his hips while the gaze is
directed to the side of the free leg ; the right
forearm was held sideways 5 from the body ; it
doubtless grasped the herald's staff, if not so ener
getically, yet much in the same manner as a tech
nically exquisite bronze Hermes in Berlin,6 which
is also quite Lysippian in style. Further, the great
—depression of the left shoulder — while the right is raised and thrust forward — is
already suggestive of that fine swing which marks not a few of the creations which

Fig.

-Bronze statuette of Asklepios
(Carlsruhe).

1 Cf. von Sallet's Zeitschr. f. Numis. ix. 140 ; Numis. Chron. ser. iii. vol. ii. PI. 14, 3 ; Wieseler, Gdtting.
Nachr. 1888, p. 152 seq. Wieseler very plausibly maintains the interpretation as Asklepios, against Wroth,
who had suggested Apollo.
2 At Carlsruhe, found at Speier : Schuster, Bronzen, PI. 27; Friederichs-Wolters, 1758. Roman work, probably
made in Gaul. Ring on finger of left hand. Back specially good.
3 It is possible, however, that a beardless Asklepios in the Braccio Nuovo (Helbig, Museums, 6), with a head
of later Skopasian type, reproduces this statue.
4 Matz-Duhn, 1046. The body has suffered severely through long exposure. The head has been broken off,
but is undoubtedly authentic ; it alone is very well preserved. The legs are ancient. Replica of head in Catajo
(Arndt-Bruckmann, Einzelv. 52, 53).
5 The stump of the arm is placed backwards and to the side, and there is no puntello on the right side of
the body. « Antiquarium, Inv. 6505 ; from Athens.

HERMES

301

-at any rate in

/SWfc;

can be traced back to Lysippos.1 Now the Herakles of
the statue of which a copy exists in the
Louvre,2 and which has the type of head
assigned to Skopas — has this same move
ment of. the upper part of the body, and
likewise rests one hand (the left) on the hip.
The extraordinary thing, however, about the
Hermes of the Terme is that the bodily forms
and the hair are still treated in the manner of
the fifth century ; the hair lies close to the
skull, and is divided into fine locks, pointed at
the ends in true Polykleitan style ; 3 yet the
forms of the face are purely Attic, and show
that the original cannot be earlier than the
fourth century. The head fascinates by its
exquisite charm and subtle refinement ; the
eyeball is slightly turned both inwards and
upwards ; it is the beginning of that forma
tion of the eye so chara^teTisTJc'ofSkopasian
heads ; the bridge of thenosTTs^rather broad,
as in the Hermes of Praxiteles, but in spite
of this the work is undeniably somewhat older
than any^ofjhe works hitherto acknowledged!
to be by Skopas or than the Hermes of Praxi-S
teles. The~motive of the statue has been
shown to be one of the prototypes that in
spired Lysippos, and previous to him it seems
to have been employed only by Skopas.
Therefore, may not the Hermes of the Terme
also be an early work by this artist ?
There is other authentic evidence to prove
that, during the middle of The fourth century,
the school of Skopas not unfrequently fol
lowed fifth-century models in its treatment of
the hair and of the bodily forms. The Hermes
on one of the sculptured drums from Ephesos4 is as closely connected with the

Fig. 129.-

Hermes from the Palatine (Muse delle
Terme).

1 Particularly the one called Alexander, and the youthful warrior in the Louvre, which may also repre
sent Alexander (Clarac, 264, 2100); further, the Herakles type of which a bronze (Brit. Mus.) is a specially
good example (Clarac, 785, 1966). Cf. Roscher's Lex. 2172. The position of the legs is thai of the
Apoxyomenos. 2 In the Salle des Caryatides, No. 1524 (published by Graf, Rdm. Mitth. iv. 1889, p. 193). The type of
head on PI. 8, 9 (ibid.) belongs to this statue, and this only. The existence of so many copies of the head and of
only one of the statue is explained by its frequent adaptation as a term (cf. p. 66, note 5, p. 235), and this also
explains the fact that so many of the heads are reversed from right to left. The Paris head is an inferior but
an authentic copy of the same original as the single heads, and is unbroken. The original certainly wore the
poplar-wreath. No copies of the Paris statue have come under my notice, but the coin of Sikyon (Graf, loc. cit.
213) probably reproduces the original of the Paris statue. The head is differently placed, but in so rough a coin
that means little.
3 This is probably what Matz-Duhn means by the 'reminiscence of Doric types' in the head.
4 Friederichs-Wolters, 1242. The burning of the temple in 356 forms the terminus post quern. The relief
is probably near this in date. Skopas himself made a pillar relief (Pliny, xxvi. 95). As no other artists are
named in connexion with the other reliefs, these were probably made by his colleagues and pupils.

302 SKOPAS
Hermes of the Terme in the treatment of the close-lying hair and the broad well-
defined surfaces of the body as it is with Polykleitan models ; at any rate, every
thing about it differs entirely from the youth on the Ilissos relief, from the
Meleager, or from the Hermes of Praxiteles. Even the 'walking motive,' with
the arm hanging idly on the side of the supporting leg, recall Polykleitos. Thus
everything conspires to commend the view that the works we have grouped
together are connected with Skopas, and testify to a phase in which the artist was
influenced by Polykleitan models.
He was, as has been very plausibly conjectured,1 the son^of An'sUndres of Paros,
who, since he worked with the younger Polykleitos on a votive offering set up at
Amyklai by the Spartans after the victory at Aigospotamoi, must have been
connected with the Argive school. Thus Skopas would naturally be early imbued
with the traditions ~oT~that school, and the peculiarities of the works enumerated
above, some of which may possibly be ascribed even to Aristandros, can be easily
accounted for.
Archaeologists have tried to trace this Polykleitan influence in the works of
Skopas, but by looking for it in the head they have gone completely astray. Only
in the^hair does a Polykleitan touch sometimes occur, otherwise the Contour of the
head and the shape of the face are absolutely the reverse of Polykleitan, and are
invariably, as was remarked in the case of the Lansdowne Herakles, dependent on
the purest Attic types.
From the Polykleitan school the young Skopas appropriated quite a number of
formal^ elements, but in that spiritual conception which finds expression in theTpose
and the presentment of a figure, and in the forms of the head, he was entirely under
Attico-Ionian influence, thus reproducing in the fourth century a phenomenon that
had its counterpart in the fifth. The Parians to whom we owe the Olympian sculp
tures had likewise been influenced by the Argive school in formal matters, though in
their spiritual conceptions they had remained genuine Ionians.2 What at that period
must still be called Ionian was afterwards entirely absorbed into Attic art. We may
therefore consider Skopas also as an Attic master.
By far the most novel and audacious of the works which we attributed to his
yearly period is the Hermes of the Terme. It introduces a rhythm full of restless
energy, hitherto utterly unknown to statuary, and which prepares the way immedi
ately for Lysippos. In this statue the individuality of Skopas bursts forth in a most
decided manner.3 Yet as if to set a check on his own audacity he shows himself
here rigidly conservative in matters of form. The Hermes exhibits a singular com
bination of fire and grace, and, though Lysippos eventually surpassed Skopas in the
expression of nervous tension and excitement, the internal tenderness and warmth
of his predecessor were always lacking in him.
Skopas, by representing figures with one foot raised on an elevation, once again
forestalled Lysippos4 in another motive full of restlessness and of broken outlines.
1 Bockh on C. I. G. 2285 b. ; cf. Ldwy, /. G. B. 287.
2 Cf. Archdol. Stud. H. Brunn dargebr. 1893, P- 69 seq.
3 A statuary composition which is known to me in two torsos seems to be an older, hesitating attempt in the
direction of the Palatine Hermes. A youth of delicate form (which however shows Polykleitan influence) is leaning
his weight on his left leg, his left hand rests on his hips, the head is turned to the right like the Hermes, but unlike
it the right leg is not drawn back, and the right arm hangs by the side (torso in Vat., Gall. Lapid. 87, and another
less well preserved in Museo Torlonia, 38, PI. 10). An inferior replica reduced in size (Vat., Gall, dei Candel. 24)
gives the boy a Satyr's tail. Closely related is one of the small athlete statues, Braccio Nuovo (97). Torso only
ancient. Head plaster.
4 According to the extant dates, the activity of Lysippos extends from 350—300 (King Seleukos after 306, cf.
Lowy, 487), so that the period of Alexander marks the middle of it. Winter (Jahrb. d. Inst. 1872, 169) is

APOLLO SMINTHEUS— PAN 3°3
The attitude had been common in relief and painting from the time of Polygnotos
onwards, but it scarcely seems to have been adopted for the round previous to
Skopas.1 Thus he represented his Apollo Smintheus2 resting, his right foot onan"=
elevation and bending forward, laurel-branch in handfas if to play with the mouse
below, while his left hand rests on his side according to a favourite Skopasian scheme
already noticed. This same attitude reappears in a statue of the youthful Pan—
(except that in this case the head is raised and the eyes look straight out) — which
presumably stood at Heraia in Arkadia, -as it is reproduced on a silver coin3 of that
town (Plate VI. 35), and in a statuette found in the neighbourhood.4 We may
surmise that fjiis_Earx^also_ is_a n earlier- work of SkopaSjjor perhaps even a work by
his father, since it must have been executed at the beginning of the fourth century.5"
Two marble copies6 have perhaps preserved for us a later creation of Skopas ; they
represent a delicate boy with forms allied to those of the statue of the striding
Hyj)nos,_which we shall later recognize to be Skopasian, and with his left foot
placed upon an elevation ; the left arm rests on the thigh, but the body is turned
somewhat to the right, producing soft lines on the whole abdominal region, and
giving a momentary character to the movement. We get an analogous though far
tamer and quieter conception in a well-preserved statue of the Capitol.7 A youth,
with his cloak round his lower limbs, raises his right hand to emphasize some quiet
impressive harangue, and lets the left one rest on his leg, which is well raised. In
style and conception this figure resembles Praxiteles rather than Skopas ; the left
hand hangs at the side, and presents a striking similarity to that of the Hermes
at Olympia, and the head is also related. We may conjecture that here too Hermes
is represented as Logics or Agoraios, and that the sculptor, follower of Praxiteles
though he was, copied Skopas in the motive of the raised foot.8
wrong in placing Lysippos at the beginning of the seventies. In editing the inscription of Troilos (Arch. Ztg.
1879, 146) I conjectured that his second victory fell in the Olympiad after 01. 102, but it may also have been
considerably later ; and above all the wording of the inscription, as Lowy points out (/. G. B. p. 76), makes it
probable that the statue was erected long after the victory. The statue of Pulydamas by Lysippos was certainly
set up long after the victory of 408 B.C. (Cf. Urlichs, Kunstsschriftsteller, p. 26. )
1 Cf. Samml. Sabouroff, to Taf. 114;. 50th Berl. Winckelmannsprogr. p. 161. K. Lange (Das Motiv des
Aufgestiitzten Fusses) maintains that it was introduced by Lysippos. Kopp (Bildnis Alexanders des Grossen,
p. 18) rightly states that it was known to art long before Ljsippos, but he does not distinguish between sculp
ture and painting. Many motives existed in painting long before they were introduced into free sculpture in the
round, e.g. recumbent sleeping figures certainly_do not appear in statuary before the Alexandrine period.
2 The statue of Apollo Smintheus found on the coins of Alexandria Troas from the second century B.C.
down to late Imperial times is not the statue by Skopaswhich Strabo mentions, but an earlier one with a different
motive and of much. severer style (Roscher's Lexikon, i. 457, Z. 53 sqq. ) A coin-type which appears only under
Commodus and Caracalla may reproduce the statue of Skopas (Plate VI. 34, 36). Weil, in Baumeister's Denkm.
p. 1669, agrees with me. Overbeck (Apollo, p. 92 sqq. ¦. cf. Coin Plate 5, 25, 27 — 33, also 5, 10, and p. 314) is
opposed to my view. According to Strabo, the mouse was in a hole under the round pedestal on which the
right foot rests, and in a coin of Apollonia ad Rhyndacum, recently discovered (Imhoof, Gr. Miinzen Abh.
bayr. Akad. 1 Cl. xviii. 3, PI. 6, 20, p. 609, No. 156), there is something undistinguishable, which may be a
mouse, in front of the pedestal. [It seems more likely, however, that this object is a raven : cf. Warwick Wroth,
Brit. Mus. Catal. Troas, Introd. p. xvii.; the raven is quite, distinct supra, Plate VI. 36 = Wroth, PI. vi. 2 ; the
Skopasian character of the statue remains none the less indisputable. — E. S.]
3 Small silver coin about 417 — 370 B.C. Brit. Mus. Catal. Pelop. PI. 34, 12.
4 Berlin, Inv. 7486, from Andritsena (Arch. Ztg. 1881, p. 251). Simple, spare bodily forms, certainly from
the first half of the fourth century. B Because of the coin and the style of the bronze statuette.
6 Louvre, Rotonde d'Apollon, Nos. 3065, 3066. Clarac, PI. 271, 2193, 2194. Life-size. Head of No. 3065
antique ; neck, chin, and lower jaw restored. Interesting type, with brushed-back hair and fillet. Head possibly
does not belong to body. 3066, head modern, support of left foot restored as a rock.
7 Helbig, Museums, 509 ; Clarac, 859, 2170. The left hand and most of the right hand are antique. The
' Sandal-binder ' is quite a different type. I know no replicas.
8 More Skopasian is a small bronze in Berlin (Friederichs, Kl. Kunst, 1850), representing the youthful
Herakles, or an athlete with a fillet. The right leg is raised in the act of striding, the head looks outwards.
I(s excited expression recalls the Skopasian Herakles.

304 SKOPAS
i There was yet another energetic system of broken lines, produced by represent
ing a figure with both hands clasped round one knee, which Skopas seems to have
been the first to transfer from painting to statuary in the round. The earlier con
jecture, which referred the Ares Ludovisi to Skopas, seems perfectly reasonable. The
statue is certainly a reduced copy of the Mars sedens colossiaeus in templo Bruti
Callaeci (Plin. xxxvi. 26). The figure on the arch of Trajan, in which Stark
thought he could identify a copy of the Mars, has turned out to be more probably
a Herakles type borrowed from a cult instituted by Trajan.1 A seated Ares is,
however, something so curious and unique that we are predisposed to imagine a
connexion between the Skopasian statue known to have been in Rome and the
Ares Ludovisi. The style settles the question. To begin with, it is not in the
least Lysippian, as is now generally supposed.2 Let the head of the Apoxyomenos
be compared to that of the Ares — and if there is any one whom the totally different
spiritual conception of the two works should fail to convince, let him compare the
foreheads. The Apoxyomenos is the earliest statue of a youth that can be dated
in which the horizontal modulation of the forehead is represented as a wrinkle ;
in the Ares, as in all pre-Lysippian works, the modelling of this part is smooth.
If it be correct, as I believe it is, to attribute the Meleager to Skopas3 (it also
passed till lately as Lysippian), then the Ares must "ai^o^6e^by__this artist. In
both the heads the powerfuljr^rrairig-^f'the^^s/'wlrich are deep-set and peer eagerly
into the far distance, the quivering mouth and nostril,4 are as thoroughly like each
other as they are genuinely Skopasian. The Ares Ludovisi cannot be an exact copy,
partly because it is a reproduction on a reduced scale of a colossal statue, partly
because the workmanship is hurried and superficial ; further, the little Eros between
the feet of the god clashes with the style of the statue, inasmuch as it is of Graeco-
Roman type, and, like a second Eros 5 that has now disappeared, is an addition made
by the copyist. We have already seen (p. 89, note 8) how fond the Roman artists
were of representing Ares as a lover, and how they introduced this suggestion into the
older Greek works by some trifling additions. Skopas certainly did not intend to
pourtray a lover lost in dreams, but a war god, of his essence restless, who when
he pauses gives no thought to his personal dignity or to the outer world, but sits as
he likes to ponder over fresh feats of arms. On the Parthenon frieze Pheidias had
already chosen this attitude as characteristic of Ares ; but it was Skopas who first
had the courage to work out the idea in a colossal work in the round.
The Meleager takes rank, it is true, among the usual statuary themes, yet it also
is genuinely Skopasian, fraught with fire and energy. The inclination of the left
shoulder, the energetic turn of the head towards the side of the free leg, and the bend
of the arm which rests on the hip, have already been often noticed in Skopas, though
1 Petersen, Rom. Mitth. iv. 1889, p. 330 sqq.
2 Friederichs, Bausteine, 436 = Wolters, 1268; Schreiber, V. Ludov. 63 ; Helbig, Museums, 883 ; Roscher's
Lexikon, i. 490 seq. M. Mayer (Arch. Anz. 1889, p. 41), arguing from the supposed Lysippian character, suggests
Piston's Mars in Rome.
3 Graf, Rdm. Mitth. 1889, 218 sqq. In addition to the replicas mentioned by him there are : Torso in Pal.
Torlonia (restored as Hermes) ; torso in Louvre (from Rome), 1884 ; replica in Pal. Doria (no chlamys, head poor,
neck new, boar's head old, wrongly adjusted on restored trunk) ; head in Jacobsen collection, 1071, mediocre.
The Meleager of Skopas probably once stood in Tegea. [For the superb head of the Meleager placed on the torso
of a Praxitelean Apollo (infra, p. 338, note 3), in the Villa Medici, see Plate XV. ; Petersen (Rdm. Mitth. iv. 186)
was, I believe, the first to recognize this head as an original.— E. S.] 4 At least the right nostril is ancient.
5 More probable than the other restorations. M. Mayer's suggestion (loc. cit.) is the least fortunate of all.
A replica (torso) in Naples (Flasch, Verhandl. d. 41. Philologen. Versamml. in Miinchen) has no puntello on the
left shoulder, hence no Eros was added. The god in this replica was seated on a pillar, not on a rock. Shield
missing, but it may have been fastened on separately, as there is a contact-surface on the right side of the drapery.

ARES— MELEAGER— ATHENA 3°S
in the Meleager the restless vitality is intensified to a degree which makes the
Lansdowne Herakles and the Palatine Hermes appear simple and even tranquil
by comparison. The motive of the Meleager reminds me of a remarkable type of Athena, which
arose I incline to think from the fiery genius of Skopas. The goddess is conceived
somewhat like a Joan of Arc, in the semblance of a young, still undeveloped girl ;
her mantle is wrapt about her in manlike fashion, her left hand is firmly placed on her
hip, while with her right she grasps the spear ; vibrant with courage and enthusiasm,
her face slightly upturned, she looks out into space (Fig. 130).1 The deep-set eyes
and their well-defined cavity, the open mouth, and the whole expression recall the
Skopasian manner. An external circumstance confirms the conjecture. Beside the
goddess on her right hand may be seen a Triton ; the figure does not serve as a
support, still less can such an unusual attribute be a copyist's addition. Now we know
that near the ancient sanctuary of Athena at Alalkomenai there was a brook called
Triton, where, according to the Boeotian legend, Athena was born and brought up.2
To this story the statue obviously refers,3 and therefore Athena is represented as still
quite a young girl. Skopas made a marble statue of her to stand in front of the
temple of the Ismenian Apollo at Thebes.4 At a time when that city was making
an energetic claim to the leadership of Boeotia, it would be only natural to lay stress
on the legend of neighbouring Alalkomenai, and to represent the goddess as the
maiden born and brought up on the stream Triton, in that Boeotian land over which
Thebes was to rule. Of course Skopas may also have worked for Alalkomenai, and,
as we know that the shrine was plundered by Sulla, the statue might have been taken
to Rome and copied there.5 The somewhat spare forms and the treatment of the
drapery suggest that the statue is one of the master's earlier creations. His later
treatment of drapery is exemplified in his Apollo Kitharoidos, of which a copy has
been preserved to us in the famous Vatican statue,6 and which is another example of
the vigour of Skopasian conception.
If we consider the treatment of the nude, we find that in this respect the Ares
and the Meleager^are again intimately connected ; the forms show, moreover, that
(supposing our attribution of them to Skopas be correct) both statues can only belong
to the artist' s_]a^er_woxks1_so entirely does the treatment of the body differ from that
observed in those statues which we ascribed to his earlier period, and which show-
dependence upon Polykleitos. A great revolution has taken place : instead of being
1 Statue in Casino Rospigliosi, Matz-Duhn, 621 ; Miiller- Wieseler, ii. 233 ; Arndt-Bruckmann, Einzelverk.
No. in. Replicas : (a) Florence, Uffizi, Diitschke, iii. No. 152. Very good workmanship ; the head is genuine, the
neck and upper part of the breast restored. Photo, ed. Alinari, 1295. (b) Berlin, Skulpt. 73 ; Bull. Comm. di Roma,
1887, p. 169; Mon. d. Inst. Supp. 27, 1. The head alone in (c) Glienecke (Friederichs-Wolters, Gipsabg. 1438),
(d) Museo Chiaramonti, 558 (Helbig, Museums, 107), and (e) Berlin, No. 80. It is purely arbitrary and unscientific
to make out the head to be Hellenistic, as is usually clone. 2 Paus. ix. 33, 7 ; Strabo, p. 413.
3 As the Triton beside an Aphrodite in Dresden refers to her birth from Ocean. Arch. Anz. 1894, p. 29.
4 Pans. ix. 10, 2. Opposite a Hermes by Pheidias. Brunn (Bayr. Sitzungsber. 1880, p. 459) refutes Klein's
hypothesis that the two statues (called Pronaoi) were necessarily contemporary in date.
5 The original was, to judge by the replicas, probably bronze.
e Cf. Hoffmann, Philologus, 1889, 678 seq. ; Arch. Anz. 1889, p. 147 ; Roscher's Lex. i. 463. The Palatine
Apollo singing appears on the coins of Nero, not those of Augustus. (Overbeck, Apollon, Miinztafel v. 47, 48,
50, 51). The Apollo with the kylix on coins of Augustus (ibid. v. 43) was probably the original of the Barberini
statue in Munich, which has been (p. 88) referred to Agorakritos (cf. Studniczka, Rdm. Mitth. 1888, 296, who
starts from the untenable premise that the coins of Augustus reproduce the Apollo of Skopas).— The Vatican
statue (Helbig, Museums, 267) is a poor copy, much worked over, mouth and chin restored.— A replica in
Stockholm (Overbeck, Apollon, p. 186, 2). The beautiful head in the British Museum (Newton, Guide Gr.-Rom.
Sculpt, ii. 99) is a better copy of the head. The hair is waved differently. Pathos is expressed by the move
ment of the eyebrows.— Overbeck, loc. cit. 186, No. 3, 1, is an older type of the late fifth century.
R R

Fig. 130. — Statue of Athena in the Ufliz

XV.
Head of Meleager, placed upon a Praxitelean Apollo.

VII.l.A MEDICI.

^

PRAXITELES 307
divided into clear, sharply defined planes, the various surfaces melt insensibly into
each other ; the chest is narrower, the navel is no longer flat as in Polykleitan works,
but it is indicated by a soft depression, and the hard modulation of the abdominal
muscles has disappeared.
This treatment of the nude is still quite foreign to the sculptures of the Mauso
leum or to the great sculptured column of Ephesos. ThechajTgecannot therefore
have taken place before the middle of the fourth century ; it is however complete
in the youth on the_Jiissos-_relief (in this respect it entirely accords with the
Meleager), and in the Hermes of Praxiteles. These works must be about contem
porary with the Meleager and the Ares. The Skopasian Herakles in the Louvre is
somewhat older, and is more nearly akin to the earlier group, and forms a kind of
mean between the two series.
The same revolution of the treatment of the nude makes itself felt in the work of
Praxiteles, where its various stages can be far more easily traced. There will be
more to say about Skopas in another connexion, but for the present it seems best to
pursue further the point of view suggested by the treatment of the male nude.
II. Praxiteles. — Date of the Hermes. — Works of the Artist's Early and Middle
Period. — Figures leaning on a support ; the Satyr and the Eubouleus. — Works
of the Artist's Later Period ; the Hermes and Kindred Statues.
Our knowledge of the art of Praxiteles rests primarily on the extant original
statue — the Hermes of Olympia. It is of the very highest importance to obtain a
more definite date for this work, and to find out whether it belongs to the youth of the
artist, as Brunn asserts,1 or to his later period.
It has just been remarked that the Hermes is connected stylistically only with
the later works of Skopas, and that therefore it must belong to the second half of
the fourth century. This can be confirmed by another most convincing, although
external, piece of evidence derived from the actual basis of the statue. All its
essential parts have been recovered and used for the reconstruction in the Museum
at Olympia.2 It entirely differs from the ordinary Greek basis of the fifth and early
fourth century in height, in both the upper and lower mouldings, and in material.
The high moulded shape of the Hermes basis occurs in Olympia and elsewhere
only on bases which cannot be dated earlier than the Alexandrine period. Such
are the three bases of the artist Sophokles in Olympia,3 which correspond to the
Hermes bases, except that their lower moulding is richer (the cornice blocks are
missing). Others are the basis of a statue by Sthennis4 dedicated by King
Lysimachos (306 — 281), and the basis of one of the works of Xenokrates,5 both in
the Amphiareion at Oropos. In the Graeco-Roman period this form of basis
became quite common in Olympia as in other sanctuaries. As regards material, the
Hermes basis is made of white limestone, only the lower block being of grey lime-
1 Brunn, in Deutsche Rundschau, viii. 1882, p. 188 seq. Kekule was formerly the only opponent of this view.
He saw in the Hermes 'no youthful work, but an example of the full maturity of his (Praxiteles's) powers of conception
and execution ' (Baedeker's Greece, p. 97). Of late Overbeck has published a more detailed refutation cf Brunn's
opinion (Sachs. Ber. 1893, p. 46 seq.), but adds no positive information about the date of the Hermes.
2 Cf. Ausgrab. v. Olympia, Bd. v. 9.
3 Lowy, /. G. B. 123—125. According to Lowy's very probable hypothesis (/. G. B. p. 384), this artist was
mentioned by Pliny among the contemporaries of Alexander in the 121st Olympiad, in Bk. xxxiv. 51, where his
name should probably be restored from the corrupt fucles of the MSS.
4 Ldwy, /. G. B. 103 a (p. 384) ; 'Eipri/i. dpx. 1885, p. 102, 3 ; C. I. Gr. Sept. 279.
5 Ldwy, 135 b (p. 387) ; 'EipVp.. dPX. 1886, p. 55, 17 ; C. I. Gr. Sept. 332.

308

PRAXITELES

stone with round dark spots on the surface. Several varieties of this white and grey
limestone occur on the bases of Olympia, but only from the time of Alexander
downwards. The limestone of the earlier bases in Olympia is always of that dark
black sort which already occurs at an early period, and was subsequently much
affected in the circle of Polykleitos.1 The earliest dated analogies in Olympia are
afforded by the bases of the six Zeus statues called Zanes, which were erected out
of the fines which the Athenians paid, 01. 112 (B.C. 332), for the misdemeanours
of one of their citizens.2 These bases follow the earlier shape without moulding,
which was chosen because the statues continued a series of six earlier Zanes,
erected OI. 98 (B.C. 388). On the other hand, the material differs from that of the
preceding Zanes ; it consists of white limestone alternating with blue-black and
with white conglomerate. Besides these there is a red and white conglomerate, never
seen in earlier times, but to be found in the bases of the statues by Sophokles.3
These facts point to the second half of the century as the date of the Hermes, so
that the statue cannot be, as some maintain,4 a record of the peace concluded after
the events of the Anolympiad and the combat between the Arkadians and the
Eleians in the Altis in B.C. 363. Nor was there any inherent, probability in this
date, seeing that the reconciliation between the two states was a hollow one, arranged
by the aristocratic party in Mantineia : the suggestion, however, that Hermes the
god of Arkadia was represented as nursing Dionysos the god of Elis, in token of
the union of these two districts, is certainly correct. I accordingly incline to refer
it to a period twenty years later (B.C. 343), when the aristocratic party in Elis, in
league with the Arkadians,6 won a decisive victory over the democratic party, and
then formed an alliance with Philip for their own protection. This was a suitable
occasion for fresh Eleian dedications at Olympia,6 and the Hermes represented as
nursing the Eleian Dionysos might well be set up as a thank-offering to the
Arkadians for their aid. At the same time that he made the Hermes at Olympia 7
Praxiteles must have certainly made a Dionysos for this god's temple at Elis ;
the reproduction of the statue on a coin of Hadrian 8 shows that in design it was
very closely akin to the Hermes.
1 Quite different from the dark blue-black limestone, out of which so many bases in Olympia in the fifth
or fourth centuries are made (also those from the circle of Polykleitos, Naukydes, Daidalos, Kleon).
2 Paus. v. 21, 5 seq.
3 All the Zanes set up in 01. 112 (Nos. 7—12) had one kind of basis, made of three superimposed cube-shaped
blocks. The lowest block is always of poros ; the middle one of white limestone (yellow and of finer texture,
like giallo antico in No. 10, missing in No. 12; cf. Arch. Ztg. 1879, p. 45 seq.); the upper block is black
limestone with white veins in 8, 9, 10, 12 ; it is missing in 7 ; red-white conglomerate in No. 11. Two blocks
of the same material belonged to the Sophokles bases.
1 Purgold, in Histor. u. philol. Aufsdtze E. Curtius gew. p. 233 seq. Dorpfeld's recent investigations have
shown (Olympia, Textbd. ii. 33, 36) that the alteration in the Heraion assumed by Purgold never took place.
Sal. Reinach, Gaz. Archiol. 1887, p. 282, note 9 ; Revue Arch. 1888, i. sqq.
5 'ApxdSav ovvayiMio-aiitvwv ro'is 'HXeiois, Diod. 16,63. Cf. Curtius, Griechische Geschichte, iii. (6th edit.)
P- 623. 6 Several larger buildings in Olympia belong to the same period, e.g. the Leonidaion, the great rebuilding of
the east porch, the so-called Echo portico, which is approximately dated by its similarity with the Philippeion
(cf. Olympia, Textbd. ii. 72, 92).
7 This statue was certainly made on the spot, for it is executed with regard to the position it was to occupy
in the Heraion.
8 Weil, in Sallet's Numism. Zlschr. xiii. 384 ; Postolakkas, Catalogue 1883—84, Taf. 2, 9 ; Imhoof-Blumer
and Gardner, Num. Comm. p. 74. The left forearm was supported, the figure rested on the right leg ; the
right arm was raised high ; the Dionysos resembles the Hermes in every particular, except that he is looking up
towards the rhyton. It is not certain that the legs were crossed. The Berlin and Athens copies are too much
defaced to afford exact evidence on this point. If the legs were crossed, it would be another reason for placing
the statue in the later period of Praxiteles (cf. Samml. Sabouroff, Text to PI. 77, 78).

DATE OF THE HERMES 3°9
The period {circa B.C. 340) to which we accordingly assign the Hermes belongs
however to the master's later period : he was already an artist of repute some thirty
years earlier. His two great groups at Mantineia must be closely connected with the
rebuilding of the city. The date assigned by Pliny to Praxiteles (01. 104) probably
refers to the artist's activity in that city in so far as it is the date of the battle of
Mantineia (B.C. 362).1 Beyond this the only other dates we possess for Praxiteles are
derived from the fact that he worked on the Mausoleum at Halikarnassos (begun
B-c- 353), and also on an altar at Ephesos, of course after the fire of B.C. 356. There
is absolutely nothing to prove that he was still working in the time of Alexander.
The date assigned by Pliny to his sons (01. 121) is obviously incorrect, and too late.
The date is that of the sons of Lysippos, who come first in the list, and Pliny,
according to his custom,2 included within the period other artists who did not strictly
belong to it. The sons of Praxiteles are connected with Lykurgos and his family ;
the statue of Menander, who died in 291, may very well have been erected in his life
time ; 3 and the palaeographic character of the inscriptions 4 shows that the sculptors
lived well within the fourth century. Pliny's date therefore could only apply to the
end of their career, and involves no contradiction of our assumption that the father's
activity extended from circa B.C. 370 — 330.
The following observations further confirm our dating of the Hermes : the hair
of the god is fashioned with a free use of the drill — one main effect of the head being
due to its employment — the hair left rough upon it with the tool-marks has a look
of rich profusion, and rises very effectively above the smoothly polished flesh of the
face. This technical refinement is yet foreign to the Skopasian heads from the
pediments of Tegea, and to the head, probably also Skopasian, found on the south
slope of the Akropolis ; 5 nor do we find it in the great number of grave-reliefs of
the first half of the fourth century ; a few only, which from their style belong to
the last reliefs before the reorganisation by Demetrios the Phalerean, show this
use of the drill in the hair.
Lastly, the drapery of the Hermes is treated with a degree of subtlety abso
lutely without parallel among the works of the first half of the fourth century. The
drapery of the Muses on the Mantineian reliefs, which belong to the artist's earlier
period, appears, in spite of the wealth of folds, quite simple by comparison. The
system of enlivening the smooth surfaces between the large folds by breaking them
up by small curving or broken lines is in the Hermes carried out to the greatest and
most lifelike perfection. Only among the later grave-reliefs of the fourth century is
1 Cf. p. 295, note 1. As Praxiteles's name is the first under the date, this passage must have reference to him
(cf. Plinius und seine Quellen, in Fleckeis. Suppl. Bd. ix. p. 21 seq.), and is not to be lightly set aside.
2 Cf. p. 41. Plinius und seine Quellen, p. 22 seq. Only a complete ignorance of the character of Pliny's
dating can have caused Overbeck (Sachs. Ber. 1893,45 seq. ) to assume that 01. 121 was the 'Bluthezeit' of
Kephisodotos II., when he was forty years old, and to found on this the date of Praxiteles. Pausanias's vague
statement that Praxiteles was three generations later than Alkamenes gives no help, for it leaves a margin of
years greater than the number about which the controversy exists.
3 The statue in the Vatican (Helbig, Museums, 201), called Menander since Visconti's time, and supposed to be
an original work, cannot be taken into consideration. The name was based on one medallion in relief which had
already disappeared in Visconti's time ; moreover, the drawing of it (Icon. Grecque, PI. 6) is very unlike the head
of the statue. The dress is against the identification as Menander ; the chiton with sheirt sleeves had not come
into fashion in the fourth century ; it points to a personage of later date — a contemporary of the Poseidippos and
of the fine-seated statue in the Capitol (Helbig, Museums, 499). The head of the double term at Bonn (Friederichs-
Wolters, 1311) is again quite different from the Vatican head. It may be remarked incidentally that the other
term at Bonn (F.-W. 1310), the so-called ' Sophokles,' has nothing in common with the portrait of this poet.
The head united with the Euripides probably represents Aischylos.
4 Cf. Ldwy, /. G. B. 112 and III ; Kohler's remarks, ibid. 109, and C. I. A. ii. 1377.
0 Cf. B. Graf, in Rom. Mitth. 1889, p. 216.

SIO

PRAXITELES

this treatment found now and again. The next stage to the style of the drapery of
the Hermes is found in the Nike of Samothrace.
Against these facts and considerations, which all show the Hermes to belong
to the artist's later period, the arguments advanced by Brunn1 to prove that it
belongs to his youth can scarcely hold. For instance, we cannot admit that 'in
the Hermes the artist has not yet attained absolutely to that perfect security of
execution which can only be the result of long practice.' There is not in all
antiquity a work showing more subtle finish or more intimate mastery of all the

Fig. 131. — Satyr in Dresden.

secrets of marble technique than the Hermes. Brunn would scarcely have passed the
verdict he did had he seen the original at Olympia.
The main point is, however, that we still possess a series of copies after Praxiteles
of a style so different and so much older than the Hermes that we are compelled to
assign them to his earlier period.
Above all there is the Satyr pouring wine (Figs. 1 3 1 and 132)2 extant in so many
1 Deutsche Rundschau, viii. 1S8 sqq,
- List of replicas (Schreiber, Villa Ludovisi, p. 93) : the Ludovisi copy, a torso of the Capitol, a torso and a
head in the Vatican are published by Gh. Ghirardini in Bull, della Comm. Arch, di Roma, 1892, Taf. 11— 14 ;
the Ludovisi copy also in Brunn-Bruckmann, Denkm. No, 376. From Schreiber's list should be omitted g, Mus.
Torlonia 35 ; there should be added to it a head in Mus. Chiaramonti, 367 (made into a terminal bust, neck very

SATYR POURING WINE

3"

copies. It is true that it cannot be identified with absolute certainty with any one of
the Praxitelean Satyrs recorded by tradition. The much-vexed passage in Pausanias,1
where he speaks of the works of art in the street of the Tripods and of the cunning
by which Phryne contrived to make Praxiteles confess that he esteemed the Satyr
and the Eros his finest works, is so obscure, cither owing to the author's perverse
manner of writing or to the unsatisfactory condition of the text, that we are not
certain whether the boyish Satyr mentioned as acting as cupbearer was by Praxiteles
or not. Either interpretation of the passage causes difficulties, which can, it is true,
be easily solved by emending the text or by assuming the existence of a lacuna, but wc
can never hope to come to any incontrovertible conclusion. Our reason for believing
that this particular Satyr goes back to Praxiteles does not, however, rest on the passage
in Pausanias, but on other circumstances. An original of which so many copies exist

Fig. 132.— Head of Satyr in Dresden.

must have been one of the most famous statues in antiquity. Now Praxiteles was
famous for his Satyrs, and, even without entering into detailed analysis, it is obvious that
the work has a thoroughly Praxitelean grace and charm. The original must have been
of bronze ; the Satyr of Praxiteles, which Pliny mentions as the periboetos, or ' far-
famed,' was bronze ; this statue, and, as it seems, the one in the street of the Tripods
also, was grouped with a Dionysos and a third figure.2 Such a combination would
suit the 'Cup-bearer' very well, for it insures the presence of another figure for whom
thick, a ribbon hanging down each side, called ' Herakles '), and a head in the Gall, dei Candelabri, 45. Both are
inferior copies.
1 For the most recent^iscussions see Wolters, Arch. Zeitg. 1885, 82; Reisch, Weihgcschcnl-e, p. Ill ;
Gherardo Ghirardini, Bull, della Com m. Archeol. di Roma, 1892.
2 Pliny, xxxiv. 69 : Liberum patrem, Ebrietatcm nobilemque una Salyrum, quern Gracci pcribocton cognomi-
nant. Paus. i. 20, 2 : Sirupjs eirri irals Kal 8t5aitnz> eicirwfxa. "Epiara 5" ea-TTjKO-ra b^iov Kal Ai6vvtsov ®vtu.lXos
eiroiritrev. In the group mentioned by Pliny, perhaps Praxiteles repeated the same Satyr which he had placed in
the group which he made together with Thymilos. It appears to me quite certain that the figures named by Pau
sanias formed a group. The Dionysos with Eros, by Thymilos, may survive in the Naples group (Gerhard,
Neapels ant. Bildw. p. 30, No. 96), Vvhere Dionysos leans on Eros (the antique basis shows that the figures
belong together). The Eros had right supporting leg, and the Satyr on the other side of Dionysos had left
¦supporting leg. In the group mentioned by Pliny the young Satyr must have occupied the centre between
Dionysos and Methe (Ebrietas).

312 PRAXITELES
the cup is to be filled.1 Certainly the Satyr is not filling it for himself; no Satyr
would use so much ceremony on his own behalf. He is here fashioned after
the pattern of the mortal youths in attendance at feasts. Thus also the youthful
cup-bearer on the so-called funeral banquet reliefs of the fourth century appears
frequently in the motive of our statue.2
We may therefore assume, with that degree of certainty that can be attained in
matters of this kind, that this Satyr is a creation by Praxiteles. Yet how different is
this work to the Hermes !
The Satyr is still constructed on the traditions of the fifth century, and, with
regard to form at any rate, betrays a strong Peloponnesian influence. In fullest
contrast to the Hermes, the body viewed from the front presents as far as possible
quiet, vertical, unbroken planes. This in itself is a mark of the same tendency so
strongly imprinted on Polykleitan figures. The abdomen and the chest are still flat,
the nipples far apart, the front of the body seems to pass into the sides almost at a
right angle, as in the fifth century, while in the Hermes a wealth of soft undulations
insensibly carry the eye on, so that there is no sharp separation between sides and
front. Then the depression between the straight and the oblique abdominal muscles
is emphasized as in Polykleitos. The boyish figures by that artist, particularly the
'boy crowning himself and statues akin to it like the Pan and the ' Narkissos,' are
the immediate predecessors of this Satyr. The legs are in that attitude of ' arrested
motion ' which Polykleitos gave to his chief works. The head is turned and inclined
to the side of the supporting leg — the left ; the left forearm is stretched out hori
zontally, just as in that old Peloponnesian type whose influence we have traced down
throughout the whole fifth century. Moreover, the figure has the self-contained
bearing of the series of Argive statues extending from Hagelaidas to Polykleitos.
The Satyr has been compared to the 'oil-pourer' at Munich, 3 and has even been
taken as derived from it. The comparison is instructive, if only because it emphasizes
the difference between the two figures.4 The ' oil-pourer ' shows lines recklessly
broken up, an oblique position of the body, and a sturdy attitude with legs well
apart ; everything is so arranged as to give a true rendering of the intention and
the action, while in the Satyr the action is a matter of indifference. The sole
object of the artist is to produce a restful, pleasing design. In this too he
followed Polykleitos, who always paid most attention to that point.
To turn now to the head. The structure of the skull differs totally from that of
the Hermes, and closely recalls the Polykleitan manner. In place of the round Attic
skull of the Munich ' oil-pourer ' and of the Hermes, we find a close approach to the
square Argive type. The skull is also considerably deeper and higher behind the
ears than in the Hermes. The difference in the hair does not arise merely from
the difference of material ; it is a real difference of treatment. The hair still lies
close to the head, so as to show the shape. The form of the separate locks
(especially of the intertwining ends in front of and behind the ear), the symmetrical
parting in front, and the small pointed locks hanging over the forehead, are all
evidence of the influence exerted by the later Polykleitan works.
1 From the Berlin copy (Skulpt. 257) it has been assumed that the vessel in the original was a drinking-horn,
but the kylix of the Palermo copy is more probably correct. In the Ludovisi statue there is preserved the end,
not of a twisted horn, but of a twisted support which connected forearm and leg ; these sort of supports are
common in the age of Hadrian and the Antonines. It seems to me that in the original the left hand more probably
held a cup ; pouring from a height seems to me more natural when the liquid falls on a broad surface, rather
than into a narrow vessel.
2 e.g. Friederichs-Wolters, Gipsabg. Nos. 1058, 1059, 1063, 1066.
3 Ghirardini, loc. cit. 4 Thus rightly Michaelis, Ann. d. Inst. 1883, 140, note I.

EROS FROM THE PALATINE

313

Beyond these formal elements, however, the Argive influence does not extend.
The spirit and expression of the head, nay, even the formation of the forehead, which
though sparer in treatment yet shows the same shape as the Hermes, the round,
soft, tender grace and kindly cheerful
ness of the small mouth and cheeks, are
thoroughly Attic, and entirely in the
spirit of the Hermes. Marked as is the
resemblance of the body to the Pelopon
nesian prototypes, the deviations from
it are no less noticeable. The navel is
sunk deeper, and below it there is no
trace of the linea alba ; the division
between the abdomen and thighs is no
longer marked with the old conventional
hardness ;4he gluteus of the free leg has
no depression ; the flesh is softer and
fuller, and the best copies show that in
the original the delicate play of the
muscles under the firm flesh must have
been a special charm of the statue.
Hence it is clear that the Satyr is
of an earlier date than the Hermes, and
it is as a youthful work of Praxiteles
that we must consider it. We learn from
it that the work of the Polykleitan
school exercised a lasting influence on
the young artist — an influence we can
well understand with a youth so alive
to beauty and grace of form. He had
opportunities enough of studying the
work of Peloponnesian artists at Athens
(we still have an inscription from a work
by Naukydes on the Akropolis), and he
soon went himself to the Peloponnesos,
where he worked for Mantineia.1 The
date of the Satyr lies between 370 and
360 B.C.
To the same period belongs a
second work by the same artist like at
all points to the preceding. It is an
Eros, of which we have a good copy
in a torso from the Palatine, which has been restored and is now in the Louvre
(Fig. 133).2 That an Eros should be the nearest counterpart to the Satyr is a
1 The group of Apollo, Leto, and Artemis in Megara must have been one of the artist's early works.
Imitations of it on coins (Imhoof-Blumer and Gardner, Num. Comm. PI. A, x. ; FF, i. ii. ) seems to point to a
group of severe character in the style of the fifth century. The work might be by the elder Praxiteles.
2 Frohner, Notice, p. 311, No. 325 ; Brunn-Bruckmann, Denkm. No. 378. Fig. 133 leaves out the restored
head and left arm. Left forearm, right leg, left knee, and right foot are restored. The beautiful large
left wing is antique. A cast of the torso, unrestored, was formerly in the small museum of the Palatine, and
was there drawn by Eichler, 1876—77, for the Institute. Steinhauser's restorations are hideous. The antique
S S

Fig. 133. — Eros from the Palatine (Louvre).

3 H

PRAXITELES

confirmation of the Praxitelean origin of both, for Praxiteles is the very artist who
was as famous for his Erotes as for his Satyrs. This Eros torso is extraordinarily
like the Satyr both in the attitude of the body and in the system of forms. The
only difference is that the head is turned, not to the side of the supporting leg,
but somewhat to the right, thus sacrific
ing the tranquil absorption of the figure.
What the hands held is uncertain ; pos
sibly a wreath or a fillet, as attributes
of the god ; x at any rate the right arm
was held very high ; the tree-trunk with
the quiver and bow is merely an addi
tion by the copyist. The formation of
the body, though still on the same basis
as in the Satyr, goes further in the
direction of softness and roundness.
The abdomen is less flat, the soft layer
of flesh over the muscles is more clearly
indicated ; the depression above the
navel is deeper ; what remains of the
hair shows a fuller, freer treatment, from
which we may infer that the Eros is a
later work of the youth of Praxiteles.
In close connexion with both the
preceding creations are two others of
less importance, which are possibly only
the work of a contemporary imitator.
These two statues, both under life-size,
represent respectively Apollo2 and
Dionysos,3 and correspond in attitude
and bodily forms to the Satyr, except
that the right hand rests quietly on the
head. In the Apollo the head has been
preserved ; it agrees in pose and in
clination with the Satyr, and, like it,
bears witness to Polykleitan influence,
in having the parted hair growing closely to the skull and hanging down over the
forehead in symmetrical curls.
On the other hand, there is a second Eros, which must have been another creation
parts are also somewhat worked over in parts. For sketch of the antique parts and description, see Roscher's
Lexikon, i. 1360, 1361, Z. 18 seq. Milani (Mus. Ital. iii. p. 767) and R. Forster (Eros, Rede, 1893, p. 10)
agreed with me in referring it to Praxiteles. Weil, in Baumeister's Denkm. p. 1401, with fig. 1 55 1. There
is a replica in the museum at Parma (Arndt-Bruckmann, Einzelverk. No. 74).
1 Cf. Roscher's Lexikon, loc. cit. Milani (loc. cit.) thought he had found an imitation of the statue on an
Athenian coin : Beitle, Mo/in. d' Ath. p. 222 (= Brit. Mus. Catal. Attica, PI. 12, 3 ; p. 39). But the motive is not
the same. The figure on the coin represents a palaestric Eros (cf. Roscher's Lex. i. 1344, Z. 7), or another
palaestric Daemon, possibly Agon, according to Weil's hypothesis (Arch. Ztg. Bd. xxxiii. 164). The Eros of
Kallistratos, with which Milani identifies it, must have been quite different, and is, I think, reproduced in a small
bronze (cf. Jahrb. d. Ver. v. Alterthums-Fr. im Rheinland, Heft 90, p. 63, Taf. 3, 3).
2 In the Louvre. Frohner, Notice, No. 74; Clarac, PI. 269, 912; photo, edit. Giraudon, No. 1202. Poor.
The stem with the tripod is only a technical support ; the left forearm does not really lean on it. This work is
quite distinct from the later Praxitelean ' Resting Apollo.'
3 In Tarragona, Hubner, Bildiv. in Madrid, No. 672 ; Friederichs-Wolters, Gipsabg. I488.

FlG. 134. — Eros in Naples:

EROS OF CENTOCELLE

3'5

of the sculptor's youth, and which, to judge from the number of extant copies, must have
been far more celebrated. I know of seven copies of this Eros, of presumably the
same size as the original (about life-size), and four statuettes. The best-known
replica is the torso from Centocelle (head in Fig. 135) 1 in the Vatican ; the statue at
Naples (Fig. 1 34) 2 is better preserved, and the work, though hasty, is fresh, and probably
dates from the first century B.C. The main effect is faithfully reproduced, but little
attention is paid to such details as the hair. On the support near the left leg are
seen the remains of a bow, once held in the left hand. This accessory is more
correctly reproduced in a third copy at Turin.3 In it one end of the big bow rests on
the ground, and the left hand holds the upper end of the weapon. ' A tree-trunk

Fig. 135. — Head of Eros of Centocelle (Vatican).

which serves as support has a quiver hanging from it ; but it is evident that it was
only added by a copyist, for it has nothing to do with the design, and is differently
treated by the several copyists. In the original the bow stood clear of the statue.
A piece of the bow is also preserved in a replica in Russia.4
1 Helbig, Museums, 185 ; Friederichs-Wolters, 1578. Holes for wings on the back.
2 Clarac, 649, 1487 ; Roscher's Lexikon, i. 1359. Right arm, left forearm, and left leg restored ; wings
mostly antique. The front of the body is somewhat worked over. The head is much more lifelike than the
one in the Vatican.
3 Diitschke, Oberital. iv. No. 49. The statue is well preserved ; the head (except the ends of the curls on
the neck) is restored ; the basis is antique, so are both legs, though broken. The right foot is drawn back, and
the heel raised. The left arm and hand are antique, though broken ; the hand rests on the end of the bow. The
bow is carved in relief in a vertical position on the front of the trunk, the upper part of which is restored. The
right upper arm is antique, and separated from the, body.
4 Stephani, Die Antikensamml. zu Pawlowsk (Mini de V Acad, de St. Petersb. xviii. 4), p. 8, No. 6. From
Stephahi's description I gather that it is a genuine replica. The bow is introduced as it is in the Turin
statue ; it is important to note that, according to Stephani, the right arm and the open empty hand are antique •
also the large wings seem to be antique. I have already (Bull. d. Inst. 1877, 153) refuted Stephani's assertion
that the figure formed a group with Psyche. Overbeck shares Stephani's opinion (Plastik, 3rd ed. ii. 34).

316 PRAXITELES
The fifth replica, recently found in Rome, is only a torso (Terme Museum) ;x the
sixth has neither head nor arms ; 2 the seventh (Palazzo Conserv.) 3 is distinguished
from all the others through the absence of wings, and moreover it is by far the worst
and obviously quite the latest of the copies. In the right hand may be seen the
remains of some attribute, apparently a small torch, which, on the analogy of a small
replica, was held reversed upon an altar. The statuette,4 however, is of very late
and poor workmanship, and is also wingless.5 The question now arises whether the
torch reversed upon the altar — an emblem of funereal import 6 — together with the
winglessness which marks the figure as a kind of genius of death,7 are, on the ground
of those two poor and late replicas, to be ascribed to the original statue. The answer
must be in the negative. We merely have here one of those transformations, effected
by some trifling external additions, such as Greek originals not unfrequently under
went at the hands of later Roman artists ; besides, in one, or perhaps even in two
copies,8 the right hand still exists, and has no such attribute. Further, a decorative
stucco at Pompeii,9 which truthfully reproduces the main points of the statue, shows
the left hand supporting the large bow placed on the ground, and the right hand
hanging down empty. Of course there is no support for the left leg, any more than
there can have been in the original ; the wings, as in the Palatine Eros, are very large.
All the evidence tends to prove that the right hand was empty, yet in the original it
may not impossibly have held an arrow, which might easily be omitted by the copyist.
The original must go back to Praxiteles. I once thought that we only had late
adaptations of it,10 because I wrongly believed that all the copies should be restored
with the inverted torch on the altar, and because I followed the commonly received
opinion that the plait of hair worn by the Eros belonged to the Roman period. This
was a mistake, for practically the same fashion of hair occurs in a boy's head of the
fifth century — probably also an Eros.11 My next view was that the original of the
statue should be sought for in the Peloponnesian school of the fourth century,12
because of the intimate connexion between this statue and the later Polykleitan
works. But I have since found all these characteristics in the early work of
Praxiteles, and various considerations will eventually show that he was the artist.
Thus the old conjecture of E. 0. Visconti seems likely to prove true.

wings

The torso has suffered by water ; the ends of the curls are preserved; large holes, in the back for the
2 Rome, Coll. Monteverde (Arndt-Bruckmann, Einzelverk. No. 141).
3 Discussed by me in Bull, dell' Inst. 1877, 151 seq. Published in Bull, delle Comm. Arch, di Roma, 1877,
Tav. 16 ; 17, 1. Cf. Helbig, Museums, 569. The head is wrongly put on.
4 Vatican, Gall, dei Candel. 203 ; Gerhard, Ant. Bildw. Taf. 93, 2 ; Helbig, Museums, 393. For the
restorations, cf. Bull. d. Inst. 1877, 152, No. 2.
» The three other statuettes are : Berlin, 139. Torso with wings. -Copenhagen, Jacobsen collection, 1051,
headless, bronze wings fastened on by two cramps. Quiver hangs on the trunk, on the top of which are
traces of the attachment of the bow as on the Naples replica. Remains of a spiral puntello on the right thigh
to join it to the wrist. -Turin. Head only. Fair work. » Cf. Bull. d. Inst. 1877, 154 seq.
Helbig (Bull. d. Inst. 1885, 71) and Fiihrer (No. 183) called it Thanatos. [His view is now modified;
cf. Museums, 185.— E. S.]
s On the copy of Pawlowsk (p. 315, note 4), and perhaps on the one at Turin (p. 315, note 3)
Mus. Borbonico, ii 53 . Brulloff, Thermes de Pompeii, PI. 4. In Bull. d. Inst. 1877, 160, I pointed out
the differences between these two publications. Brulloff gives the right hand empty, in Mus. Borb. it holds an
arrow. The original is in Pompeii, on the wall of the Tepidarium of the Thermae of Fortuna. It is clear that
the hand was empty, although only the thumb and forefinger remain. The stucco relief corresponds with the
statue in at itude turn of head etc., but the head is not bent, and the plaits of hair are not exactly reproduced.
• ¦ t f u V\ ' Where SU'eSS is yet Iaid on the fact that an OTiginal of the best period is its
clpTab. pj™ '"') aSSIgnS th£ m°tiVe '° 'he R°man ep0Ch' W°UerS d0eS this sti11 more definitely in
» Archiiol. Studien H. Brunn dargebr. 1893, P- 88 seq. Taf. 3. 1. Roscher's Lex. i. 1362.

THESPIAN EROS 317
It is the obviously great reputation of the work which tells most in favour of
Praxiteles, the most celebrated pourtrayist of Eros. In the second place, the bodily
forms present great similarity to the Satyr and the Palatine Eros, though the age
represented is certainly somewhat more advanced, since the muscles are more power
fully developed.
The motive, like that of the Satyr, is connected with the series of Argive types.
The attitude of the body and the turn and bend of the head agree with the Polykleitan
' boy crowning himself and the Ildefonso figure ; the motion of the arms, however, is
different, and the upper part of the body is slightly twisted, the right shoulder being
slightly thrust forward. Even this attitude, though reversed, is found in the Poly
kleitan 'Dresden boy' (Plate XII.), who is almost identical^with our Eros; even the
arms correspond — one hanging down beside the body, the other being outstretched ;
and if the Eros held, as we conjectured, an arrow in his right hand, the analogy
with the Polykleitan boy, who also held something in the hand towards which the head
is turned, becomes so much the closer.
It is evident that the youthful Praxiteles had fallen under the charm of those Poly
kleitan figures, and his Eros shows an admirable adaptation of their chief characteristics.
For Praxiteles, as for Polykleitos, the rhythmic grace of the whole motive depends
very largely on the bend of the head. But this bend of the head has also been used
by both sculptors for the purposes of characterization : in one case it serves to
emphasize the modesty of the boy victor ; in the other, a slight change of attitude
has sufficed to produce a very different effect : it is the demoniac nature of Eros
which is expressed in that bent head, in that face peering up from amid its profusion
of locks.1 This is no longer the joyous, innocent Eros with wreath and fillet, as we can
imagine him to have been represented in the Palatine torso : he has become the
bewitching daemon, the captivating, irresistible god, airavTrnv Baip,6vwv vTrepTaTos,
glorified in the poetry of Euripides,2 who gave him as symbols of his all-powerful
sway the bow and arrow 3 which he carries here. We can now understand why
Praxiteles gave to this Eros more developed forms than to the other.
The god holds his bow (and arrow ?) ready for immediate use, yet the artist
only intends them as an outward symbol, for he has essayed the difficult task of
giving visible expression to the inward might of the god : he does not allow him to
use his weapons, his Eros is irresistible through his upward glance and bewitching
charm. . . . We have thus unconsciously arrived at a presentment of the god which
agrees strictly with a well-known ancient epigram on the Thespian Eros of Praxiteles,
to the effect that he excited the transports of love by hurling not darts but glances,
(pLArpa Be fiaWco ovk eV olaTevcov, aKX' aTevL%6p,evos.i Further, since this statue was,
as we saw, an early work of Praxiteles, made at the same time as the Satyr ; and
1 That the bent head means grief is a modern misunderstanding of Friederichs (Bausteine, 448). His earlier
opinion (expressed in his exquisite monograph on Praxiteles, p. 22), that the head is bent in a dreamy mood, and
is characteristic of Praxiteles, is more correct.
2 Cf. Roscher's Lex. i. 1348. 3 Cf. ibid. 1348, and Arch. Anz. 1890, p. 89.
4 For the epigram, cf. Benndorf, in Btill. della Comm. Arch, di Roma, 1886, p. 69 seq., who shows that
arevi&pevos is in the middle voice. His proofs seem to me absolutely convincing, and Robert's arguments (Arch.
Mdrchen, p. 167) are powerless to refute them. By the basis in the theatre, however, on which, according to
Athenaeus, the epigram was inscribed, I incline to understand, not as Benndorf does, the original basis in Thespiai
(for Athenaeus does not mention this city at all), but a basis of later date supporting a replica in Athens. Near
the theatre was found a torso of an Aphrodite which appears to be an admirable replica of the Aphrodite in
Thespiai (cf. supra, p. 319, 1,. 1) I have already shown (p. 271, note) that Benndorf is wrong in referring to the
Eros of Thespiai a statue of Ares in the Capitoline Museum.

3 IS PRAXITELES
since in the story of Phryne's stratagem the Thespian Eros actually figures in
company with a famous Satyr ; and, finally, since a statue preserved in so man}'
copies must have been one of the sculptor's masterpieces — there can be hardly any
doubt that we arc right in identifying it with the Thespian statue.
What dates we have of Phryne's life agree very well with the conjecture that the
Eros which she dedicated at Thespiai was one of the works of the earlier period of
Praxiteles, between 370 and 360 B.C. The only reliable chronological information we
have about her, beyond her connexion with the artist, is that she inspired the ' Anady
omene ' of Apelles, and there is nothing to prevent our placing this about 350.1
When Phryne offered to rebuild Thebes in 335 B.C., after its destruction by Alexander,
she may very well have been advanced in years, for this was an enterprise for which
only money was needed. The lawsuit, in which she was defended by Hyperides,
cannot be closely dated.2 Writers of the Middle Comedy speak of her as a
contemporary,3 while in the New Comedy she is alluded to as a hetaira of the
previous generation.4 It has been plausibly suggested that it was not till after the
destruction of her home, Thespiai, by the Thebans that she came to Athens, at a time
when her countrymen were turning to Athens for succour (probably B.C. 373 — 372).5
At any rate, in order to be known as a Thespian, she must have been born before
that date, and need no longer have been a child" at the time. Her connexion with
Praxiteles, which anyhow belongs most naturally to the artist's youth, would take
place in the period immediately following. The temple at Thespiai was of course
spared by the Thebans," and Phryne, in recognition of her successes in Athens, sent a
handsome votive offering to the god of her native town, though its walls were razed
and the town itself had passed into the power of Thebes.s The Thebans, however,
would certainly not interfere with private piety or dedications in the temple, their
only object being to destroy the political existence of the community.
Phryne's offering apparently consisted of three statues, for beside the Eros stood
a statue of herself, and one of Aphrodite, both by the hand of Praxiteles : the Eros
probably stood between the two,9 and the group was analogous to that of the Satyr
between Dionysos and Methe (p. 311). This presentation of her own portrait by
the hetaira was simply in accordance with an old custom. We are expressly told that10
her gilded portrait at Delphi, also by Praxiteles, was dedicated by the hetaira herself.
The material of this statue and the place where it was set up mark an advance
1 Owing to his relations with Philip, Apelles must have been active at this time.
- Sal. Reinach points out (Rev. Crit. 1894, p. no) that Hyperides was very little older than Phryne was
according to our assumption. I see in this no reason against the dating proposed above.
3 Timokles, apud Athen. p. 567 c In Aristoph. Eccles. IIOI, Phryne is only a name given to any
courtezan. 4 Poseidippos, apud Athen. p. 591 e.
5 Xenoph. Hell. 6, 3, 1. Cf. Von Stern, Gesch. d. spartan, u. theban. Hegemonie, p. 119.
'' As Sal. Reinach (Gaz. Archiol. 1887, 283) assumes, by placing her birth about B.C. 375. In 335 she may
just as well have been fifty as forty. Overbeck (AVA. Ber. 1893, p. 40 seq.) places Phryne's birth (in connexion
with his wrong late dating of Praxiteles) after B.C. 372, assuming that Thespiai was recolonized again imme
diately, a view that cannot be reconciled with tradition. (Cf. infra, note 8.)
7 In the similar case of Plataia, this is expressly staled. Paus. ix. 1, 8.
8 As Thespiai seems, like Plataia, not to have been rebuilt till after Chaironaia — for that this had not yet
taken place in 343 is proved by Demosthenes (F. L. m, 325)— Phryne's votive gift must fall in the time when the
walls were in ruins and the citizens scattered. For to place her gift after the rebuilding would contradict all we
know for certain about Praxiteles. And why should Phryne have delayed her gift to her ancestral god till the
Thespians had rebuilt these walls and were again politically independent ?
9 When Alkiphron (Overbeck, S. Q. 1271) makes Phryne say that she stood in the middle between Eros and
Aphrodite, he is evidently merely indulging in a rhetorical period.
10 Pans. x. 15, 1. The statements that the irepiKrioves (Athenaeus) or the "EXXyves (Aelian) had dedicated
the statue are in each case due to misunderstanding.

APHRODITE 3T9
upon the Thespian offering, so that it must have been the later work. We now have
to consider whether there exist any copies of these statues.
The well-known 'Venus d'Arles ' (Fig. 136)1 is certainly to be referred to
Praxiteles. This is mainly proved by the extraordinary likeness of the head to the
authentic copies of the Knidian Aphrodite {infra, p. 322). The arrangement of the
hair with its double fillet, the forehead, the contour of the face — and we may
add the bracelet on the left upper arm — all- correspond perfectly. It is only in
details that the forms of the head are harder and more constrained ; the eye is
less deeply set, the eyelids are rendered in the old manner, and the drawing of the
mouth is more severe.
An examination of the body confirms the impression that the Venus of Aries
belongs to the early period of Praxiteles. The attitude is very like that of the
Eirene and of the ' Satyr pouring out wine ' : the left leg bears the weight of the body,
the left forearm is outstretched bearing some object towards which the head is turned
and inclined, and the right arm is raised. As in the Eirene the right foot is not drawn
behind the other, but only placed to the side in Pheidian fashion with the heel off the
ground. One of the muses on the Mantineian relief — the one playing the flute — is
also closely related in the whole rhythm of her attitude. The simple broad treatment
of the drapery corresponds to that of the Eirene and to the relief just mentioned,
which is a work of Praxiteles's youth. Specially characteristic is the treatment of the
nude, which so far as youth and maiden can be compared is closely related to that of
the Satyr ; for the abdomen is still flat, and the whole front of the body is still treated
in great broad planes, ending in a harsh line at the side ; the thorax is broad, and the
breasts, very youthful in form, are far apart ; 2 the undulation of the hip above the
supporting leg is as yet only slightly indicated.
There can be little doubt as to the motive of the Venus of Aries : the left hand
held a mirror ; the right was raised towards the head. An engraved gem 3 of the early
fourth century shows a woman, doubtless an Aphrodite, with a mantle wrapped in
the same way round the lower part of the body and a mirror in her left hand. The
terra-cottas of that period frequently reproduce the same design.4 The partial nudity
of the figure is explained by the fact that the goddess is at her toilet.
The statue belongs to the period of the Satyr and the Thespian Eros ; therefore
it probably represents the Aphrodite which stood beside the latter at Thespiai.
A statue in the British Museum from Ostia (Fig. 137)5 is in close connexion with
the one just described ; it is a variant of the same motive, with the sides reversed. The
head is again so like the Venus of Knidos that we must refer this work also to
1 Frohner, Notice, No. 137. ' Thasian ' marble. Right arm and left hand new. Drapery worked over head
well preserved (cf. supra, vignette on title-page). Cf. Bernouilli, Aphrodite, p. 182 ; Brunn-Bruckmann, Dcitkmdter,
No. 296. An excellent replica (torso) was found in Athens near the theatre (Friederichs-Wolters, 1456). Cf'
Roscher's Lex. i. 415, Z. 30. . Brunn-Bruckmann, Denkm. No. 300 A. The workmanship of this torso (in the
finest Parian Lychnites) is so wonderful that it is impossible to resist the notion that it is a replica from the studio or
Praxiteles himself or of one of his sons. (For a replica of the Thespian Eros in the theatre at Athens, cf. supra
p. 317, n. 4.)
2 The breasts are much more lifelike in the Athens torso than in the Paris statue. The Athens torso gives a
much better idea of the original.
3 In Berlin, 4631 ; scaraboid from Sparta; judging from style certainly not later than the first half of tlie
fourth century ; even the attitude of rest on the left leg corresponds ; the right hand moves towards the head, but
does not hold anything. The hair is bound with three fillets like the statue. Broad chest.
4 Cf. e.g. Dumont-Chaplain, Ciramiques, ii. PI. 28, 1 ; here the raised right hand holds a band for the hair
The style shows that this figure must be dated in the first half of the fourth century. The attitude of rest on the
left leg and the wide flat chest are very like the Aries statue.
6 Friederichs-Wolters, 1455. The head is unbroken ; the left arm is restored, but it was raised ; the richt
hand is new. Frohner, Notice, p. 180, calls the statue an ' imitation ' of the Aries statue,

PRAXITELES

Praxiteles. But the original must have been a later work than the Venus of Aries.
The attitude and action are less constrained, the free leg is drawn more back, and,
above all, the head, which is relatively smaller, is turned towards the side of the free
leg, and raised, thus destroying all the repose and tranquillity of the earlier conception.
The drapery is richer and more agitated, and falls down lower on the one side ; the

Fig. t3$. — 'Venus d' Aries' (Louvre).
bodily forms are rounder and closely related to the Knidian statue, the chest is no
longer so broad and flat, the breasts are closer together, the navel is sunk deeper,
and the lower abdominal line is treated precisely as in the Knidian statue.
The motive also was certainly identical with that of the Venus of Aries ; the left
hand held up the mirror, and the right may have held some other toilet requisite. The
whole conception lacks the dignity and repose of the other statue ; this maiden has a
questioning, self-conscious look ; rejoicing in her beauty she raises the mirror — and
cares not if her mantle slips down a little lower.
It seems to me that a Phryne by Praxiteles must have looked just like this — ideally
beautiful and noble, yet different enough from a goddess. The statue of Phryne so

PHRYNE 321
famous in antiquity was the one at Delphi; as we saw, it was later than the
Thespian offering. It is this statue that I should like to imagine as the original of
the ' Townley Venus.' Of both the portraits, also, between which she was placed —

Fig. 137.— The 'Townley Venus' (Brit. Mus.)
King Archidamos of Sparta and Philip 01 Macedon — copies seem to exist, found in
that villa at Herculaneum whose owner was so devout an admirer of Greek philosophy.
1 An Archidamos is preserved, as Wolters (Rom. Mitth. 1888, 113 seq.) has shown, in a terminal bust of the
Herculaneum Villa. Wolters remembers only the two Archidamos statues of Olympia, and not the Delphi
portrait, which is much more renowned ; he concludes from the style that the terminal bust represents
Archidamos III., but admits the similarity of the portrait to that of Euripides. The resemblance is indeed so
strong that we may assume the same artist for both portraits. The stylistic treatment of the hair in the Archi
damos points to the period of the Peloponnesian war, therefore the statue presumably represents the Archidamos
by whose name the first period of the war was designated. It is not impossible that Demetrios, who belonged to
T T

322 PRAXITELES
After the Cynic Krates had mocked at this juxtaposition of a courtesan and a king,
later ages cited it as a typical instance of Greek licence.1
The Aphrodite of Knidos must, from the preceding considerations, be assigned
to the middle period of the career of Praxiteles.2 This conjecture is strengthened when
we remember that works executed for three towns so near each other, and so far from
home as Knidos, Kos, and Halikarnassos, must needs be of about the same date. The
Mausoleum was begun in 353 B.C., so that the statue for Knidos should be dated about
350 B.C. It has been preserved in a series of copies presumably of the original size,
and also in statuettes. All these doubtless reflect the style and design of the
statue, but can give us no idea of its subtler qualities. We must be careful,
however, not to reckon among the copies, as has been done hitherto, works which
are really later independent adaptations, such as the Munich statue and the little
head at Olympia.3
Praxiteles has represented the goddess disrobed for the bath, in the act of laying
aside her last covering. She turns her head quickly round in the direction of the
free leg,4 an attitude only comprehensible if we imagine her as looking about her to
make sure her privacy is not threatened. This also explains the instinctive movement
of the right hand. The dominating note of the expression is its absolute freedom
from any stronger emotion — from any yearning or languishing. It was only in
later adaptations that voluptuous suggestions 5 were introduced, but the goddess of
Praxiteles produced her effect by the purity and innocence of her expression, by
her simple grace and noble naturalness. As regard bodily forms, the statue stands
about midway between the older type of the ' Venus of Aries ' and the later Medicean
Venus. Much as she differs from the former, she yet also differs from the latter,
especially in the breadth of the chest and in the modelling of the contours, which are
far less full and rounded.
In addition to the nude Aphrodite of Knidos, an entirely draped Aphrodite by
Praxiteles — probably the one described as velata specie at Kos — is known to us in a
poor and late reproduction, which however has the advantage of being attested by docu-
that period, might be the artist (cf. p. 122, n. 2). The Archidamos was found in the Villa Ercolanense in the great
peristyle as a companion piece to a term representing a beardless man, who, judging from style and type, might
very well be Philip of Macedon, who stood beside the Archidamos at Delphi. Gercke (Bonner Studien, p. 141)
suggests Philhetairos, instead of Philip, but the lower part of the face, which deviates from the fixed type of
the coins, speaks against this assumption.
1 Trjs ra>v 'EXXt\viiiv aKpaatas avd6n/j.a, Krates, apud Athen. p. 591 b.
2 Our dating of Phryne does not clash with this in the least, as Sal. Reinach asserts (Rev. Crit. 1894. no),
for of course no reliance can be placed on the anecdote that Phryne sat as a model to the sculptor.
3 As Michaelis has done in his last treatise on the statue (/. H. S. 1887, 324 seq.) Cf. my remarks in Arch.
Anz. 1891, p. 140 seq. Michaelis's list of copies is in need of revision. The following are true copies of the Venus
of Knidos, and correspond in their measurements : (1) = Mich. A, in the ' Magazino' of the Vatican; bronze
cast in the Louvre (cf. Arch. Anz., loc. cit.); head unbroken ; best preserved copy; workmanship good.— (2) = Mich.
D; Helbig, Museums, 316; head wrongly adjusted; modern neck. Poor, leathery work.— (3) and (4) Museo
Torlonia, No. 106, 146 (but not No. 26, which Mich, reckons with the others) ; both much restored.— (5) = Mich. C ;
Pal. Pitti ; body good ; head wrongly adjusted, but genuine.— (6) Villa Ludovisi, Helbig, Museums, 869 ; head
readjusted according to the line of breakage ; fair copy, which passes incorrectly as modern.— (7) Torso in
Kaufmann Coll. (Berlin) from Tralles ; the head (Ant. Denkm. i. 41) is the best copy of the head.— (8) Torso,
Louvre, No. 2885. — (9) Torso in Villa Medici, wrongly restored.— (10) Bust in Louvre = Frohner, 164; neck
unbroken.— (11) Head in Louvre, wrongly given to a reclining nymph (Frohner, 454).— (12) Head in Museo
Chiaramonti, No. 254 ; fair work, but much rubbed over.— (13) Head in Capitol, Gal. No. 39.— (14) Head, Mus.
Boncompagni-Ludovisi, No. 21.— Among the statuettes there is one very good and well-preserved copy in the
Museo Chiaramonti, No. 119 c ; an inferior one in Potsdam in the picture gallery. A bronze not quite exact in
detail, Frohner, Coll. Greau, Bronzes, PI. 26, No. 927.
4 The turn has been kept especially well in 1, 6, 7, 10, II.
5 Especially in the Munich statue, where the thighs are pressed together.

KOAN APHRODITE

323

mentary evidence. It is a statue in the Louvre (Fig. 138),1 round whose plinth runs an
ancient inscription, LLpa£«Te\»7? eirolr}o-ev, dating from late Imperial times. Until a
short time ago this inscription was only known from a notice in Visconti, and had
disappeared because the ancient plinth had been let in or
sunk in such a manner that the inscription was supposed
to be lost.2 Lately the edge of the plinth has been set
free, and the inscription is again legible.3 According to
every analogy, the inscription can only mean that the
work is a copy after Praxiteles, and the style certainly
confirms this view. The goddess wears a long thin
chiton, girded deep about the hips ; the right shoulder is
lowered, allowing the garment to slip slightly off, but not
so as to expose the bosom. The attitude is as simple
as that of the Knidian statue ; the weight of the
body is thrown on to the right leg, the left shoulder is
much raised, the right hip curved outwards. In the
copy the left hand rests on the head of a little Eros
obviously of late Roman origin, which cannot possibly have
belonged to the original ; it doubtless replaces some acces
sory which was not to the liking of the copyist ; but
if the hand be imagined to have rested on a terminal
figure or an idol, we obtain a motive especially popular in
the time of Praxiteles. The head is unfortunately missing.4
The main charm of this statue must have resided in the natural treatment of the
thin semi-transparent drapery. The copy at any rate still allows this to be felt, and
shows how differently the artist solved the problem which Alkamenes had essayed
before him {Venus Genetrix, p. 82). Those clinging wet draperies, those beautifully
arranged if conventional folds of the earlier artist, are quite abandoned in the Praxitelean
period, which condemned them as untrue ; thus here also Praxiteles, in the multitude of
fine flowing folds, is striving entirely after truth of effect. The lines of the folds are
not beautiful in themselves, as they were in the older system, but they are charming
(or were so in the original), because they reflect the rich abundance of nature. The same
drapery, treated in the same way, as well as the deep girdle and the slight displacement
of the drapery on one shoulder, is seen in another Praxitelean work, in the chiton of
the Artemis Brauronia, a statue which I agree with Studniczka in thinking we have
a copy of in the ' Diane de Gabies ' (Louvre).5 Further, many a work of later times
(like the celebrated Flora Farnese) 6 was evidently inspired by the Aphrodite of Kos.
According to Studniczka's very probable suggestion, the Brauronia of Praxiteles
was set up on the Akropolis, in B.C. 346. It therefore belongs to the later period
of his artistic activity. He must evidently have returned from Asia Minor about

Fig. 138. — Statue in Louvre. (From
Clarac, Mus. de Sc. PI. 341.)

1 Published Bouillon, t. iii. Statues, PI. 6, 7. Clarac, Musie, PI. 341, 1291 (= our Fig. 138, which omits
the restored head). Cf. Bernouilli, Aphrodite, p. in, No. 2.
2 Cf. Frohner, Notice, 151 ; Ldwy, /. G. B. 502.
3 The epigraphy is of the later Imperial period. Sal. Reinach ( Gaz. Arch. 1887, 259, note 5) called attention
to the inscription. The statue is not, as R. supposes, a replica of the ' Genetrix.'
* Head and right shoulder are modern, not ancient and readjusted as Clarac and Frohner assert.
5 Studniczka, Vermut. z. Kunstgesch. p. 18 seq. Robert (Arch. Mdrchen, p. 144 seq.) has adduced nothing
decisive against this supposition ; cf. p. 102, note 3. To the known replicas of the statue should be added a
head in Berlin, Skulpt. 625. The Praxitelean Dionysos, in the Hope Coll. at Deepdene (Surrey), is analogous to
this Artemis in the drapery (Clarac, 695, 1614; Roscher's Lex. i. 1 133).
6 Clarac, 438 B, 795 D ; cf. 437, 792 (Munich Glyptothek, 298).

324

PRAXITELES

Fig. 139. — Artemis in Dresden.

this time. Shortly before this
he had brought to a close his
work in Ephesos, Halikarnas-
sos, Knidos, and Kos.
Belonging apparently to
the earlier part of his career
is another statue of Artemis,
of which the best copy is the
one at Dresden1 (Figs. 139,
140). It represents the god
dess as a quite young, still
undeveloped girl, wearing a
long ungirt tunic, and stand
ing in a simple, tranquil at
titude, her bow in her left
hand, while the right feels for
the quiver at her back. The
Praxitelean authorship is
proved by the type of the
head, which corresponds in
contour and arrangement of
hair with the Venus of Aries
and the Knidian Aphrodite,
and merely introduces into
the type a maidenly, almost
childish element. It is above
all the attitude and drapery
of the Dresden statue which
clearly prove its early origin.
The attitude, with the un
weighted .foot placed to the
side, is that of the Eirene and
of the Venus of Aries. Like the
Eirene, too, the drapery ¦' falls
in straight folds from the
1 Dresden, Becker, Augusteum, 45 ;
Clarac, 569, 1214 A; Miiller- Wieseler,
ii. 162 ; Roscher's Lex. i. 606, No. 2.
Excellent preservation. Faithful but
rather coarse copyist's work. P. Herr
mann and P. Arndt have recognized the
Praxitelean character (cf. Arndt-Bruck-
mann, Einzelverk. to No. 133). — Three
replicas in Cassel, heads not authentic ;
in one (ii. 19) a piece of the raised right
arm is antique. — A good replica in Ber
lin, Skulpt. 60 (remains of the raised
right arm antique). — Good statue in
Villa Borghese (Arndt-Bruckmann,
Einzelverk. No. 133 ; head foreign).
An Ionic under-chiton is added. The
drapery is finer and better executed than
in the Dresden statue.

ARTEMIS

325

right knee, but whilst the Eirene seeks to imitate the large simple folds characteristic
of Pheidian drapery, the Munich Artemis shows a wealth of charming detail care
fully studied from nature. There is an astonishing similarity, except that the sides
are reversed, between the drapery falling from the knee and the corresponding portion
on one of the Muses, leaning on her right arm, of the Mantineia relief.
This statue was afterwards slightly altered and turned into an Isis-Tyche, by
lowering the right arm, and by doing away with the quiver, which was no longer
needed to explain the action of the hand, though the quiver-strap was retained

Fig. 140. — Head of Artemis (Dresden).

because it gave a special charm to the drapery ; the bow in the left hand was replaced
by a cornucopia, and an Ionian under-chiton, visible at the shoulders, was sometimes
added. The best copy of this variant is at Munich.1 Perhaps the basis of this con
ception was a Tyche by Praxiteles himself, with features borrowed from his Artemis.
1 Glypt. 113. Brunn-Bruckmann, Denkm. No. 123. The ends of curls on the shoulders are antique.
Thorwaldsen rightly restored a horn of plenty in the left hand. The replica noted by Winter in the Pal. Sciarra
(Berlin, Skulpt. p. 529) gives the true interpretation. The goddess holds a horn of plenty and a snake ; the child
Harpokrates below. Replicas, known as such by the lowered right arm, the Ionic under-garment, and the
absence of quiver, are : Clarac, 410 H, 837 A (with horn) ; Museo Torlonia, Taf. 64 and 251 (wrongly restored
in a group) ; Louvre, 1910 (Frohner, Notice, 390), etc. The type referred to on p. 60, note 1, is related to
this one.

326 PRAXITELES
The later Orphic identification of Tyche and Artemis1 may be derived from an
earlier, presumably Eleusinian conception.
The same type was adopted, actually by Praxiteles, for an Athena.2 A small
transverse aegis now takes the place of the quiver-strap ; the right arm is raised and
rests upon a lance, the left upon a shield. The goddess is conceived as a graceful
girl, bright and gentle, but not quivering with excitement, like the youthful Athena
attributed to Skopas {supra, p. 305). A head in Berlin,3 belonging to a large statue
now lost, bears so marked a likeness to the Knidian Aphrodite that its original must
have been a Praxitelean work.
We must not pass over a remarkable statuette of Artemis from Kition in Cyprus
(Fig. 141 ).4 Here also the goddess is conceived as a graceful but quite young girl
in a long chiton. She is leaning against an idol.5 The folds of drapery round the
lower part of the body are quite in the style of the Dresden Artemis. The head
likewise resembles that of the Dresden statue, save that the knot of hair is rather
different. But this Cypriote statuette has the freshness and bloom of a work which,
however slight, is yet an original. It is so thoroughly Praxitelean that we would
gladly call it a irapepyov by the master himself, made perhaps when he was in the
neighbourhood of Cyprus, at Kos or at Knidos.
In addition to what has been described, a good deal of Praxitelean work might
certainly be recovered from the statues and heads of other female divinities.6 But
this master's creations have given rise to such countless variants and adaptations
that it is difficult to pick out what was originally his. Such an inquiry would lead
us too far from our immediate object, which is to trace the main points in the
successive developments of the artist's style. One example however of what is meant
may be given. Praxiteles certainly also pourtrayed older, more dignified goddesses,
such as Leto for example. Some such type it is that underlies the famous so-called
' Hera Ludovisi.' 7 The arrangement of the front hair, the forehead, and also the
lower part of the face clearly show Praxitelean forms. In spite of this, the work is
only a Roman creation based on a Praxitelean model ; for we must not attempt to
deceive ourselves ; this famous ' Hera,' which passes for a type of ideal beauty, is
in reality only a Roman empress or a lady of the Imperial family, represented as a
goddess, as a Ceres or Fortuna. There is external evidence to prove this. The
arrangement of the hair at the back, with the typical plait on the nape, is never given
to goddesses. It is the special fashion of the Claudian age, and, so far as we know, it
occurs at no other period, either before or after.8 Further, the knotted fillet that
1 Orph. Hymn. 72.
2 Statue of the Villa Borghese, Helbig, Museums, 935 ; Berichte d. Sachs. Gesell. 1861, Taf. I, 2 ; 1865, Taf.
I, 2, 3. The head is foreign. The right arm was certainly raised. The drapery corresponds in all essentials
with the Artemis ; on the shoulders an Ionic chiton is visible. Poor work. Replica in Turin, staircase to the
Armeria ; right arm raised ; the neck new, but the head seems genuine : Corinthian helmet, hair simply taken
back; general Praxitelean character. — Museo Chiaramonti, No. 403 ; right arm raised, head foreign. In the
later statues of Athena, all sorts of Praxitelean influences may generally be distinguished.
3 Berlin, Skulpt. No. 78.— A head in the Vatican (Helbig, Museums, 229 ; Mus. Pio Clem. vi. 2, 2) is
similar, but differs in the proportions and in the turn of the head.
4 In Vienna, Jahrb. d. Kuusts. d. Oesterr. Kaiserh. v. 1887, Taf. i. ii. ; p. 1 sqq. (v. Schneider). Ohnefalsch-
Richter, Kypros, the Bible and Homer, PI. 203, 5 ; p. 318 sqq.
5 Cf. the Dionysos at Deepdene mentioned p. 323, note 5.
6 Note especially a Kore statue (head preserved) in Vienna, quite lately publ. by Rob. v. Schneider (Jahrb.
d. Kunsts. d. Oesterr. Kaiserh. )
7 Schreiber, Villa Ludovisi, 104; Friederichs-Wolters, 1272; Helbig, Museums, 872; Roscher's Lex. i.
2122 (Vogel) ; Brunn, Gdtterideale, p. 9.
8 Cf. Bernouilli, Rdm. Ikonogr. ii. 1, 181, 193, 182, Fig. 27, 190, Fig. 30. Taf. 14, 15, 21. The elder
Agrippina has on her coins just such stiff locks as the 'Hera.' In the Ludovisi inventory for the year 1633

Fig. 141.— Statuette of Artemis from Kition in Cyprus (Vienna).

328 PRAXITELES
passes in front of the diadem and falls down at the sides is certainly not the attribute
of any goddess, but it is characteristic of the Imperial ladies during the early Empire,
and marked them as priestesses,1 and it is precisely in authentic heads of the Claudian
age that we find this same combination of diadem and sacerdotal fillet.2 There are
various degrees of idealization in these portraits of deified women. In the Hera
Ludovisi it is strongly emphasized, and so is the imitation of Praxitelean style. But
a careful copy from any Greek original the ' Hera ' is not,3 and the additions in
Roman taste are clear enough to be unmistakable to us, though a former school of art
criticism might fail to distinguish them.
But to return to the statues of Artemis. If we assign the Diana of Gabii to
Praxiteles, then the ' Diane de Versailles ' cannot be very far removed from
him. For the types of head are nearly related. Above all, there is nothing in
this celebrated statue which could not belong to the master's later period. Judging
from coins, his Artemis at Antikyra 4 must have been very similar. The ' Diane ' is
certainly not, as was formerly believed, a creation of the Hellenistic age. No such
invention remained for that age, for all these types of divinities had been created
long before. Even the Belvedere Apollo has lately been shown to be a fourth-
century creation.5 The Diana may well be derived from a work by the master of
the Apollo — a point to which we shall have to return.
Still more closely related to Praxiteles and to his Artemis of Antikyra is a torso
in the Jacobsen Collection.6 It is very like the Diana of Versailles, though it is
simpler, with less of a gliding gait and a more decided stride ; the garment is open at
the side and clings closely to the figure, like the drapery of the Aphrodite of Kos.
Another Artemis type also belongs to the later period of Praxiteles, and is the
creation of the master himself. It shows the goddess arrayed as in the Versailles
statue, with a scarf twisted round the short chiton ; in an easy unstudied pose
Artemis is leaning her left elbow on a tree-trunk, while her right arm rests on her
hip. A picturesque negligence of dress and of attitude and a certain genial natural-
the head is called ' Giulia ' ; the headdress, characteristic of a portrait-head, had evidently been noticed at
that time. Since Winckelmann, the interpretation as Hera has been accepted, and sanctioned by Goethe, Schiller,
and Humboldt. Only Conze (Familie d. Augustus, p. 15) doubted the Hera interpretation, and suggested Venus
Regina. From the work on the diadem he ascribes the head to the early Imperial period.
1 Cf. e.g. the Livia from Pompeii, Bernouilli, loc. cit. Taf. 5 ; the Roman lady from the Heraion in Olympia,
Ausgrab. ii. Taf. 30, etc. The knotted fillet is known to be a priestly attribute ; cf. e.g. the Archigallus
relief, Schreiber, Kulhirhist. Bilderatlas, Taf. 16, 9.
2 Cf. especially the Munich head in Bernouilli, loc. cit. 193, where diadem, fillet, and the knot in the neck
correspond with the ' Hera,' but the front hair and face are like a portrait. The head of the Berlin colossal
statue (Skulpt. 587) is closely related to the ' Hera,' though it is a very inferior work ; it has never been separated
from the statue. The Ludovisi statue probably, like this one, had a horn of plenty on the left arm. Perhaps it
represented Antonia, the consort of Nero Drusus. She appears on the reverse of her gold and silver coins as
Constantia Augusti in chiton and mantle, with horn of plenty on the left arm, with torch raised aloft in her right
hand,, and wearing a diadem and side curls. The head on the obverse always has the knotted hair in the neck.
3 The head with diadem and lied fillet, which is placed on the statue of the Villa Albani, 711 (Helbig,
Museums, 844), is much more markedly Praxitelean (photo. Bruckmann).
4 Imhoof-Blumer and Gardner, Numism Comm. PI. Y 17 ; Bauermeister, Denkm. p. 1405 ; Ath. Mitth.
xiv. 229, Taf. 7. The hair seems to have been dressed in the same way as that of the Versailles statue. For
head-type cf. the coins of Ephesos of the third century, Brit. Mus. Catal. Ionia, PI. x. 4, xi. I, 2.
5 Winter in Jahrb. d. Inst. 1892, p. 164 sqq. Cf. infra.
* Ny Carlsberg Glypt. No. 1048 ; no drapery round the hips ; where the quiver-band and girdle meet, a
ram s head serves as buckle ; this, the arms of Delphi, would be specially appropriate in the statue of Antikyra.
The statue has the same turn of the head as the one in Versailles. The coin gives the head in profile simply
because the full face is not suitable to a coin. The more marked stride is clearly rendered on the coin. The left
hand must have carried the torch, the right the bow. On the left knee and behind on the torso are protuberances
of uncertain purport. The Jacobsen torso, No. 1049, is of Pergamene style, and quite different from No. 1048.

THE RESTING SATYR 329
ness of treatment show us the sculptor in his later development. This work is
preserved in a marble torso of the Museo Chiaramonti.1 It is highly probable that
a marble bust in the Hermitage 2 belongs to this type, and affords an idea of
the lofty beauty of the head. The hair is combed back as in the Brauronia, but
coiled round at the back in a plait. The head is bent, and has a gay, gracious
expression. This pose of the head and the facial forms resemble the Hermes of
Olympia as nearly as may be, considering the different personalities represented.
The composition as a whole appears to be preserved only on a Greek votive relief,
which itself is probably a fourth-century work.3 A Tanagra terra-cotta gives a later
adaptation of the design.4
With this statue and the statuette from Cyprus, mentioned above, we come to
another group of Praxitelean figures — those which lean with one arm on some
support. In his first period the artist does not seem to have been attracted by this
attitude, which we meet for the first time in his middle and later period ; it is
coincident with a broader, more pictorial style, as opposed to the more studied, almost
constrained motives of his earlier works. Peloponnesian influence has now fallen
into the background, and the artist follows only his own natural bent.
The motive was specially curious in the case of Artemis, for it seems unsuited
to her active energetic personality. So much the more clearly does the artist's ten
dency declare itself. The same remark is true of the Satyr, that frolicksome creature
whom Praxiteles yet presents to us in a graceful, contemplative attitude. After he
has shaken himself free from the older, more conventional statuary themes the master
gives the fullest expression to his own longing for cheerful realism and repose — as
contrasted with the Skopasian love of violent movement and excitement. In the
' resting Faun ' 5 we have a work belonging to his middle period. Its Praxitelean
origin can hardly be doubtful, although the statue cannot be identified with any
Satyr handed down by tradition. (It certainly is not the periboetos, which was part of
a group.) The Praxitelean authorship is not proved merely by the evident renown of
the work and by the gentle grace of the figure : another and more definite reason lies
in the likeness of the head to that of the Eubouleus (Plate XVI.) 6 The breadth of the
face, due to an exclusive attention to the front view, the shape of the forehead — with
its well-marked horizontal modulations, the luxuriant crown of hair confined by the
1 No. 122. Clarac, 573, 1228. Good workmanship, badly restored. Drapery specially lifelike. The
raised left shoulder shows that the forearm was raised and supported.
2 188 B. Made into a bust in ancient times. The breast shows the chiton and the mantle on the left
shoulder, but the folds are not the same as in the torso, and the quiver band is absent.
3 In Villa Albani, against the wall in the garden, No. 295. To the left three worshippers, to the right
Artemis in the type described (head broken off); below a dog. Drapery and attitude correspond with the
Chiaramonti torso. 4 Samml. Sabouroff, PI. 125, 126.
5 Cf. Friederichs-Wolters, 1216; Helbig, Museums, 525; Berlin, Skulpt. 258, 259; Brunn, Glypt. 105.'
Among the replicas note the Capitoline statue, smooth and elegant, apparently of the period of Hadrian. The
copies of Villa Borghese (Helbig, Museums, 943), of Villa Albani, No. no, and of Museo Torlonia, No. 113, are
very good. The two in the Vatican (Helbig, 211, 55), and that of the Lateran (Helbig, 639), are inferior ; on
the Braccio Nuovo copy the antique fragment of a pedum on the right upper arm is an addition of the copyist.
The best copy of the body is the Palatine torso in the Louvre. Among the heads, one in Vienna is to be noted,
because the beautiful nose is antique. There is a celebrated head in the Hope collection at Deepdene, in Surrey
(cf. Michaelis, Anc. Sculpt, p. 270), but I have unfortunately never seen it. [I was enabled to see this beautiful head
in the summer of 1893 : every effort, however, made to obtain a photograph of it for the English edition of this book
has proved of no avail. — E. S.] Of smaller replicas we may mention the one in Museo Chiaramonti, 582, and
the one in Tripolitza (cf. Ath. Mitth. 1879, p. 144, 1). The owner told me in 1878 that it came from Sparta ;
it is a good copy ; the part under the eyes is more hollowed out than usual.
6 Benndorf (Anz. d. Wiener Akad. 1887, Nov. 16) laid stress on the rounded form of the skull, the full
hair, and the fillet that presses into it as points of resemblance between the Satyr and the Eubouleus. U U

330 PRAXITELES
chaplet, and the way it falls at the sides and on the nape of the neck, are strikingly
alike in both heads. We must mention, too, one special point of likeness to the
Hermes — in the form of the foot, which has the characteristic great toe, slender
above and broadening at the end, separated from the others by a wide interval.
If we compare the body, on the one hand, to the works of the early period, and,
on the other, to the Hermes, we shall find that this Satyr comes between the two.
In the body the artist has certainly not attained x to the same degree of roundness
as in the Hermes : on the contrary, the older manner is still quite evident in the broad
planes, in the harsh depression between the straight and the oblique abdominal muscles,
and, above all, in the flat form of the abdomen itself. The treatment of the body is still
far removed from the stage reached in the Hermes ; still it is more advanced than in
the Satyr pouring wine ; the pubic region, for instance, is enlivened by little furrows
treated with great delicacy. In the unconstrained easy design, in the picturesque
draping of the panther-skin, which lies across the figure, in the exclusive attention paid
to the front view, and above all in the whole formation of the head, Praxiteles has given
unrestrained expression to his own new method, and thrown aside all that had bound
him to older models ; but in the formation of the body he had not yet gone quite so far.
With this conclusion it is, however, impossible to reconcile Brunn's view,
founded on generalizations, that the Hermes is an early work, the Satyr a mature
one.2 Brunn even thought that he had found the original of the Satyr in a
torso from the Palatine in the Louvre. The torso however is unfortunately only a
copy — a good one, it is true, but no better than other good copies ; even the material,
a coarse-grained marble cold and white in tone, and possibly Thasian, but certainly
not Parian, is never found in Greek works of the good period, though it is common in
later copies. The work too is in many places quite lifeless, and the modelling lacks
the finer touches ; it is only necessary to compare the part round the collar-bone
and the shoulder with the corresponding portion of the Hermes in order to grasp
thoroughly the difference between an original and a copy.
The blithe beauty of the conception and the entrancing sweetness of the head,
with the delicate smile on its lips, easily explain the reputation of the statue.
Protogenes, a painter whose originality was in inverse proportion to his admirable
technique, early utilized the design for a picture,3 and there are still later adaptations
in marble which give to the Satyr coarser forms, a prominent abdomen, and crossed
legs, and make him play the flute.4
We have already reckoned with the Eubouleus (Plate XVI.) as with an authentic
work of Praxiteles, for such we must maintain it to be, in spite of a contrary opinion 5
that has lately been gaining ground. The Roman copyists, as an inscription tells us,
copied the head of the ' Eubouleus of Praxiteles.' Now, to assume with Kern that
this Praxiteles may be any other artist who bore the name6 would be as.perverse as if
1 The copies all agree in this particular. 2 Deutsche Rundschau, viii. 200.
3 Strabo, 14, p. 652; Plin. 35, 106. Stephani (Compte Rendu, 1870—71, p. 99) wished to refer to this
statue the marble figures of the resting Satyr, which is impossible. The painter made use of the sculptor as
Eupompos made use of Polykleitos.
4 e.g. Berlin, Skulpt. 260, 261 ; Samml. Sabouroff, Text to Plate 77, 78, p. 2, note 6.
5 Kern, in Ath. Mitth. xvi. 1 sqq., also Helbig, Fiihrer, ii. p. 413. Sal. Reinach expresses doubt in
Chronique d' Orient, xxiv. 21, and in Daremberg et Saglio, Diet, d'antiqu. if. 850 ; Rubensohn differs, with
good grounds, Die Mysterienheiligtiimer in Eleusis und Samothrake (1892), p. 10, 36 sqq., 197 sqq. Kalkmann
(Prop. d. Gesichts, p. 82) cites the Eubouleus as non-Praxitelean.
6 Loc. cit. 20, with reference to the Praxiteles inscriptions of Roman times, Ldwy, /. G. B. 318, 319 (to
which should be added AeAnW dpx. 1888, 177; 1889, 32). These, however, all come from Greece, and were
written on bases not of copies but of original works, mostly portraits, and are worded like original inscriptions,
The wording EiifiovXevs Upa^ireXovs proves that the work was a copy.

XVI.
Head of Eubouleus.

THE EUBOULEUS 331
in the case of an inscription, from a copy of a ' Madonna by Raphael ' we chose to
doubt whether the celebrated Raphael or some other artist of the same name were
intended. The Eubouleus in question must have been by the celebrated Praxiteles,
whose works were so popular among the Roman copyists, and, since Eleusis is the
only place where we know of an independent cult of Eubouleus without the addition
of some other name, it is at Eleusis that the original of the Roman copy must have
stood. There are two extant copies of the work,1 in addition to which the original
itself has been found at Eleusis. The identification of this head, which evidently
represents an Eleusinian divinity and is eminently Praxitelean, with the Eubouleus of
the inscription, suggests itself at once, and is confirmed by subsequent evidence.
The identification has lately been opposed on mythological grounds, and it has
been assumed that the ' Euboulos,' or ' Eubouleus,' of Eleusis was only a surname of
Zeus. But it will be seen that the Eubouleus of Eleusis is neither Pluto nor Zeus,
nor any other divinity, but Eubouleus himself.
An inscription of the fifth century 2 shows that he was worshipped with Kore,
Demeter, Triptolemos, and two divinities called 0 (9eo? and r) ded, who are represented
on a votive relief ; 3 and a later inscription 4 gives the names of the priests of this
divine couple and of Eubouleus. This ' Eubouleus ' was also worshipped inde
pendently at Eleusis,5 where he must have been regarded as an under-world divinity,
since the pig and the serpent are known to have entered into his ritual ; 6 but he had a
separate existence, just like the couple #eo? and ded, who were probably akin to Pluto
and Persephone, though at Eleusis they had their own independent cult,7 even as
Klymenos and Pluto8 existed side by side at Hermione. The name Eubouleus,
whatever its meaning, be it the wise counsellor, the well-disposed or well-intentioned,
or a combination of both,9 belongs to that series of euphemistic names given in
shuddering propitiation to the powers of the under-world,10 in order to induce them to
exert only the beneficent side of their natures.
The tendency to connect and identify separate divinities with one or other of the
great canonical gods was at work from the earliest days, and is assuredly a main factor
in the progressive simplification of the primitive complexity. But the individual, the
special, the living elements peculiar to a cult are always there from the beginning, and
it is only later on that they get incorporated into another cult and eventually absorbed.
' Eubouleus,' who in the conservative ritual of Eleusis retained his individuality, is
older than the ' Zeus Eubouleus ' of the islands, where the male divinity worshipped
with Demeter and Kore was affiliated to Zeus, who in his character of Chthonios was
also a deity of the under-world. There are countless examples to show that the lesser
1 So-called Virgil in the Capitoline Museum (Helbig, Museums, 463) and in Mantua (Arndt-Bruckmann,
Einzelverk. No. 17), both recognized by Benndorf as exact replicas of the Eleusinian head. The head published
by Heydemann (Marmorkopf- Riccardi, Taf. 1) is of a different type. 2 Dittenberger, Syll. 13.
3 'E(pi)iJL. dpx. 1886, Taf. 3, 1. The relief belongs to the late fourth or third century B.C.
4 Ibid. Taf. 3, 2 (relief of Lakrateides). 5 Ibid. p. 262.
6 Cf. Kern, loc. cit. n.
7 In Sammlung Sabouroff, i. Einl. Skulpt. p. 22, note I ; p. 29, I pointed out that Be6s and Bed cannot
be identified with Pluto and Persephone, and that it is futile to try to name them at all. Tdpffer (Attische
Geneal. p. 33) arid Rohde (Psyche, i. 196) express the same opinion. It is retrogression to identify them again
as Kern does (loc. cit. p. 6) with Pluto and Persephone. On one relief Demeter and Persephone are seated beside
the Bed, and on the other Persephone was represented in addition to the Bed, as the complementary fragment
found through Heberdey and Reichel proves (Ath. Mitth. xvii. 127) : Be6s and Bed are here an absolutely
separate pair beside Persephone and Pluto. 8 Paus. ii. 35, 10. Cf. Rubensohn, loc. cit. 198.
9 Sal. Reinach in Daremberg et Saglio (Diet. ii. 849) lets both interpretations stand.
10 Cf. Sammlung Sabouroff, loc. cit. 22. Cf. ibid, note 2, where Kern could have seen that the inscription from
Lebadeia (mentioned by him, p. 10), which seems to prove that the name Eubouleus was applied also to
Trophonios, was already quoted.

332 PRAXITELES
deities partly retained their original individuality, and partly became annexed to some
one of the greater gods ; for instance, one of the mother goddesses was worshipped
in Attika separately as Kourotrophos,1 and also as one withGe or Demeter; the same
is the case with Pandrosos and Athena Pandrosos, Hygieia and Athena Hygieia,
Erechtheus and Poseidon Erechtheus, Peitho and Aphrodite or Artemis Peitho,
Hekate and Artemis Hekate, Eileithyia and Artemis or Hera Eileithyia, Iphigeneia
and Artemis Iphigeneia, Eukleia and Artemis Eukleia,2 Aristaios and Zeus Aristaios
(in Keos), Asklepios and Zeus Asklepios, Maleatas and Apollo Maleatas, etc.
Legend, which is wont to enter into facts of belief and ritual as a secondary
element, called forth into existence the distinct mythological personality of Eubouleus
in the ritual of Eleusis. The story grew out of the customary offering and burial of
young pigs, and Eubouleus was made into a swineherd, whose swine had been swallowed
up in the chasm when the earth opened at the rape of Kore. Kern has well shown how
the myth arose from the ritual,3 but it is purely arbitrary to call the myth late, and
to assume without the slightest foundation an older stage in which the swineherd of
Eleusis was no other than Zeus himself. Legends that arise out of rituals are by no
means usually of late origin.4 On the other hand, Triptolemos the neatherd is only a
later duplicate of Eubouleus the swineherd.5 Now, as legend could find no place for
Eubouleus as a god of the under-world, it accordingly changed him into an heroic
attendant on the great goddesses -; he became brother to Triptolemos, and also the
son of Demeter, and as such he came, even as Triptolemos, to be conceived of as
youthful. The monuments will now show that this aspect of Eubouleus is familiar
in the fine period of art.
On one of those Attic statuette vases (Fig. 142) on which the boy lakchos appears
so often there is a youth with thick locks of hair, high diadem and wreath, standing like
a god, his mantle draped about his lower limbs, holding a sucking-pig in his right
hand, and in his left a bundle of the twigs or the ears of corn 6 used in the mysteries
of Eleusis.7 The same figure, only represented in movement, appears on the cele
brated Attic hydria,8 decorated with a frieze in relief of the Eleusinian gods ; here
the youth stands close to the great goddesses. The statuette vase is sufficient to
show that he is not a mere sacrificial attendant ; he must be a divinity, and as a fact
the figure can only be satisfactorily and fully explained by reference to the myth of
Eubouleus, which must accordingly have been current in the fourth century, the
period to which the vases belong.
1 A votive relief from Athens in Dresden shows the goddess as Kourotrophos with a torch (in the Hekate
type), conjoined with Apollo, Leto, and Artemis (Jahrb. d. Arch. Inst. Anz. 1894, p. 26). Dedication to the
goddess Kourotrophos on a cup from the Peiraieus, 'E^rj/i. dpx- 1884, p. 194, 6. Sacrifice to her, C. I. A. ii.
481, 59. Prayer in Aristoph. Thesm. 299.
2 Cf. the interesting passage in Plut. Aristeid. concerning Eukleia at Plataia ; some identified her with
Artemis, others not. 3 Loc_ ciL l6,
4 After the publication of O. Midler's Prolegomena, it should no longer be necessary to say that the fact of a
legend only appearing in later sources proves nothing against its antiquity.
5 As Kern shows (loc. cit.), Triptolemos the neatherd is connected with Alexandria and Osiris. This
naturally proves nothing with regard to the swineherd.
6 For its use in the Eleusinian cultus, see Arch. Anz. 1892, p. 106, notice to No. 19. Of great importance
in this connexion is a fragmentary B. F. vase in the Museum at Eleusis. It represents a procession of men and
women, all bearing the bundles of twigs which belong to the mystic cultus.
7 The vase is published in Frdhner's Catalogue of the Piot Collection, Paris 1890, No. 153, p. 42 (= our
Fig. 142). Frohner mistakes the pig for a hare. A replica of the figure was in the Coll. Greau (Frohner, Coll.
Griau, Terrescuites Gr. 1891, PI. 40, No. 462) ; the form is founded on the Piot example, but the bundle on the
left arm (it was put on separately) is missing ; the pig is better preserved.
<• stephani, Compte Rendu, 1862, Taf. 3 ; Gerhard, Ges. Akad. Abh. Taf. 78; Heydemann, Marmorhpj
Riccardi, p. 14 seq. The vase is certainly Attic.

THE EUBOULEUS

333

A striking confirmation of this Eubouleus type is afforded by a marble statuette
at Eleusis x which corresponds with the terra-cottas in all essentials. The boy, whose
soft round face and luxuriant hair resemble the marble head, carries on the left arm
the bundle of ears,2 and in the right hand may be seen
the fragments of a small pig.3 Here too he wears the
mantle, as befits a god. The workmanship shows the
figure to belong to the fourth or third century B.C.
Another Eleusinian monument represents a different
type of Eubouleus ; it is the newly reconstructed relief
of Lakrateides.4 The boy at the end on the right, with
long hair and short-sleeved chiton, who holds a torch,
can scarcely be other than Eubouleus, whose priest
Lakrateides is. He stands beside the pair, #eo? and
6ed, as Triptolemos stands beside Demeter and Kore.
Triptolemos wears the same sleeved chiton as he, and
high boots, which we may also assume for Eubouleus
(the legs are missing). This type, established for
Eubouleus by the Lakrateides relief, occurs on five Attic
vases in conjunction with the divinities of Eleusis.5 It
is a youth who appears, either alone or with Triptolemos
and Dionysos, as a companion of the two great god
desses. Like lakchos in the group of the elder Praxi
teles, he is their torch-bearer, or he holds the sceptre,
like the lakchos issuing from the earth, on another vase.6
He usually wears an embroidered festal robe with
tight sleeves, and high boots. This type is influenced
by the type of lakchos, but the former one was created
for Eubouleus. Finally, we must mention another Eleusinian monu
ment which represents Eubouleus, this time shortened
into a bust 7 like the Praxitelean work. It appears in
relief on a tablet set up within an aedicula which is supported on a pillar, just the
arrangement which I imagined we must assume for the Praxitelean head.8 The head
has again the full soft features and almost feminine neck, also long shoulder-curls like
the Eubouleus of the Lakrateides relief.

Fig. 142. — Attic statuette vase.

1 Noted by me in 1894 in the Eleusis Museum.
2 The lower part of the bundle shows plastic stripes, above it is painted red.
3 The forelegs are still recognizable. 4 In the Museum at Eleusis.
0 Gerhard, Ges. Akad. Abh. Taf. 77 (youth with two torches). Ibid. Taf. 71, I (youth to the left of Demeter
with one torch). Hydria in Athens, Coll. of the Arch. Society, No. 2722 (youth, with sceptre, short-sleeved chiton,
and high boots, beside Demeter, Kore, and the bull-horned Dionysos. Style of fourth century, polychrome and gilt).
Hydria of the same style from Coll. Castellani (Frohner, Coll. Cast. Rome, 1884, PI. ii.), now in the Tyszkiewicz
Coll. (youth in long chiton and mantle, with twisted curls, carrying a sceptre). Vase in Athens, Arch. Soc. No.
2382, fine fifth-century style ; three figures separated by palmettes and the handle : Demeter, Kore, and youth wearing
boots and an embroidered garment. This vase alone refutes the assumption that the figure is a priest, for it
undoubtedly represents a triad of divinities. I have already explained the figure as Eubouleus in Arch. Anz.
1889, p. 47 ; Roscher's Lex. i. 2185.
6 Tischbein, i. 39; Jahrb. d. Inst. 1891, p. 12 eq.
7 Relief in the Museum at Eleusis (Photogr. in German Ath. Institute, Eleusis, 29).
8 Cf. Arch. Anz. 1889, 147. Even after J. P. Meier's remarks in the Jahrb. d. Inst. 1890, 209, I still
think this hypothesis the most likely. The side wall of the aedicula probably joined on to the left shoulder, as is
shown by the perpendicular contact-surface. The left shoulder was free. Perhaps there was a second bust on the
left side of this one, which would explain the turn of the head of Eubouleus.

334 PRAXITELES
But replicas of the Praxitelean head itself have been found in Eleusis. The
only complete one, now in the Museum at Athens,1 is neither poor nor late, but is
evidently only a copy from the Praxitelean original which has been preserved to us.
In the original we can trace the master's touch, in the copy we only get a spiritless
and tame imitation. This contrast is specially noticeable in the hair. The amazing
mastery of the Praxitelean head lies in the indefiniteness of the hair, in its curling
disarray and shimmering undulations, qualities which the copyist was powerless to
catch, although he imitated the technique of the drill. In the Eleusinian copy,
just as in the Roman, the lines of the hair are arranged more according to scheme
and rule, a process in which all the original charm is lost. So, too, the fresh untram
melled hair of the superb Skopasian head from the south slope of the Akropolis 2
lost all spontaneity in the Roman copy. The existence of replicas at Eleusis
affords a proof of the canonical value attached to the original, by showing that
whoever in later times desired to set up an image of Eubouleus at Eleusis only
required to have a copy made of the head by Praxiteles.
The Eubouleus presumably belongs, like the ' Satyr at rest,' to the artist's middle
period. In it he made free use of the drill — as he must have done in the original of the
Satyr — to make the hair look loose and lifelike. Two other technical peculiarities
show that it is somewhat earlier than the Hermes : the lower eyelid is not treated
with the characteristic delicacy,3 but its hard and clearly marked outline is more like
what we notice in the original heads attributed to Skopas. Further, the corners of
the mouth are not as in the Hermes worked with the drill, but, like the Skopasian
heads (where the drill is not employed even for the hair), they show no traces of
its use. Since Praxiteles did not employ all the technical resources of his later
style in the Eubouleus, it follows that the work must be earlier than the Hermes.
We are actually watching the growth of that refinement in the working of marble
which distinguishes Praxiteles even from Skopas.
There is a Roman copy of a superb head (Fig. 143) 4 so like Praxitelean work,
and more especially the Eubouleus, that we must needs be referred to an original
of the master. The breadth of the full delicate face, the cut of the eyes, the chin
with the shallow dimple, and the broad fleshy neck resemble the Eubouleus ; so does
the thick loose hair, which is worked with the drill, and was of course still more
effective in the original. The two heads, however, also offer marked contrasts to one
another. The short locks are parted in the middle and fall evenly and simply down
on either side of the face, but not over the forehead. The head is thrust somewhat
forward, and the expression is not in the slightest dreamy ; rather is it energetic and
expectant. The personage is more youthful than the Eubouleus, and the forehead is
smoother ; but the way of wearing the hair and the muscular neck show without
a doubt that the head is male.
I think that no name seems to fit this head so well as Triptolemos, the brother of
Eubouleus, whom Praxiteles had, according to Pliny, represented in a group {Flora,
Triptolemus, Ceres in hortis Servilianis). A comparison of the Triptolemos on the
1 Ath. Mitth. xvi. Taf. 2. (The other replica is only a fragment.) The eyes were inserted; this was
enough to make the Praxitelean refinement in the rendering of the eye impossible. Kern claims the head. of a
statue in Rome (loc. cit. Taf. 1) as a repetition of the same type, but the resemblance is only general. The sym
metric fall of curls over the forehead, characteristic of an under-world god, is absent.
2 Ath. Mitth. i. Taf. 13, 14; Skulpt. Berlin, 610; Rom. Mitth. 1889, p. 216.
3 Cf. Graf, in Rdm. Mitth. 1889, 204.
4 Pal. Pitti, Diitschke, ii. 27. Nose and lips restored. Length of face 0-17. Arndt and Amelung also
thought of Praxiteles as the author.

Fig. 143.— Head in Pal. Pitti (Florence).

336 PRAXITELES
great Eleusinian relief,1 which belongs to the period and the school of Pheidias, shows
that his hair was parted in the middle and fell on either side, and that even the
attitude and expression of that head (after making all due allowance for difference of
style) are extraordinarily like this one. Praxiteles must have taken that older type
of Triptolemos as the basis of his creation.
The Sauroktonos can be but very slightly older than the Hermes.2 The body
comes appreciably nearer to it than did the ' Satyr at rest' The rich round modelling
of the abdominal region and the rendering of the navel with a fold of skin above it
are already quite in the style of the Hermes, and absolutely different from the
Satyr pouring wine and the Eros ; though the chest with its quiet broad surfaces
passes without much transition into the sides, still somewhat after the older
method. The copies do not allow us to judge of the subtler beauties of the head ;
the face with its soft setting of hair recalls the Aphrodite of Knidos ; yet the
forehead broadens considerably at the base in order to characterize Apollo, in
accordance with a type elaborated, as we have already seen, as early as the fifth
century (p. 194).
The statue possesses a wonderful charm, which is due entirely to the attitude, on
which the artist has expended all his skill. It would have been more natural had the
right hand with the arrow been nearer to the object aimed at, but this would have
spoilt the beautiful front view of the body.
The true interpretation of the attitude seems to lie about midway between
the fluctuating theories which have been propounded on the subject — i.e. the
action is a playful one without any mantic meaning, yet the lizard was probably
an attribute of Apollo, in the ritual for which this statue was made ; and the
Sauroktonos is playing with a lizard, just as the Smintheus of Skopas was playing
with a mouse.
The Eros of Parion, which as I had already shown from coins' 3 was leaning on his
left arm, has been happily identified 4 in the so-called ' Genius Borghese ' in the Louvre.6
This statue is of course a very poor copy, but it enables us to assign to the original
its place in the Praxitelean series. The hair with the loop tied up above the ears is
very like the Sauroktonos ; yet the little knot of hair over the forehead, and the
symmetrical division of the locks just at that spot, recall the Thespian Eros. How
ever, even the chest no longer shows any trace of the earlier type, for it is as softly
rounded in every direction as that of the Hermes. There is a wide gap between it
and the master's earlier statues of Eros. This difference is not merely a difference
of form but also one of expression. The whole effect conveyed by every detail
of attitude, form, and face is soft and round, and the feeling excited in us by this
' Genius ' is one of voluptuous delight.6 On the other hand, the Thespian Eros, with
1 I consider Grafs attempt (Ath. Mitth. xv. 36 seq.) to see in this relief Boeotio-Peloponnesian style to be
completely mistaken.
2 Replicas, Overbeck, Apollon, 235 seq. In judging of the style we should study the marble copies executed
on the scale of the original, not the reduced and less faithful Albani bronze (whose Roman origin is sufficiently
proved by the technique, especially by the inlaid band of silver).
3 Roscher's Lex. i. 1358.
4 By Benndorf in Bull, della Comm. Arch. 1886, p. 74.
6 Frohner, Not. 326. Photo, ed. Giraudon, 1201. So-called Thasian marble. The connexion between
the figure and the antique stem and drapery is new ; the stem probably stood rather nearer. The head is un
broken. The workmanship is poor, flat, and spiritless, yet the beauty of the original shines through. The motive
of the Eros of Parion recurs on a Greek mirror and gem (cf. Roscher's Lex. i. 1359).
6 Winckelmann (Hist, of Art, bk. v. ch. 1, § I2 = tr. Lodge, vol. i. p. 325) felt deeply the beauty of the
work. In saying that it was ' fashioned after the beauty of angels ' he was evidently influenced by the voluptuous
angel of seventeenth- and eighteenth-century art.

HERMES— APOLLO— DIONYSOS 337
its well-knit forms, is still under the spell of the constrained rhythm of Peloponnesian
creations,1 and seems consumed with repressed longing.
We have now arrived at that period of the artist's life to which the Olympian
Hermes belongs. Many monuments can be grouped about this statue, but only the
more important need be mentioned here. First of all comes an ' Apollo at rest ' with
his right arm over his head — a famous work, to judge by the many copies and later
adaptations.2 Both the attitude and figure are extraordinarily like the Hermes, and
the spirit of the work is so thoroughly Praxitelean in the quiet beauty of its flowing
lines, so radically different from the work of Skopas for instance,3 that we naturally
assign it to the master of the Hermes. The position of the head is like that of
the Aphrodite of Knidos ; the eyes gaze into the distance, and even the features
show such a family likeness to the Aphrodite that the authorship of Praxiteles
cannot be doubted. The hair is a fresh example of the master's skill in inventing
rich and characteristic coiffures. The god is only lightly leaning on his left arm, for
the left leg is not absolutely free and drawn behind the other as in the Hermes, but
is planted flat on the ground, so as to give the figure more dignity of bearing.
A Dionysos, of which copies likewise exist, corresponds so closely to this 4
Apollo that it must be the work of the same artist. The attitude and the bodily
forms are identical ; on the left shoulder a nebris is knotted, and clings closely and
softly to the figure so as to reveal its shape. The bacchic mitra is fastened into the
hair, which falls in long curls over the shoulders. The type of face also, with the far-off
look, is very like the Apollo, though it is modified to characterize a more effeminate,
sensuous nature.
There is another Dionysos at Madrid,5 which is certainly Praxitelean, and belongs
! I may just mention at the close of this section that a series of Kanephoroi is preserved to us as Caryatids
in copies which must go back to originals from the circle of Skopas and Praxiteles. Three types may be dis
tinguished : (a) Villa Albani, 725; Friederichs-Wolters, 1557; Helbig, Museums, 837; Clarac, 444, 814 B;
left supporting leg, head related to Knidian Aphrodite. A replica without head in Athens, found near the Metro
politan Church, (b) Villa Albani, 628; Friederichs-Wolters, 1556; Helbig, 834; Clarac, 444, 814. Rests on
left leg. Hair in furrows. Replica in Athens found with replica of a. (c) British Museum, Newton, Guide Graeco-
Rom. Sculpt. 1, 126; Clarac, 444, 813. Rests on right leg. Hair in furrows. The types are magnificent and
worthy of the greatest artists. The replicas in Athens were perhaps made to replace the originals which the Romans
took away. According to a suggestion kindly communicated to me by Bulle, these types can be traced back to the
Triopian sanctuary of the Chthonic divinities near Knidos. This would answer admirably to our assumption of a
Praxitelean origin, for Praxiteles worked in Knidos. There was a Kanephora by Skopas in Asini Monumentis ;
the Kanephora of Praxiteles, Plin. xxxiv. 69, is, however, only a conjecture, the reading oporan of the cod.
Riccardianus being very probably the correct one. [Cf. now Bulle's article, Rdm. Mitth. ix. p. 134^^. His
own view is that the Roman statues belonged to a sanctuary of the Triopian Demeter on the Via Appia
dedicated by Herodes Atticus in memory of Regilla, and that the Athenian copies possibly belonged to the
sanctuary of Serapis and Isis, in the neighbourhood of the Metropolitan church. Neither series stood free,
according to Bulle, but were placed pillar-wise against a wall. — E. S.]
2 For the type and its Praxitelean origin see my Satyr von Pergamon, p. 21, A, 2 ; Roscher's Lex. i. 460—
462 ; Samml. Sabouroff, to Taf. 36. Collection of replicas (including some wrong material), Overbeck, Apollon,
209 seq. Good copies are Berlin 44 and 512 ; Louvre, Frohner, 75, 76, and a torso, No. 3013 ; a torso in Palazzo
Mattei. A good copy of the head is falsely placed on a seated statue of Apollo in the Museo Boncompagni-
Ludovisi (Helbig, Museums, 877). Schreiber (V. Lud. 116) wrongly explains it as modern; another copy is
in the Conservat. Pal. Salone,, No. 31, placed on a female statue. Two replicas of the head in the Brit. Mus.,
Friederichs-Wolters, 1292 — 93. One in the Coll. Barracco (Coll. B. PI. 59, 59 a). Kalkmann, misled by his
measurements (Prop. d. Gesichts, p. 62), places this Apollo in the fifth century, and, on the other hand, calls a
late Roman copyist's modification (in the Brit. Mus. from Kyrene) Praxitelean ! Cf. my remarks in the Berl.
Philol. Wochenschrift, 1894, p. 109. 3 Birch and Wolters, loc. cit., thought of Skopas.
4 The so-called Bacchus de Versailles in the Louvre, Frohner, Notice, 218; photo, ed. Giraudon, 1189.
Well preserved, with head. Ibid. Frohner, 216.
6 Friederichs-Wolters, 1485. I know no replicas ; the bronze Dionysos published by Milani (Museo Ital.
iii. 752 seq.-) as Praxitelean has no connexion with Praxiteles. Cf. infra, p. 353.
X X

338 PRAXITELES
to the same period of the artist's life. In conception it most resembles the Eros of
Parion. The god is resting his arm on an archaic terminal figure.
But belonging to this later period there are also figures standing free without
any support. First of all comes, as was noticed immediately on the discovery
of the Olympian Hermes, the so-called Antinous (Hermes) of the Belvedere.1
The head and the bodily forms, even down to minute details like the hair over the
forehead, correspond most closely to the Olympian statue. Still there are differences
which cause one to suspect that this work is a still later development of the
Praxitelean manner, and may be by one of his sons ; these differences are the
modelling of the chest, the greater curve of the hips, and the smallness of the head,
which however seems to have lacked the ideal grace and refinement of the Olympian
Hermes. The right arm is placed on the hip — an attitude often found in statues by
Skopas ; here however it assumes a different character. Skopas made it express
fierce impatience and restlessness, but here it conveys an idea of graceful nonchalance.
Compare for instance the Meleager of Skopas. It is untrue to say that the difference
arises from the subject-matter ; for the Hermes belonging to the earlier works of
Skopas was, as we saw, represented in like unrest (Fig. 129). This again is a point
in which Lysippos is linked to Skopas ; a beautiful bronze statuette from Athens 2
gives us a Hermes of purest Lysippian style, in the swinging posture of the
Apoxyomenos, the left arm placed behind the back, the right holding the kerykeion
sideways, the head jerked impatiently to one side. How totally different is the
thoughtful, quiet Praxitelean figure !
Of another Praxitelean Apollo3 we get at least a glimpse in a torso {supra,
Plate XV.) Like the Belvedere Hermes, he rests full on the right leg, with the hip
curved outwards, and does not lean against any support ; the right arm is raised to
feel for the quiver at his back. The head is turned towards the side of the free leg.
The contours agree with those of the ' Apollo at rest ' and the Hermes of Olympia,
but they are somewhat softer.
There is another statue, a Hermes (Fig. 144),4 which is particularly interesting
owing to the novelty and originality of the motive. The artist has for once broken
through his practice of only representing figures at rest, and has attempted some
thing more vigorous. Even here however the master's bent is unmistakable, for the
design is a model of perfectly rounded beauty and internal repose.
Besides a full-sized copy in Florence, there are two statuettes, which however
are not quite exact in details.5 The attitude is the same as that of the ' Apollo at
rest ' and of the Dionysos, except that there is no support for the left arm. However,
1 Treu, Hermes mit dem Dionysosknaben (1878), p. 8 seq. Cf. Friederichs-Wolters, 1218 — 20; Helbig,
Museums, 145. The replica from Andres is not faithful stylistically, but mixes in Lysippian elements ; the Roman
copies are more faithful. A good one in the Lansdowne Collection (Michaelis, Anc. Sculpt, p. 454, 65), and one
in the Pal. Vecchio in Florence (Diitschke, ii. 511; the head antique, but foreign). Amelung (Florentiner
Antiken, p. 37) mentions replicas of the head.
3 Berlin, Inv. 6305 ; cf. supra, p. 300. The bronze corresponds very closely with the Apoxyomenos in
proportions and details of form. It appears to be an original of the same school.
3 In Villa Medici, Matz-Duhn, 215. This is the torso on which has been placed the beautiful Meleager
head published in Ant. Denkm. i. Taf. 40.
* Uffizi, Diitschke, 98 ; Gall, di Firenze, Ser. 4, 135 ; Clarac, 666 C, 1512 A. Cf. Arch. Anz. 1889, 147.
The drawing by Heemskerk, mentioned by Michaelis, Jahrb. d. Inst. vi. p. 140, 26, C, seems to represent this
statue. The left arm and the beginning of the right upper arm are antique ; the kerykeion is restored ; head
unbroken ; nose, lips, and edge of petasos restored ; legs, wings on feet, and plinth are antique.
6 a. Marble statuette, Gall, dei Candelabri, 17 ; only the torso antique ; the chlamys is fastened on the right
shoulder, b. Bronze statuette in Naples, No. 5208. No chlamys ; the right hand, bent and empty, seems to
grasp the hat. The left hand is held as in the Florentine statue. No kerykeion.

HERMES

339

since in those other two statues the support is almost without influence upon the
position of the body, seeing that the left foot is planted flat to the front, while the
left arm is only slightly supported, the actual difference is very slight. The close
connexion of the Uffizi Hermes with the Apollo, the -Dionysos, and the Olympian
Hermes is proved above all by the correspondence of the bodily forms, somewhat

Fig. 144. — He:

in the Uffizi.

hard and dry though the copyist's work was. In this case too the right arm is raised ;
unfortunately only the stump is left in the large statue, but the bronze statuette
shows that the hand was empty. The left one holds a kerykeion like the
Belvedere and the Olympian Hermes ; the chlamys falls over the arm as in the
latter, an arrangement which we sometimes find on fourth-century Attic grave-reliefs.

340

PRAXITELES

The head reminds us of the ' Satyr at rest ' and of the Eubouleus, for the hair grows
thick and full on the crown of the head, then falls away on either side of the face,
and there is the same breadth of forehead, while face and neck agree completely.
There are traces of a growth of beard on the cheeks — he is irpSiTov vTrr)vrjTr)<;,
as Homer represents him (II. xxiv. 38).
In fact, the whole conception is Homeric — this Hermes is no longer the youth
trained in the palaestra, with close shorn hair and ephebic type of face ; he is a free
child of nature, a young
man in the prime of his
strength and beauty, a
ruler's son, /covpat alo-vp,vr\-
T?)pi eoi/cax;, a son of Zeus,
as we are reminded by the
typical growth of the hair.
The Hermes is the messen
ger between gods and men,
rejoicing in his strength as
he lifts his arm and looks
upwards, and we feel that
his winged feet are about
to bear him and his mes
sage to Olympos.
This Hermes recalls
the Eros described by
Kallistratos, as about to
cleave the air with his
wings. I believe I have
discovered a reproduction
of this statue in a small
bronze,1 which shows Eros
with both arms raised pre
paring to soar upward.
This audacious conception
would of course belong to
the artist's later period.
In close connexion
with this Florentine Hermes comes a statue of Herakles in the Villa Albani (Fig-
145).2 The attitude is identical, and the statue has the typical -late Praxitelean
forms. The hero holds his club in the left hand just as the Hermes holds his
kerykeion, and the right hand too is upraised ; the restorer has put a cup into it,
and it is not improbable that it actually once held a drinking-vessel. Unfortu
nately a portion of the neck is new, but the head itself is genuine, and its turn
to the right is shown by what remains of the neck to be accurate. The paws
of the lion-skin are knotted together on the right shoulder so as to produce
1 Jahrb. d. Ver.f. Alterth.-Fr. im Rheinland, Heft 90, p. 61 sqq. Taf. 3, 3. Cf. p. 314, note I.
2 Clarac, 804 B, 2007 A ; Helbig, Museums, 840. The front of the right leg above the knee, the front of the
foot, the left right arm (except at the shoulder, where sufficient remains to show that the arm was raised), the left
hand, and most of the club, except a small piece in the middle, are restored. The neck is partly restored. The
body has suffered much from exposure. The support with the quiver is antique. Emil Braun (Ruinen und
Museen, p. 706 seq. ) well appreciated this little-known statue.

Fig. 145. — Herakles in Villa Albani.

HERAKLES

341

the effect of a chlamys, and the head of the animal falls over the breast ;
this is an innovation, for it is usually either drawn over the hero's head or hangs
over his arm. On the other hand, it is easy to see that this draping of the skin,
with the head in front, is dictated by the same taste as the panther-skin of the ' Satyr
at rest.' All these points justify us in assigning this work also to Praxiteles.
Like the Hermes of the Florentine statue, Herakles is gazing up brightly and cheer
fully, his victories won.1 He is the hero, who by sheer prowess has attained
Olymposy and with conscious pride
looks upwards to the heaven where
eternal youth and happiness await
him. How different to that rest
less hero thirsting for new feats
whom 'Skopas conceived !
A well-known statue in the
Museo Chairamonti (Fig. 146) 2 is
a modified replica of the one in
the Villa Albani ; the bodily forms
are considerably altered ; a Per-
gamene layer has been superadded
to the Praxitelean substratum.
The superficial realism, the turgid
manner of the Pergamenes, has
obscured the beautiful simplicity
of the original design ; even the
lion-skin has become more effect
ive and pathetic. The attitude is
changed — the club is transferred
to the right hand, and held down,
not up, and therefore the head is
less raised. The child Telephos,
about whom his father is but little
concerned, is awkwardly adjusted
on the left arm. The addition of
Telephos shows that we have to
do with a Pergamene adaptation-
adaptation.3 The head of the statue (Fig. 147, from a cast) shows, as compared with the
Albani head, some traces of later modification, though mainly in externals ; the beard
is less full and deeply undercut, after the Pergamene fashion, and the twisted
fillet is an addition. However, in its main lines the head does not differ

Fig. X46. — Herakles with Telephos (Mus. Chiaramonti).

-that is to say, with a Roman copy of such an

1 A bearded Herakles, looking up, but more pathetic than the Albani one, is in Dresden (Augusteum,
Taf. 91); skin over the head; in bodily forms rather older than the Albani statue. Beardless Herakles
(excellent) with the same motive in the Coll. Jacobsen in Copenhagen (No. 1050). The attitude and build
of the body are very similar to the Apoxyomenos of Lysippos, but the head is like the Skopasian type. One
hesitates whether to call Lysippos or Skopas the author. The composition is far more pathetic and restless
than the Praxitelean.
2 No. 636. Helbig, Museums, 113. Very well preserved. Unimportant restorations. The upper end of the
club with the right hand is antique. Head unbroken, only the point of the nose restored. I can see absolutely
no grounds for connecting the statue with the Munich Diomede, as Brunn has done (Bayr. Sitzungsber.
1892, 662).
3 For Telephos with Herakles, cf. Roscher's Lexikon, i. 2246 seq,

342

PRAXITELES

appreciably from the Albani Herakles, and its presupposed Praxitelean model ; for
instance, it is quite uninfluenced by the exaggerated Lysippian ideal of Herakles.
Therefore we may believe that the main points of the Herakles of Praxiteles are
faithfully reproduced in the Chiaramonti head. For even the system of forms
corresponds throughout with the Hermes of Olympia ; the modelling of the forehead
is strikingly similar, though rather more powerful ; further, the. eyes and their
surrounding parts, the ancient portion of the nose, and the profile completely
correspond. It is only in details, like the hair and the lower eyelid, that those
differences of treatment appear which always distinguish a copyist's work from an
original. In presence of this profoundly felt conception of Herakles, undoubtedly the
most beautiful image of the hero that exists beside the beardless Skopasian type,

Fig. 147. — Head of Herakles Chiaramonti.

many will feel reminded of another famous ancient type— the ' Zeus of Otricoli.'
There is an undoubted resemblance in the system of forms, but the likeness
is due to their being not the work of one master but of one development and
age. It has recently been more than once suggested that this famous head is a
specimen of the work of the second Attic school, and the suggestion is correct.1 It
is now possible to conjecture what the statue which belonged to this head was
like. At Constantinople there is a bronze statue of Zeus (one third life-size)2 whose
head exactly reproduces the Otricoli type. The powerful nude figure has nothing
Lysippian about it, and is connected rather with the Praxitelean manner. The figure
is in the walking motive, with all the weight thrown on the right leg, and the left
1 U>sdik.e, Jahrb. d. Ver. v. Alterth.-Fr. im Rheinland, Heft 88, p. 266 ; Helbig, Museums, 294.
" Found at Janina.— B.C.H. 1885, PI. 14.

THE PETWORTH APHRODITE 343
hand raised aloft. According to the marks on the basis, this was precisely the
attitude of the six Zanes which the Athenians erected at Olympia (B.C. 332).
These Zanes must assuredly have resembled the Constantinople figure in style as
well as in motive, and the Otricoli mask may very possibly go back to one of
them. We have reserved for the end a very beautiful and interesting work. This is,
as I hope to be able to prove, a real original by Praxiteles himself, a head of his
favourite goddess Aphrodite, belonging to his later period (Plate XVII. and Fig.
148). The head, which belongs to Lord Leconfield, was till lately in the Petworth
collection, but has now been removed to the owner's London house.
This superb life-size head l was found last century. When I saw it for the first
time in 1888 2 I was absolutely enraptured with its beauty. Unfortunately the head is
somewhat damaged, the nose and part of the upper lip being restored, and, worse than
this, the whole face, except for a few small portions, having been rubbed smooth, so
that the freshness of the ancient surface is lost. The corners of the mouth and the
adjacent parts seem to have suffered most from the polishing process, hence we may
suppose that the drill-marks in the mouth were originally less conspicuous than they
are now. Still, it is only the last delicate touches that have disappeared, and in all
essentials the head is well preserved. Even the antique polish survives uninjured in
some parts. Originally the whole neck was polished, but the hair left rough — the
Hermes of Olympia being another instance of the same method. The marble is
Parian, of that fine variety known as lychnites, which was also employed for the
Hermes. As the block was not large enough, the back of the head with the bunch
of hair was made separately and joined on. The join is straight, and so neatly
effected that it is to-day as good as ever, and must have been quite invisible when
the hair was painted. It is well known that in the Hermes several pieces — not,
however, parts of the head — are put on in the same way, with straight surfaces.
The head is intended for insertion into a statue, which we may suppose was
draped and made of marble less fine in quality.
In Specimens of Ancient Sculpture (i. 45, 46) there are engravings from the head.
The editor, by suggesting that it may be ' a relique of the parent statue from which so
many have been derived,' shows that he looked upon it as a Greek original. Ottfr.
Miiller {Handbuch, §375,3) classes it among the older and more majestic types of
Aphrodite.3 It seems to have disappeared from the collection after this. Michaelis
did not see it,4 nor did Conze.
We are now able, in the light of recent discovery, to say with certainty that this
is a real original work by one of the first fourth-century artists. The very best copies
of Praxitelean heads — even, for example, the good copy of the Aphrodite of Knidos
in a private collection in Berlin 5 — seem lifeless empty masks beside it. Nothing is
worthy to be compared with it except an original such as the Hermes of Olympia.
There is only one period of Greek art to which the hair technique of this head can be
assigned, and that is the period of Praxiteles. I am speaking now not of the arrange
ment of the hair, though that too is purely Praxitelean and corresponds in essential
1 Length of face, 0'2I ; brow, nose, and lower part of the face each measure 7 cm. Distance between
outer corners of the eye, 0-115 ; length of mouth = 0-056.
2 Arch. Anzeiger, 1890, p. 164.
3 Mentioned again by Ottfr. Miiller in Kunstarch. JVerken, ii. 89.
4 He is mistaken in mentioning it as a replica of the Venus of Medici (Anc. Marb. p. 616, No. 73), from
which it differs in size and pose.
5 Antike Denkm. i. 41. Cf. p. 322, note 3.

344

PRAXITELES

features with the Aphrodite of Knidos, but of the manner of execution. The roots of
the hair are rendered with marvellous delicacy, and the loose waving locks produce a
wonderfully natural effect, much more so even than does the hair of the fine
Skopasian head from the southern slope of the Akropolis. This difference in effect is
due to the use' of the drill, which was not employed either in tho Skopasian work
mentioned above, in the Tegea heads, or in any other original sculptures related to
Skopas.1 On the other hand, the hair of the Praxitelean Hermes and of the Eubouleus

Fig. 148. — Profile of Aphrodite in the collection of Lord Leconfield.

is very deeply drilled, a process to which is partly due the effect of life and vigour.
The same method has been employed for the Aphrodite, except that in accordance
with the different character of the hair the tool has been used more sparingly, but the
masterly lightness and sureness of touch are the same.
The hair at the back of the head is, as in the Hermes, indicated rather than
worked out. A breakage behind each ear shows perhaps where the ends of a fillet
hung down, as in the Venus of Aries.
1 Such as the grave-relief from the Ilissos and the sculptures of the Mausoleum. Cf. p. 309 seq.

XVII.
Aphrodite.

collection of lord leconfield.

THE PETWORTH APHRODITE 345
The whole contour of the face, a long oval with harmoniously rounded lines and
high triangular forehead, is at once recognized as Praxitelean by its resemblance to
the Knidian Aphrodite. The round skull and the ear, with its long lobe well
detached from the cheek, its wide opening, and the narrow fold of the cartilage,
exactly resemble the Olympian Hermes.
But the master hand is above all manifest in the surpassing beauty of the eyes,
which are a veritable mirror of the soul. In this respect too only the Hermes
can stand comparison. In both we find the same rounding of the ball and the
same treatment of the lids, which are not sharply detached from the eyeballs ;
the under lid is peculiarly characteristic in its exquisite delicacy, being almost
imperceptibly defined against the ball and the cheek (cf. p. 334). It is drawn slightly
upwards, thus giving to the eye that essentially Aphrodisian expression which,
from the evidence of copies, we already know to have been one of the charms of the
Aphrodite of Knidos. Imitators of Praxiteles carried this tendency to excess — witness
,the small head from Olympia (which however, being only a small, slight decorative
work, comes under another category), with lower lids that appear almost swollen.1
Again, the mouth of the Petworth Aphrodite finds its closest analogy in that of
the Hermes, especially in the delicate transition between the lips and the surrounding
surfaces, which is an unmistakable mark of late Praxitelean art. The fascinating
dimple in the chin reminds us further of the Hermes, while the Eubouleus is recalled
not only by this charming trait but more especially by the treatment of the neck,
the flesh of which is rendered with the softness of nature, but without a trace of feeble
ness. It is so real and living that we are almost inclined to call it the best part
of the bust.
Although we can bring forward no literary witness to help in the identification
of this head, we are, I think, completely justified on technical and stylistic grounds in
maintaining that it is a real Greek original by the artist of the Hermes. We cannot
for an instant suppose that Praxiteles made no statues of Aphrodite besides those
handed down by tradition, or that Pliny gives even an approximately complete
list of the many works by the master which Roman spoliation brought to Italy.
From the carriage of the head and the slight elevation of the shoulder I imagine
that the right arm was raised and supported. The statue was probably draped, and
perhaps it lost popularity and fell out of notice on this very account. Yet it seems
to have been copied ; a head in the Giardino Boboli, Florence,2 and another in
Dresden,3 seem to be replicas, though very poor ones, of the Petworth head.
How strong was the influence exercised by this work of Praxiteles on his
successors is shown in an interesting manner by the Venus of Medici. This type is
merely a translation of the magnificent Praxitelean ideal into a languishing and
coquettish form. The original of the Medici type must have taken an amazingly
firm hold of popular fancy, to judge by the great number of copies and variants in
existence. I think we may safely assign it to the generation after Praxiteles, perhaps
to his sons Kephisodotos and Timarchos.
In proving that the Petworth head is an original we have at the same time
ascertained its date : its special correspondence with the Hermes shows that it
belongs to the artist's later period. That it is somewhat later than the Eubouleus
(cf. p. 334) seems evident from the fact that it shows the technical peculiarities which
1 As remarked above (p. 322), this little head is not a copy of the Aphrodite of Knidos.
2 Diitschke, ii. 80. Has been wrongly placed on a Hera statue.
3 Wrongly attached to the so-called Urania. Hettner, No. 228 (Becker, Augusteum, Taf. 69).
Y Y

346 PRAXITELES
distinguish the Hermes from the Eubouleus, such as the use of the drill for the
corners of the mouth and the ' liquid ' treatment of the lower eyelid. Side by side
with the youthful Hermes we may now claim to have an original female head
from that later period of Praxiteles in which his chisel had attained to the highest
mastery. At this point it is necessary to suspend our inquiries about Praxiteles, which
only profess to lay down the main lines of his artistic development, and to turn our
[To the Hermes of Olympia, the one authenticated original from the hand of Praxiteles, and to
those other two works which on every analogy of form, of style, and of technique— the Eubouleus
(Plate XVI.) and the Leconfield Aphrodite (Plate XVII.)— may also be attributed to the master, I now
wish to add a fourth, a head in the British Museum (Plate XVI IL), known as the ' Aberdeen head,' from
the collection to which it once belonged. The stages by which its Praxitelean authorship has come
to be recognized are sufficiently instructive for a brief account of them to be in place here. The head
is described in the British Museum Guide to the Graeco-Roman Sculptures, Part II. 1876, p. 44 (97),
as ' Head of young Hercules '— ' the type is probably one which prevailed in the Macedonian period.'
It was Dr. Wolters who, in 1886, had the signal merit of redeeming the head from its obscurity, by
adequately publishing it in the Arch, fahrbuch, i. PI. v., and by calling attention {ibid, p. 54) to its
Praxitelean character. It may, however, be added, if only as proof of the rapid progress which
archaeological criticism has made within the last few years, that in 1 886 Wolters still showed that same
lack of discrimination between Skopas and Praxiteles as, to Pliny's bewilderment, did the ancient
critics, since along with the Aberdeen head he published as equally Praxitelean the fine terminal bust
of Herakles from Genzano which Graf, and others after him {supra, p. 296), have since shown to be a
copy of a Herakles by Skopas. In 1891, when on my return from Greece I first turned to the compara
tive study of form, my attention was attracted by the essential similarity which the hair and the eyes
of the Aberdeen head offered to the hair and eyes of the Hermes. In the course of my demonstrations
at the British Museum I pointed to the head, timidly at first, as to an original, probably by Praxiteles.
It was not however till, for lecturing purposes, I had this head and that of the Hermes photographed
side by side on a lantern-slide that I was able definitely to make up my mind. When the two heads,
much enlarged, were projected on the screen, the agreement between them of form and technique was
exact beyond all expectation ; the indubitable likeness made it at once clear to me that the alternative
which I had at times contemplated— namely, that the head might merely be a studio work, traceable
to a pupil of Praxiteles — must be abandoned ; the head could only be by the master himself.
Subsequently I found that this was also the opinion of Professor Klein of Prague (whose book on
Praxiteles will, I have reason to believe, throw much light on this magnificent work) and of Professor
Furtwangler, at whose suggestion I now publish the head afresh, in connexion with his own
researches. Further, Dr. Kalkmann (as I lately learnt in conversation with him) inclines to the same
view. Turning to a comparative analysis of the forms, it is at once obvious that though the hair in the
Aberdeen head is conceived as a lighter, less abundant mass than in the Hermes, the form of the
curls that play about the forehead, especially in the region of the temples, is identical in both heads,
while on the top it is treated in the same broad masses, calculated to produce impression, rather than
to render formal detail. One point, however, should be noted to which, at present, I can instance no
parallel : it is the deeply scooped-out grooves which separate the masses of hair, and which are
doubtless intended to help out the effects of light and shade. In both heads the structure and
modelling of the forehead correspond in their every part ; further, the nose, with the great width
between the eyes, the form of the eyeball, and the modelling of both upper and lower lids, are precisely
similar, even the little furrow indicated between the eyebrow and the eyelid being rendered in both
heads with identical personal touch. Although the cheeks of the Aberdeen head are fuller than those
of the Hermes, the depression in the fleshy region between cheek-bone and jaw-bone is rendered pre
cisely in the same manner as in the Hermes, where the treatment of the cheek contributes in so high
a degree to the dreamy grace and spiritual delicacy of the face (well seen in the fine photograph of
the profile by Rhomaides). The Aberdeen head has fuller, more sensuous lips than the Hermes, but
their form is the same, and the corners of the mouth are, like those not only of the Hermes but also
of the Leconfield Aphrodite, worked with the drill, the marks of which are quite distinct. The chin
has the same full rounded character as the lips ; it is shorter, and has a more marked dimple than
that of the Hevmes. Finally, the close correspondence in the measurements of the two heads cannot

XVIII.
The ' Aberdeen ' Head.
BRITISH MUSEUM.

Printed by Giesecke & Devrient, Leipzig (Germany).

Hermes.

Aberdeen head

112 mm.

. . 108 mm.

129 „

• • 127 .,

i9°i „

i94i „

165 „

162 „

2IO „

• ¦ 202

THE ABERDEEN HEAD 347
attention to one of his contemporaries, who is an imposing figure in the traditions
of the time, though we have so far been unable to obtain a distinct picture of
his art.
be without significance : these have been lately accurately taken and published by Dr. Kalkmann in
his Proportionen des Gesichts in der Griechischen Kunst (53rd Winckelmannsfestprogramm, 1893), on
p. 98, under Nos. 123 and 124, as follows : —
Distance between [outer corners of the] eyes
Eye to chin 
Height of face  ....
Distance between ears . . .
Nose to neck . . 
(Cf. also ibid. p. 109 and text p. 41.)
In looking at the silhouette of the two heads, it must however be owned that the Aberdeen head
presents a much broader outline at the top than does the Hermes. This divergence is, I think, only
superficial, and is solely due to the difference alluded to above in the conception of the hair as a mass.
The ears of the Aberdeen head are much knocked about, which makes comparison difficult, still the
cartilage seems flattened, or rather as it were folded back, and kept close to the head, thus presenting
a variation from the treatment of the ears in the Hermes. The lobe too is rather larger than in the
latter, though the drawing about the hollow of the ear and the large hollow itself are points of
resemblance. Only a small portion of the neck is preserved, but this suffices to show a breadth and
strength which are eminently Praxitelean. The head has a slight inclination to the left, on which side
the muscles are relaxed. The marble of the Aberdeen head is the finest Parian ; in the hair are holes
for the attachment of a metal wreath.
The great likeness notwithstanding, a glance at the Aberdeen head shows that the hero or god it
represents is of a different nature to the Hermes. Kalkmann (loc. cit.) calls the head a ' Praxitelean
Herakles,' while we have seen that in the British Museum guide-book it was called a ' young
Hercules.' Such in effect I believe it to be. The greater emphasis laid on the sensuous physical side
suits this interpretation. Nor is it, I think, fanciful to detect a likeness between this youthful beardless
Herakles and the bearded Herakles Chiaramonti (supra, Fig. 147). When Dr. Wolters considered
that, in spite of resemblances, the ' Aberdeen head ' only proved after all ' wie unendlich hoch der
Hermes in der feinen Durchbildung aller Teile iiber den anderen Werken derselben Zeit steht' — the
difference he detected was, I believe, not one of quality but of conception.
The technique, the slight working of the under lid, the use of the drill for the corners of the
mouth, show that the work belongs to the artist's later period. At the back, on the right-hand side, a
portion of the head has been cut away in a slanting direction ; we have there a tooled surface showing
that the head probably underwent this maltreatment in order to be fitted into a pediment or some
other architectural frame. As regards ihe. provenance of the head, Wolters had well pointed out (loc.
cit.) that from what we knew of the Aberdeen collection it was likely the original had come straight
from Greece ; it is probable that by gaining access to the still unpublished Journals kept by Lord
Aberdeen during his tour in Greece (cf. Life of the Earl of Aberdeen, by the Hon. Sir Arthur
Gordon, p. 10) interesting facts with regard to this head might come to light. Its significance for
our present purpose, however, lies not so much in its provenance or in its interpretation (two points
on which Professor Klein, it can hardly be doubted, will have much to say) as in its formal and
stylistic kinship to the series of Praxitelean works discussed above by Professor Furtwangler. —
E. S.]

348 EUPHRANOR

III. Euphranor.— fmitation of Polykleitos and of tke older Argive Types.— Bonus
Eventus; Dionysos; Apollo Patroos ; Paris and Aphrodite ; Athena.— Lysippos.
Tradition names Euphranor as a celebrated painter as well as sculptor. His
versatility is especially praised, for he distinguished^ himself in every branch of
the plastic arts, and moreover busied himself with their theory as well as with their
practice. Still, nothing is ever said to lead one to suppose that he had a pronounced
individuality. Here, however, as elsewhere, we must guard carefully against trying
to extract too much definite criticism from the statements made by ancient writers
on art. For instance, when Pliny says that in looking at the Paris by Euphranor
the spectator could realize all at once ' the judge of the goddesses, the lover of Helen,
and even the murderer of Achilles,' he is evidently merely quoting a neatly turned
epigram. Critics ambitious to shine have ever shown themselves adepts at reading
the most incredible meanings into works of art ; but their utterances are of no value
for the objective understanding of the subject, and we need not conclude that the
Paris was a psychological conception.1 Nor, again, when a saying by Euphranor
himself is quoted to the effect that his Theseus was fed on meat, but that of
Parrhasios on roses, should we be justified in assuming any special realistic tendency
in the artist.2 The saying possibly arose from some comparison of the two pictures :
in the one the flesh-tints may actually have been rosier, in the other duskier and
less transparent. Pliny, however, makes one significant criticism, belonging to that
series which is almost certainly borrowed from Xenokrates,3 to the effect that the
bodies of Euphranor's figures were too slight, while the heads and limbs were too
large, a fault which he also attributes to Zeuxis. The criticism, like all those of
Xenokrates, refers probably only to some one principal work of the artist, and is
made from a Lysippian standpoint, according to which the heads of all earlier statues
appeared too large for the bodies. It teaches us this much only, that Euphranor, like
Zeuxis, adopted the older, pre-Lysippian proportions. On the other hand, the dis
paraging statement about the disproportionate slightness of the body is of greater
importance, for it implies that the figures of Euphranor's statues must have had
something spare and hard about them, since no critic of the Lysippian school could
ever consider actual slightness of figure as a fault. The statement at any rate
contradicts most emphatically the modern conception of the figures of Euphranor,
which — on the ground of that saying concerning the Theseus, and also because the
artist was said to have well expressed the ' worth of heroes ' — have been imagined as
powerful and of imposing proportions.4 On the contrary, we should rather imagine
the bodies of his statues to have been spare.
Ancient writers date Euphranor by the battle of Mantineia (362 B.C.), an episode
of which — the famous Athenian cavalry engagement — he had painted in the Stoa
Basileios at Athens (Paus. i. 3, 4). It is of course unlikely that so considerable
a public commission would be intrusted to a very young artist, and, since among
Euphranor's bronze works mention is made of figures of Alexander and of Philip
in quadrigis, for which no more suitable occasion could be found than the battle of
Chaironeia (where Alexander had distinguished himself by the side of Philip), it
seems likely that Euphranor's artistic activity lies between 375 and 330 B.C., and that
1 Cf. Brunn, Bayr. Sitzungsber. 1892, 663. 2 Brunn, loc. cit., and K. G. ii. 187^.
1 Cf, Robert, Arch. Mdrchen, p. 68 seq. J Brunn, K. G, ii. 189 seq.

BONUS EVENTUS 349
he was thus throughout his career a contemporary of Praxiteles. His son Sostratos,
according to extant inscriptions,1 worked in Athens and the Peiraieus, and he is
mentioned by Pliny among the artists of the epoch of Alexander.
Euphranor was a native of the Isthmos, i.e. of Corinth. Considering the intimate
relations of Corinth and Athens at the beginning of the fourth century, it is not
surprising that a Corinthian should settle for a time at Athens. Nevertheless, he was
a pupil of the Peloponnesian school. His master was Aristeides, not, as Pliny
incorrectly states, the celebrated painter, but the bronze-worker, himself a pupil of
Polykleitos.2 Therefore Euphranor too belongs to the outer circle of the Polykleitan
school.3 One — and only one — work by him, the Leto carrying her children in her arms,
has so far been doubtfully recognized in copies.* But it has lately been shown on
convincing grounds 6 that the two statuettes in which archaeologists saw copies
after Euphranor really reproduce a work of the severe style of about the middle
of the fifth century, and that Euphranor's Leto consequently still awaits discovery.
We are not, however, entirely without any landmarks to enable us to get some idea
of his style. Among his bronze statues Pliny mentions a Bonus Eventus, holding
a patera in his right hand and ears of corn and poppies in his left. From this
description it has been rightly conjectured that the statue originally represented
Triptolemos, and was therefore of necessity a youthful figure.8 Since it bore the name
of a Roman god, the statue had evidently been brought to Rome.7 A Bonus Eventus
by Praxiteles was also at Rome, together with a Bona Fortuna? In this case the
original Greek names must certainly have been Agathodaimon and Agathe Tyche,
and the former, judging from his received Attic type 9 such as we see it on an Attic
votive relief when he appears in" company with Agathe Tyche}0 must have been
represented something in the manner of Pluto, as a draped bearded figure holding
a cornucopia. Roman Imperial coins, however, from Galba onwards, and likewise gems,11 very
frequently bear the figure of a youth described in the coin legend as a Bonus Eventus
{supra, Plate VI. 37 ; infra, Fig. 149). Now, since this figure corresponds strictly with
the description of Euphranor's work, and is obviously copied from a statue made to
stand alone — a statue moreover which must have been celebrated and in Rome, and
the main points of which are always carefully reproduced — it becomes as certain as
1 Ldwy, /. G. B. 105, 106.
2 Kroker, Gleichnamige Kiinstler, p. 25 sqq. ; Robert, Archdol. Miir. p. 83.
3 It is uncertain whether Sostratos, who is mentioned as the sixth successor of Aristokles of Sikyon, was the
father of Euphranor. In that case, Pantias the Chian, who made statues of Olympian victors, must have been his
brother. This would not be impossible. Lowy on /. G. B. No. 105 also suggested a family connexion between
the elder Sostratos and Euphranor.
4 Schreiber, Apollon Pythoktonos, pp. 70, 88; Overbeck, Gesch. d. Plastik, 3rd ed. ii. 86; Apollon,^. 371 ;
Helbig, Museums, 421.
5 E. Reisch, ' Ein vermeintliches Werk des Euphranor,' in Festgruss aus Innsbruck an die Philologenver-
samml. in Wien, Innsbruck 1893. 6 Cf. Kern, Ath. Mitth. xvi. 25 seq.
7 In Pliny's account it stands between two works whose position in Rome is accurately given.
8 Plin. xxxvi. 23. On the Capitol. 9 Cf. Kern, Ath. Mitth. xvi. 24 seq.
10 Anc. Marbles of the Brit. Mus. xi. 47. The beardless figure carrying a horn of plenty on a votive relief
rom Kyrene (Archaeologia Scotica, iv. PI. 17 ; J. H. S. v. 157, 3) is closely analogous.
11 Some of the more important gems are : carnelian in the British Museum (our Fig. 149), Catal. 929, already
given by Natter, Traiti, PI. 23, also Cades, cl. ii. G, 57, 58. Several instances in Vienna and in Copenhagen
(Abdr. der Dan. Samml. in Berlin 295 — 297), a good emerald-plasma in the Thorwaldsen Museum, No. 610, several
stones in St. Petersburg (Abdr. in Berlin, 20, 16 ; 26, 14 ; 32, 86 ; 35, 47 ; 44, 59), Coll. Schdnberg (Abdr. 127),
etc. Coins : cf. the index of Cohen, Med. Impir. 2nd ed. viii. p. 365, Bonus Eventtis. In the Berlin collection I
have compared pieces of Titus, Hadrian, and Antoninus ; they continuously show exactly the same type as the gems.

350 EUPHRANOR

such things can be that this statue was actually the work of Euphranor. This is
of the highest importance, for an examination of the figure shows that Euphranor
followed that Argive type of which we found the Idolino to be the finest example.
The youth rests on his right leg, and places the other to the side without drawing it
back ; he bends his head to the right, holds the cup in his right hand,
while in the left, which hangs down, he holds a bunch of ears of
corn, or of poppies and corn mixed. The better reproductions reveal
a certain spareness of form which is specially conspicuous in the
Idolino, so that the Bonus Eventus has all the characteristics of style
which tradition ascribes to Euphranor.
Taking this Bonus Eventus as a basis from which to recover the
fig. i4p. standing male figures of Euphranor, we may now venture further, and
Carnelian in the & ° A
British Museum. turn our attention to a small group of figures which reproduce the
Idolino type so popular, as we have seen, in Argive circles of the
fifth century, in a style which savours more of the fourth. Among these by
far the most important is a statue of Dionysos (Fig. 150),1 found at Tivoli. A
number of traits combine to suggest that it must be taken as a copy after
Euphranor. It may even be the very statue of Dionysos by him which epigra-
phical evidence shows to have been at Rome, or at any rate to have been copied
there.2 The god stood exactly in the attitude of the Idolino. The left leg has been
wrongly restored ; 3 it originally was planted sideways with the foot flat on the
ground. The right hand held a cup, the left hangs empty by the side. Imme
diately on its discovery the connexion of the statue with Polykleitos was noticed,4
but too much was made of it,5 and those who attributed it to Polykleitos himself
wandered far from the truth. On the other hand, Michaelis, to whom we are indebted
for an excellent analysis of the statue, has made too little of this Polykleitan influence,
and has certainly not hit on the right solution when he talks of late eclecticism. It is
to be hoped that this is only a last flicker of that unfortunate theory, now practically
extinct, of an independent ' Pasitelean school' — a theory by whose delusive light it
was formerly hoped to illumine all dark places.
The close connexion of this statue with the works of the Polykleitan circle is
absolutely unmistakable, especially in the forehead, eyes, mouth, and profile, and in
the nebris, which is disposed precisely as in the Dionysos of the Polykleitan school
mentioned on page 272. But a comparison with the Polykleitan Dionysos and with
the other undisputed works of the same school shows that we have before us no
production of the actual inner Polykleitan circle. The Dionysos of Tivoli can only
belong to the fourth century,8 for its refined, delicate, and natural modelling is not
found in earlier works. Compare, for instance, the formation of the toes (Michaelis
has already drawn attention to their Praxitelean quality) or the richly and softly
modelled knee with that of the Polykleitan Amazon, where this part is harder and
more simple, or the hair, which in this statue has a much more natural effect than in
the more severely stylistic works of the fourth century. Further, in that Polykleitan
Dionysos there is not the slightest admixture of those feminine forms which, as
1 Mon. d. Inst. xi. 51 ; Ann. 1883, 136 sqq. (Michaelis) ; Thramer in Roscher's Lex. i. 1138 ; Friederichs-
Wolters, 520 ;"^Helbig, Museums, 967. Good copyist work of Hadrian's period.
2 Lowy, /. G. B. 495. It seems to me likely that the inscription stood under a copy rather than under the
original. 3 First pointed out by Petersen, Rdm. Mitth. 1891, 238.
4 Robert, Arch. Ztg. 1882, 137. ¦' See Wolters, loc. cit,
6 Helbig (loc. cit.) is of the same opinion,

DIONYSOS 351
Michaelis has shown, are so noticeable a feature in the Dionysos of Tivoli — as, for
instance, in the glutei and the upper arm.
Notwithstanding this, the bodily formation is still that of the Polykleitan circle, with
which the artist must be connected. Apart from the attitude, the Dionysos resembles
the Idolino in having a less rounded lower abdominal line than is usual in Polykleitan
statues. The way too in which the hair lies close to the skull, and allows its shape to
be seen, is Polykleitan. A resemblance to the Electra at Naples has been noted in

Fig. 150. — Dionysos from Tivoli (Museo delle Terme).
the method of dressing the hair : the likeness consists in the employment of the roll
of hair so characteristic of earlier Argive art.1 In the Electra, however, the roll is
arranged in front of the ears according to the older fashion, while in the Dionysos it
passes behind them, in an original manner. In this hair-roll we can anyhow recognize
the influence of early Argive art.
All these facts fit in admirably with what we have ascertained about Euphranor,
and, supposing we place the Lysippian Apoxyomenos beside this statue, have we not
the best possible illustration of the criticism passed by Xenokrates on Euphranor ?
1 Cf. 50th Berl. Winckelmannsprogr. p. 128 sqq.

352

EUPHRANOR

As a matter of fact, beside the Apoxyomenos the Dionysos looks as if he were all
head and limbs, while in the former the first thing which catches the eye is the torso,
and then the small head and nervous limbs as mere appendages. If traces of
Euphranor are anywhere to be found — and he must be represented among our copies
— this Dionysos is by him.
By its means several other statues are drawn into the group of works by this
master. I formerly thought that the beautiful bronze representing a youthful Apollo

Fig. 151. — Bronze statue of Apollo (Brit. Mus.)

Fig. 152. — Statue in Dresden.

in the Sabouroff Collection 1 was a Peloponnesian work of the fourth century, which,
starting on the lines of the Idolino, was closely connected with the Dionysos of
Tivoli. We can now recognize in it an original work in the style of Euphranor ;
attitude and position are the same, except that the left leg is drawn slightly back
so that the heel is somewhat raised. The Apollo is appreciably younger and
more boyish than the Dionysos, and naturally shows absolutely no admixture of
feminine forms. With these exceptions the likeness of the two figures is remarkable ;
1 Head unfortunately missing ; Saturn. Sabouroff, Pi. 8 — it ; Skulpt. in Berlin, No. I ; Overbeck, Apollon,
p, 227 ; Brunn-Bruckmann, Denim. No. 278.

STATUES OF APOLLO 353
the Argive substratum pierces through in every part : the torso, especially the
front view of it, is rather flat, spare, and dry j1 and the lifelike, finely formed limbs
are in strong contrast with it. The fingers, toes, and the modelling round the knee
recall the Dionysos, and are also related to Praxitelean work. Now that we have
brought the Dionysos into connexion with Euphranor, the historical position of the
Apollo also becomes intelligible.
A bronze statuette, formerly in the collection Greau,2 seems to go back to the
same original as the Sabouroff Apollo, and can thus afford a notion of the face and
hair. A Hellenistic imitation of an original closely connected with the preceding
is preserved in a large bronze statue from Lower Egypt in the British Museum
(Fig. 1 5 1).3 The attitude and the position of the arms are like the Sabouroff" bronze ;
the bodily forms resemble the Dionysos, though the copyist has very much weakened
his model : the head is not bent, the eyes gaze straight in front of them, the hair
is very individual ; it lies close to the skull ; it is confined by a twisted fillet, whence
it falls in loose curls all round the head. A work by Euphranor must have been the
basis for this statue also.
With these is connected a statue at Dresden (Fig. 152),4 with sufficient remains of
the arms and legs to show that the attitude was analogous. It is the figure of a boy,
similar in form to the Sabouroff bronze. In this case too the front of the figure is
rather flat and spare, although remarkable for its softly blended transitions of surface.
The head, which is unbroken, turns naturally to the right, and is slightly bent ; the
hair, which lies close to the head, has something Polykleitan in its arrangement over
the forehead, and at the back escapes in short twisted curls. The interpretation of
this figure is uncertain, since it belongs to no known type. From what has been said,
however, it is a natural step to conjecture in it the Triptolemos {i.e. Bonus Eventus) of
Euphranor, shown to exist on coins and gems, and to restore it with a cup in the
right hand and ears of corn and poppies in the lowered left ; at any rate, it will be
difficult to find a better explanation for this remarkable type.
Further, we now perceive that it was on prototypes in Euphranor's style that the
artists of Hadrian's day based certain portraits of Antinous, which are among the
most celebrated of those extant.5
In this connexion must be mentioned a bronze statuette of Dionysos which
1 Cf. the fuller analysis in the text of Saturn. Sabouroff, p. 3 seq.
2 Frohner, Coll. Griau, Bronzes Ant. No. 913, PI. 20. Now in Berlin; cf. Jahrb. d. Inst. 1886, p. 157 ; Samm.
Sabouroff, vol. ii. Nachtrdge, p. I ; Overbeck, Apollon,^. 229. Mention must be made of a statue in Cassel,
under life-size (ii. 12), in attitude and bodily forms remarkably like this Apollo ; no remains of curls on the
shoulders ; the neck restored ; the head antique but foreign ; the latter is copied from an interesting older
Argive work, with a roll of hair in front and behind. Closely allied to the Sabouroff bronze is another
marble statue, with foreign head and no curls on the- shoulders, in Catajo (Arndt-Bruckmann, Einzelverk.
No. 59).
3 Anc. Marbles of the Brit. Mus. xi. 33 ; Clarac, 482 C, 929 B. From the Coll. Mimaut. Mentioned by
Michaelis, Ann. 1883, 141. The legs are restored, left leg too far drawn back ; the attitude was that of the
Sabouroff bronze, to which the hanging left hand corresponds exactly ; thumb and forefinger, little finger, and hand
are fastened together for security by bars (not the remains of an attribute) ; the right hand differs somewhat from
the Sabouroff bronze ; it appears to have held something upright (perhaps a laurel-twig). The loose curls are
made separately and put on. Eyes of silver.
4 Cat. Hettner, 89 ; Becker, August. Tl. 87 ; Clarac, 809, 2026. The restorations (legs from above the knees,
both arms from the middle of the upper arm) have been removed ; the left arm was lowered, the right probably
extended outwards. Length of face, 0-152 ; distance between the nipples, 0'I94 = height of head. Our illus
tration (Fig. 152) shows the statue in the same characteristic half-back view as the London bronze and the Tivoli
Dionysos. 0 The Capitoline (Friederichs-Wolters, 1659 ; Helbig, Museums, 524) and the Farnese statue in Naples (Mus.
Borb, 6, 58)4 The former most certainly represents Antinous as Hermes,
Z Z

354

EUPHRANOR

has been referred to Praxiteles.1 Though different in conception, and more full
of action than the works just described, it is yet connected with them first of all by a
coiffure similar to that of the small bronze replica of the Sabouroff Apollo, and
secondly by the clear reminiscences of Polykleitos in the formation of the face, the
arrangement of the nebris, which is merely knotted on the one shoulder, like the one
worn by the Tivoli Dionysos, and lastly also by the essential qualities of the bodily
forms. For the rest, the statuette is of Roman workmanship, and is in no sense an
exact copy.2
So far all the male statues referred to Euphranor and to his influence have been
based on the scheme of the Idolino. There exist however copies of a remarkable
work which at first sight seems strange when seen among the preceding ones,
but which is nevertheless closely connected with them, except that it is based on
another and older Argive type, which is no other than the one created by the school of
Hagelaidas, and which, as we saw, was also used in the Polykleitan circle (p. 287 sqq.)
The figure in question represents a youthful nude Apollo ; the best-known copy
is the so-called ' Adonis ' of the Vatican (Fig. 153).3 The identification as Apollo is
confirmed by a number of other copies 4 which have preserved undoubted Apolline
attributes. The god, in complete harmony with the Argive scheme, stands on the
left leg, with the right foot placed with full sole to the side and somewhat in advance
of the other ; he looks with bent head towards the left, and stretches his left forearm
horizontally in front of him, while letting the right hand hang straight down. The
left hand probably held a bow. In one copy the right holds the remains of a strap,
which can only have belonged to a quiver which was dangling from it ; in another copy
the quiver hangs on the tree-trunk beside the left leg, and in yet another the quiver-strap
passes round the chest. From these variations we may safely conclude that the quiver
was not a feature of the original. The notion of making the right hand hold the
quiver by the strap was introduced by the copyist from another older statue of Apollo 5
1 By Milani in Museo ital. di antich. classica, iii. Tav. 7, p. 753 sqq. ; Sal. Reinach (Gaz. des Beaux-Arts,
1891, ii. 265) follows him ; Heron de Villefosse, Bulletin des Musics, 1892, p. 29 ; Michon, Mon. Grecs,
1891—92, p. 2. Cf. p. 337, note 5.
2 In the Louvre, from Italy, supposed to have been found on the Akropolis. The appearance of the bronze
is against this view. The metal has the yellow colour peculiar to the Roman bronzes, more especially to those
found in Gaul. It never occurs, to my knowledge, on Greek bronzes. The patina is, moreover, quite different from
that of the Akropolis bronzes, and the workmanship is of Roman style ; the hair is bluntly worked, the
superficial modelling of the bodily forms lacks insight and refinement. The statuette has altogether been
much overvalued.
3 Helbig, Museums, No. 255, and vol. i. p. 545, for the literature of the subject.
4 Lansdowne House, in the dining-room ; Clarac, 906 C, 476 A. The head is genuine, and wears a laurel-
wreath and the loop of hair which designates Apollo ; the wreath is of course merely an addition of the
copyist. The right leg is wrongly restored, set back. — St. Petersburg, Hermitage, 340; D'Escamps, Marbres
Campana, PI. 36. Head well preserved ; inferior work. In the antique right hand is a fragment of * strap
(wrongly restored as a taenia), which must belong to the quiver-band ; puntelli on the thigh show its continua
tion ; both legs are restored. The left forearm, stretched straight out, is antique, the hand with the diskos
lately restored.— Richmond, in possession of Sir Francis Cook; noted by me in 50th Berl. Winckelmannsprogr.
1890, p. 152, note 92. Trunk with quiver partly antique; head preserved; right hand and left forearm new ;
feet antique ; mediocre work.— Rome, Pal. Chigi : Guattani, 1785, p. 7 ; Clarac, 489, 947 ; Matz-Duhn, 184 ;
Overbeck, Apollon, p. 135, 13. Quiver-band round the breast, not continued over the back. Trunk with snake,
feet and forearm new, the rest antique. Very poor copy, inexact in the hair, effeminate expression. Torso in
Berlin, No. 511, good copy. Torso restored as Hermes, formerly in Villa Casali, now in the Jacobsen Coll. in
Copenhagen, No. 1059 ; right shin wrongly restored, as if drawn back ; good copy. Head without fillet, placed
on the Munich 'Jason' (Brunn, Glypt. No. 151).
5 The Mantua Apollo (cf. 50th Berl. Winckelmannsprogr. p. 141 seq. ) and the later modification in the Museo
Torlonia (Taf. 32, 126 ; Overbeck, Atlas d. K. Taf. 23, 24 ; Apollon, p. 225, 1), and the coins there referred to.

' ADONIS ;

355

very like this one ; in the original of the Adonis-Apollo, however, the right hand
most probably held a laurel-branch.
This statue again is entirely built up on the lines of a fifth-century creation ; nay,
the artist has been so penetrated with the popular canon of Hagelaidas that he has
imported into his own work some slight reflex of its proportions : for example, he has
made the chest too broad for the narrow hips.
But above all he has everywhere retained the
practice of treating the front of the body in
simple, broad planes. The work cannot how
ever belong to the fifth century. The transitions
from plane to plane are soft and smooth, and
the formation of particular parts such as breast
or knee find their closest analogy in works
belonging — like the Thespian Eros and the
' Satyr pouring out wine ' — to the early period
of Praxiteles. The relation of the head to
this series is likewise manifest — its likeness to
that of the Thespian Eros being specially
striking. All this is of weight so far as the date
of the statue is concerned ; but now, passing to
the more delicate personal traits, we find they
all speak for Euphranor. In common with the
bronze Apollo in the British Museum, the
'Adonis' has hair lying close to the skull and
then escaping in a rough tangle of short curls —
a peculiar and original arrangement, not met
with elsewhere in statues of Apollo. Some
copies, though certainly not all, even have the
twisted fillet. Here too, as in the Dionysos of
Tivoli, earlier Peloponnesian influences are at
work. With the customary type this Apollo
has as little in common as the Dionysos. Both
however are closely related in their facial forms ;
they both aim at typical beauty rather than at
spiritual characterization. Both have the same
heavy eyelids, treated after the earlier fashion,
and similar ends of hair falling about the nape.
In face and figure, and even in hair, the
'Adonis-Apollo' resembles the curly-headed
Dresden youth (Fig. 152) still more strongly.
Finally, if we survey the composition as a
whole, we recognize in the Apollo that self
same spirit and temper which pervades the works grouped round Euphranor's name.
The spirit of the old Argive creation lives again, rendered in the softer manner of
a later age. Unoccupied, unconcerned by any inward and personal emotion, these
figures stand absolutely at rest, with heads gently bent, basking as it were in
their own beauty.
The early works of Praxiteles — the Thespian Eros and the Satyr pouring wine1 —
1 In proportions, especially the small head, these are closely related to the Apollo,

Fig. 153. — Apollo ('Adonis'), Vatican.

356 EUPHRANOR
are, as already stated, specially related to this Apollo, and we may conclude that the
two artists were closely connected. Both were influenced by Peloponnesian models.
But the active emotional Attic spirit in Praxiteles can never be disguised ; in spite of
their tranquil motive, those Praxitelean statues are all doing something, even down to
the Thespian Eros, employed in spreading the meshes of his glances.
If the Apollo of the so-called Adonis type belongs to Euphranor, as every
thing seems to indicate, it must, like those works of his contemporary Praxiteles, be
placed in his earlier period, so that the difference between the Apollo and those other
works which form a group with the Dionysos becomes explicable as a difference of
time. In his early period the artist was permeated by the old types created by
Hagelaidas, and fashioned his genius after them. Later, he gave the preference to the
Idolino type, and developed its bodily forms in the direction of softness and roundness,
as shown in the Dionysos and the Sabouroff Apollo.1 The criticism passed by
Xenokrates deals only with these later works.
The Apollo of the ' Adonis ' type must, to judge from the replicas, have been a
well-known and celebrated work. A similar figure resting on the left leg, the right
somewhat advanced, the bow in the left hand, a twig in the right, appears on later
Athenian coins,2 whence it is safe to conjecture that it is a reproduction of one of the
principal statues in the city, such as for instance the statue of Apollo Patroos, the
special god of the Attic race, in his temple on the Agora.3 Now, since the cultus-
statue of the Patroos actually was the work of Euphranor,4 it seems not impossible
that our copies reproduce that very statue.
The style of the Corinthian Euphranor must have been much appreciated in
Athens, since he was commissioned to execute the Patroos. The fact testifies to the
enthusiasm for Peloponnesian art which must have reigned in Athens at that time, and
which enables us to understand still better the early work of Praxiteles. It certainly
must have been a great surprise to the Athenians to see Euphranor represent their
Patroos almost as a boy with a round curly head, and in a tranquil attitude. Yet
the creation found favour and reflected the more effeminate spirit of the age,
even as the powerful, virile Apollo of Kalamis and Myron had reflected the spirit
of theirs. An original work from the Akropolis, unfortunately only a torso,5 bears witness
to the influence of Euphranor on contemporary Attic artists. It too is an Apollo,
with soft, boyish forms, showing in attitude and conformation the influence of the
figure surmised to be the Patroos;6 yet the artist has, in Attic fashion, replaced
the spareness and flatness affected by Euphranor by more softly rounded flesh, and he
has obviously been influenced not only by this master but by Praxiteles. He was,
however, one of the lesser artists ; his work must be dated at about 360 B.C.
The Apollo has its exact counterpart in a Dionysos,7 evidently belonging to the
1 The Adonis-Apollo is flatter in breast and abdomen than these. The fold which separates the pubic
region from the abdomen is carried directly into the sulcus inguinalis as in the Polykleitan statues of boys. In
those later works this is no longer the case.
2 Imhoof-Blumer and Gardner, Num. Comm. PI. CC. xv. xvii. ( = our Plate VI. 31) ; Overbeck, Apollon,
Miinztaf. 4, 33. Cf. p. 196, note 2.
3 Cf. Wachsmuth, Stadt Athen, ii. 418 ; Aristot. 'AB. toX. c. 55, 3 ; Beule (Monn. d'Ath. p. 271
connected the Patroos with these coins.
4 Statues by Kalamis and Leochares stood outside the temple : Paus. i. 3, 4.
5 So-called Elgin ' Eros' in the British Museum ; Friederichs-Wolters, 129 1. With quiver-band ; of course
Apollo, not Eros, who is excluded by the absence of wings.
6 Weight on the left leg, right foot a little drawn back.
7 It is the type of the horned Dionysos, discussed by Amelung, Florentiner Antiken, p. 15 sqq. I had always
supposed the head in Florence (Amelung, pp. 16, 17) to be a variant of the ' Adonis,' but the replicas enumerated

DIONYSOS— HERAKLES— PARIS 357
same period of Euphranor's activity, and of which the head, at any rate, has been pre
served in several copies. We may venture to restore the body on the analogy of that
of the Apollo, to which it must have offered a complete resemblance. Like the Apollo,
this Dionysos differs from the received types in that his thick, short, curly hair, in
which we again find a twisted fillet, lies close to the skull. The modelling of the face
and the self-absorbed expression are extraordinarily like the Apollo. The earlier
works of Praxiteles are also akin to it ; but this master always makes the features
less massive and simple, and gives them a more delicately emotional grace.
It further appears that we possess copies of yet another work by Euphranor, in
which he clung close to the Hagelaidas types. On a term found in Rome in the
seventeenth century, and which afterwards disappeared, ran the inscription 'Hpa/cXy?
'Evcfrpdvopos.1 This proves that a terminal Herakles by Euphranor was well known at
Rome through copies. In the Villa Ludovisi there is a series of terminal figures which
must once have adorned a gymnasium of some kind, and which are copies of works
in divers styles.2 Among them is a Herakles holding a cornucopia, in which I had
already recognized a later adaptation of the Alexikakos of Hagelaidas.3 Since then a
statuette has been found in Rome which in all main points corresponds to the idea
which one would naturally form of the Herakles by the early Argive master, except
that the copyist seems to have replaced the cornucopia on the left arm by the more
familiar lion-skin.4 The Ludovisi term repeats the main features of this type, but in
a later style, which manifests itself less in the body than in the head, and which
seems to correspond to the artistic stage reached by Euphranor. The skin is
here slung round the figure, and knotted on the shoulder just as in the Praxitelean
Herakles. Finally, it is an important confirmation of our results that the only head of Paris,
which judging from copies and later adaptations enjoyed a great reputation, fits so
admirably into our sketch of Euphranor's art — nay, belongs so entirely to the series
of works we have been discussing, that we are justified in thinking that it goes back to
Euphranor's celebrated work. The youthful face in the Phrygian cap 6 has a convinc
ing likeness to the Adonis-Apollo ; and here again the Peloponnesian influence makes
itself felt even in the symmetric arrangement of the curls on the forehead. At the
back the curls are twisted, and specially resemble those of the Dresden Triptolemos.
The original underv/ent adaptations and alterations, but was apparently rarely copied,
by Amelung make it quite certain that it is Dionysos. The variations in the Lateran head (Amelung, p. 22) are
probably assignable to the copyist. Amelung suggested Praxiteles, but the illustrations he gives are peculiarly
instructive, as showing the finer points of difference from Praxiteles.
1 Ldwy, I.G.B. 501. Several such inscriptions are known : there is therefore no reason for doubting the
genuineness of this one.
2 Mon. d. Inst. x. 56, 57. Cf. Helbig, Museums, 861—865 ; Brunn, Bayr. Sitzungsber. 1892, 660.
These terminal figures are of different date and style (cf. p. 248, note 4), but mostly early. Cf. a similar term
from Rhamnus, 'EiJhj/i. dpx. 1891, Taf. 7.
3 In Roscher's Lexikon, i. 2158^.
4 Marble statuette (in the Roman market) preserved intact (only known to me from the photograph) ; attitude
like the Stephanos athlete ; the lowered right hand rests the club on the ground, as in the term ; bearded head
bent to the left ; hair like the Stephanos figure and the Olympian sculptures ; lion- skin on the outstretched
forearm. The skin was probably absent in the original.
5 As the style of this work fixes its date, we can hardly be at a loss about the interpretation. Ganymede
was as yet not represented as a Phrygian, nor presumably was Attis (cf. Samml. Sabmiroff, Text to Taf. 137,
p. 4 and Supp.), but Paris was. A statue in Rome which may be reckoned among the free later modifications
shows by the attributes that it was meant for the shepherd Paris (Bull, della Comm. Arch. 1887, Taf. 2, p. 27).
The seated Paris in the Vatican, conjectured by Helbig (Museums, 188) to be after Euphranor, is in conception and
form of Hellenistico- Roman origin. The figure seems to have been adapted from a painting or a relief. There
are no replicas of it,

358

EUPHRANOR

which accounts for the lack of unanimity shown by the extant examples in the hair
and even in the position of the head. The most complete reproduction of the original
composition seems to be the statue in Lansdowne House (Fig. 154).1 The youth
stands just like the Polykleitan 'Narkissos' {supra, Fig. 115), except that the sides are
reversed and the legs crossed. The right hand is supported on a tree, the left placed
behind the back ; the head looks downwards to the right. The same pose appears in

iu km

1

\ U

^1
I rt 1
1|H
1^B 1
; Bk H
1 Hi|
• 'jeM-'mMm !
Fig. 154. — Paris in Lansdowne House.
Fig. 155. — Torso of Aphrodite (Naples).
a head at Munich,2 a really good copy ; a head in the Villa Albani 3 differs somewhat
in the hair. On the other hand, there is a statue at Copenhagen 4 with a beautiful
head that corresponds fairly closely to the Munich head, though it is inclined in the
opposite direction,5 while the attitude of the body is unaltered, and only a tree-stem is
1 Clarac, 396 E, 664 L ; Michaelis, p. 447, 39. Head broken, but genuine. [Photographed for this edition
by kind permission of Lord Lansdowne. — E. S.]
2 Glyptothek, 135 ; Liitzow, Miinchner Antiken, 27.
1 No. 218. The head is placed on a statue foreign to it. Photo, by Bruckmann.
4 Jacobsen Collection, No. 1052, head unbroken.
5 Also the head in the British Museum (Specimens, ii. 17 ; Friederichs-Wolters, 1580) is bent to the left ; this
agrees in the hair most nearly with the Albani head,
PAN— ATHENA 359
introduced at the back. The pose of the head of the Lansdowne statue is the more
natural, and certainly more likely to be the original one. The body, at any rate in
the Copenhagen copy,1 is closely related to the Apollo, and shows the same cha
racteristic broad smooth masses. The attitude, while influenced by that Polykleitan
prototype, has gone beyond it in the direction of pictorial nonchalance. The complete
self-absorption of the figure would encourage any one with a turn for epigram to find
all possible contrasts united in the figure.
A feminine counterpart of the Paris exists in a nude female torso at Naples
which must have stood precisely in the same attitude (Fig. 155).2 As far as a
comparison of the figures is possible, the stylistic treatment of the body is in all main
points that of the Apollo and the Paris.
A strong family likeness unites with the preceding works another beautiful head,
which has already called forth many a modern psychological description : I refer to the
so-called ' Faun ' (Pan) of Winckelmann (Fig. 156), to whom it once belonged (No. 102
Munich Glypt.) The likeness of the features to the Adonis-Apollo and the youth at
Dresden is most striking. Remains of Polykleitan rendering of form are also plainly
to be seen — for instance, in the close-lying hair, and in the incision at the nape of the
neck. The figure which must have belonged to this enchanting head should be con
ceived after the fashion of the Dresden youth ; the head should be more inclined to
the front ; the eyes look to the right, doubtless to the side of the supporting leg. The
expression of the head, as Brunn rightly states, is one of still unconscious but awaken
ing desire— 'a soft entrancement,' as Winckelmann calls it. Brunn considers the
' Faun ' to be a pendant to the Praxitelean Eros, and as a work by Euphranor this is
actually its historical position.3
Moreover, we can, I think, also distinguish among our copies a dignified goddess
by Euphranor. .None among the extant types of Athena has been more difficult
to explain than the so-called Athena Giustiniani (Fig. 157), of which several replicas
exist.4 It contains a host of peculiarities which at first seem to indicate a fifth-
century work,5 side by side with others which point to the fourth.6 The solution is
easy now that we have learnt to recognize Euphranor's style ; for this statue also
belongs to the series of his works.
Once more the attitude is influenced by the early Hagelaidas type, and is like
the Adonis-Apollo, except that the right foot has less weight thrown on it and is
somewhat drawn back, while the right hand is. supported on the lance. The steady
1 Not on the Lansdowne copy, where the forms are rounder and softer.
2 Friederichs-Wolters, 1468. Cf. Bernouilli, Aphrodite, p. 279 sqq. ; Roscher's Lex. i. 415, Z. 62 sqq. The
left hand was supported against the side, not raised as in the paste compared by Friederichs.
3 Closely analogous to the Pan is a head of a youth in the Jacobsen Collection, No. 1095. Finally, the
so-called Theseus of Ince Blundell Hall should also be mentioned in the present connexion (Spec, of Anc. Sculpt.
ii. 19 ; Arch. Ztg. 1874, Taf- 1; Michaelis, Anc. Marbles, p. 351, 43; cf. p. 94, note 1), the body of which
seems to me very like the Adonis. The attitude is the same. I have, however, not seen the original.
4 I know the following copies : (a) Vatican, Braccio Nuovo, 1 14 ; Helbig, Museums, 5 1. Aegis on the breast,
and snake below ; remains of the sphinx on the helmet are antique. Aegis, snake, and sphinx are copyist's addi
tions, and are absent in b. (b) Capitol, Helbig, Museums, 438. (c) Museo Torlonia, 298. The head and the left hand
with the shield are modern ; with aegis ; an interesting copyist's addition is the olive-tree with the snake (which
here replaces the lance and is mostly antique) touching the sleeve ; the right upper arm is raised higher than in
a and b, the forearm restored. Very late work. Cf. the similar copyist's invention of the laurel-trunk beside the
Mantuan Apollo, (d) Palazzo Pitti, Diitschke, ii. 28. Head modern ; aegis ; poor work, (e) Naples, Inv. 6399 ;
no aegis; head modern, restored as 'Talia.' (/) Good copy of the head in Berlin, 77. (g) Head in St.
Petersburg, Hermitage, 325 ; much worked over ; poor, (h) Head in the Terme in Rome. Poor, (i) Head on
the Cassel replica of the Lemnia (supra, p. 5).
5 Hence my dating in Roscher's Lex. i. 702, 1. 5.
8 Helbig (Museums, 51) refers to an Attic relief of 375—374, which, however, bears only a superficial
resemblance to the statue.

H
"I

ATHENA 361
pressure on the left foot, the turn of the head to this side, and the horizontal projection
of the left forearm are all part of that early scheme. No convenient attribute could
be found for the left hand, for it would have been unsuitable and clumsy to place
something in it when the right hand was already holding the spear ; it only re
mained to make it play with the edge of the cloak.1 The holding up of the lance
involved a variation from the Argive scheme which required the arm to be held
downwards ; the artist managed it by raising the arm as quietly and slightly as
possible, and he arranged the wide sleeves of the Ionic chiton so as to fill the whole of
the triangle between the elbow and the body with vertical lines. Thus the variation
from the pattern type is compensated for and passes almost unnoticed. The uniform
large surface which the broad folds of the Ionic chiton form over the breast is
thoroughly intentional, and contributes to the impression of dignified repose. The
aegis, shown in some copies, is a disturbing addition made by the copyist ; it is
absent in other copies whose evidence should be insisted on ; for copyists, while apt
to introduce popular and typical attributes,2 never left them out if they were to be
found in the original. By leaving out the aegis the artist seems to recall that earlier
type 3 of peaceful Athena which so often appears without this attribute. The folds of
the chiton, on the other hand, only superficially resemble the fifth-century manner ;
their real likeness is to the drapery of the Praxitelean Aphrodite of Kos {supra,
Fig. 138). The folds of the cloak are treated in a simple and dignified fashion ;
there is even a severe touch about the straight fall of the drapery over the left
shoulder. Still there is nothing clear-cut or large about them, and we feel that they
are somewhat conscious and intentional in arrangement.
While all this fits in admirably with the works already referred to Euphranor,
the accordance of the head is still more conclusive. It too has a purely superficial
resemblance — due mainly to the low forehead — to the types of severer style ; but it has
none of their harsh strength, and it finds its real analogy in the Adonis-Apollo and
kindred works, except that it is less kindly in expression, owing to the air of dignified
severity produced by the slightly drooping corners of the mouth. The treatment
of the hair, which contrasts with the strongly stylistic methods of the fifth century,
is related in principle to the Dionysos of Tivoli ; but . in arrangement it is almost
identical with that of the Athena we assigned to Skopas {supra, Fig. 130) ; the same
arrangement, in conjunction with the ram's head on the helmet, was seen in an
Athena head of Praxitelean style (p. 326). Another detail too which definitely shows
the head to belong to this later period is the soft leathern lining showing under the
helmet at the sides and at the back, and which makes its first appearance on Corin
thian coins of the end of the fifth century.4 Lastly, a good extrinsic support to our
statement that this statue is referable to a Corinthian artist, i.e. to Euphranor, is
afforded by the tiny roll-shaped attachments which form the uncommon decoration
of the plumeless helmet ; 6 they are exclusively a characteristic of Athena's helmet on
Corinthian coins of the fourth century, and hardly ever appear elsewhere. Thus
1 This motive, as also the attitude and fall of drapery, has been used by the artist who made the statue 01
Antinoos in the Eleusis Museum.
2 Like snake and olive-tree in a and c.
3 Cf. Roscher's Lex. i. 697, 1. 6 ; Preller-Robert, Griech. Mythol. i. 191, note 3.
4 First in the new, quite free style, Brit. Mus. Catal. Coins, Corinth, PI. 2, 21 ; PI. 3 sqq.
6 In the Berlin copy of the head they are preserved on both sides, just in the place where they appear on the
Corinthian coins. In the original, as the coins show, there must have been a similar roll on the top of the helmet.
The sphinx, shown in the Vatican copy, is certainly an addition of the copyist ; in the Capitoline and Berlin heads
the helmet is smooth above. 3 A

Fig. 157.— Athena Giustiniani (Braccio Nuovo, Vatican).

SOSTRATOS— LYSIPPOS 363
Euphranor took his native Corinthian type of Athena with all its details as the
basis of his creation.
Finally, when we call to mind that there was a celebrated statue of Athena by
Euphranor dedicated below the Capitol 2 which must certainly have been copied, we
shall have no hesitation in considering it to have been the original of the Giustiniani
type. This Athena, like the Apollo Patroos, would belong to Euphranor's earlier period,
even as the Dionysos and the works akin to it belong to the later. The older group is
intimately connected with the earlier works of Praxiteles and also of Skopas. But
Euphranor never got beyond that clinging to the types of the Argive school which
with his two contemporaries was only a transient phase. He was obviously a man
of formulas ; not a strong, fiery nature, but thoughtful and quiet, docilis ac laboriosus,
active and versatile — doubtless capable of exquisite invention, in dependence upon
older types, but evidently lacking the freshness of inspiration or the inexhaustible
fantasy of a Skopas or a Praxiteles.2
It would be interesting to know whether Euphranor ever adopted that roundness
of modelling and that expressive formation of the eye and its adjacent parts which
mark the style of Skopas and Praxiteles in their later period. It seems probable
that Euphranor remained conservative, and it is not likely that he assimilated much
of this newer method. A statue of a youth in the Capitoline Museum 3 is of interest
in this connexion. It is most closely and indubitably connected with the later
creations of Euphranor, such as the Sabouroff Apollo and kindred works. The
attitude is the same {i.e. it is that of the Idolino), so is the position of the arms and of
the head. But the body is modelled in a perfectly round, smooth manner, with soft
undulations of surface ; the head is appreciably smaller, and shows in forehead, eyes,
and the whole formation of the face and hair a style similar to that of the later
Praxitelean or even of Lysippian heads. It is quite plain that we have before us
the last outcome of Euphranor's art, though it is uncertain whether the artist himselt
lived to accomplish this progress. The small head makes this improbable, but his
son Sostratos, a contemporary of Lysippos, might very well have worked in this
manner. A youthful Hermes (Dresden),4 with wings fastened to his fillet, is related
to the Capitoline statue.
The coming dominion of the Lysippian style soon pushed to one side these
offshoots of Euphranor's art. While Euphranor clung to the old Argive types, and
to the less intense Polykleitan creations, Lysippos broke away entirely from these
traditions. He certainly studied the old masters to a certain extent — nay, he called
the Doryphoros his master, and how he adapted it we may still conjecture from the
monuments 5 — yet by giving himself up entirely to the novel, energetic motives which
Skopas had introduced he infused new blood into the stagnating Peloponnesian art.
1 Plin. xxxiv. 77. The statue was called the Minerva Catuliana, as it was set up by A. Lutatius Catulus,
who after the fire of 83 B.C. dedicated the newly rebuilt Capitoline temple with great splendour.
2 Plutarch's information that Euphranor painted the cavalry engagement at Mantineia in a very lifelike
manner is naturally quite reconcilable with this view.
3 Salone 12. Photographs of the head by Bruckmann. Head unbroken ; right arm, left hand, legs, and
support restored. Right supporting leg, left thigh turned slightly out ; the head, with short hair, looks out
quietly towards the right.
4 Hettner, 151 ; Becker, August. PI. 42. Poor execution. Attitude as in statues discussed above ; chlamys
on left shoulder ; head with short hair turned to right ; fillet supporting antique small wings.
6 The Herakles statue of the Museo Chiaramonti, No. 294, shows an interesting combination of the proportions
and forms of the Doryphoros with the Lysippian realistic rendering of form. The pose of the head just differs
slightly, inasmuch as it is turned a little toward the other side (the neck seems to be rightly restored) ; the head

364 EUPHRANOR
It is interesting to compare with the aforementioned Capitoline statue a Lysippian
statue of a youth, at Berlin,1 that tallies with it exactly as regards age and pose.
Although it is in more complete repose than any other Lysippian figure known to me,
yet the comparison forcibly brings out the stronger stir of life in head and limbs.
How much stronger therefore will it not appear in those Lysippian figures, whether of
gods or athletes,2 which are represented in active movement.
Still we must confess that Euphranor infused an inner meaning into the old
Peloponnesian types, whilst Lysippos externalized the inner conceptions of Skopas.
In the art of Lysippos lay the beginning of the pathetic style, which spread to such
an extent in Hellenistic times.
is of the bearded Lysippian type. It seems to me not improbable that Lysippos himself used the canon of Poly
kleitos in this way, adapting it in his own manner. His procedure would be the same as when Rubens copied an
older work.
1 Skulpt. 471. The body, without pubes, is analogous to the youth of the Capitol. The attitude and head
are quite Lysippian. This is doubtless a copy of an athlete by Lysippos.
2 The majestic motive of resting the hand high on the sceptre, seen in the greater gods, was invested with a
pathetic character by Lysippos, who made the hand grasp the staff higher up. Cf. the rock-relief from Alyzia,
where Lysippos worked, Heuzey, Olympe et Acarnan. pi. 12, p. 412 (cf. Roscher's Lex. i. 2173, line 33), and the
same figure used as a symbol on an Athenian coin, Imhoof-Blumer and Gardner, Num. Comm. pi. EE. 13. The
same motive is shown in the large bronze statue (Ant. Denkm. i. 5 ; Helbig, Museums, 965), which, as Rossbach
has recognized, represents Alexander Balas ; I think, however, that the motive of this statue is only borrowed from
the famous Alexander with the spear by Lysippos. As regards athletes in motion by Lysippos, the two statues in
Dresden (Hettner, 245, 246) are worthy of special notice. The beardless head is surprisingly similar to that of the
sandal-binder of the Akropolis (Ath. Mitth. 1886, Taf. 9, 1). A head analogous to the bearded one, but much
more beautiful, is in the Jacobsen collection at Copenhagen (No. 1072) ; it belongs to an athlete in similar motion,
and is one of the finest and most purely Lysippian works in existence. It probably belongs to an earlier period of
the artist than the Apoxyomenos, for the hair is flatter, and worked in an older manner.

THE VENUS OF MILO

THE VENUS OF MILO

B

Y subjecting the famous statue from Melos
in the Louvre to a fresh critical examina
tion, we shall, at the same time, have
occasion for enlarging and extending in more
than one direction the knowledge we have gained
in the preceding essay of the art of Skopas and
Praxiteles. The ' Venus of Milo ' is still a centre of eager
controversy, and only recently a distinguished archae
ologist pronounced the whole question to be an
insoluble riddle.1 Before resigning myself to this
conclusion, I should like to be sure that no means of
solution has been left untried. Since questions of
very great interest in the history of art arise in
connexion with this statue, it will be well to see
what further light can be obtained from a sober and unbiassed examination
both of facts and of tradition.

I. The Lost Inscribed Fragment: Discussion of the Statue's Provenance.
Our first inquiry must be concerning the fragments belonging to the statue.
It is now almost universally acknowledged that a left upper arm and a hand
holding an apple, which were found with the Venus, really belong to it. Even
F. Ravaisson, the author of the most recent reconstruction, admits this fact, although
1 Sal. Reinach, in the Gaz. des Beaux-Arts, 1890, i. 376 sqq. Among the more recent literature on the
subject the following may be mentioned : F. Ravaisson, in the Revue Arch. 1890, t. xv. p. 145 sqq. ; criticism of the
same by Sal. Reinach in the Chronique des Arts, 1890, p. 294 ; F. Ravaisson, La Vinus de Milo (extract from Mini.
de l' Acadimie des Inscr. et Belles -Lettres, tome xxxiv. 1), Paris 1892 ; my review of the same in the Berliner
Philol. Wochenschrift, 1893, 1 107. Less recent works are : Saloman, Die Plinthe der Venus von Milo, 1884 ;
Ldwy, /. G. B. (1885), p. 209 sqq. ; Wolters, Gipsabg. (1885), p. 560 sqq.; W. Henke in Liitzow's Zeitschr. f. bild.
Kunst, 1886, p. 194, 222, 257 ; Overbeck in the Renuntiationsp-ogi-amm der Phil. Facultdt, Leipzig 1887,
Archaol. Miscellen. vi. ' Die Plinthe der A. v. M.' ; Heydemann, Pariser Antiken (1887), p. 5 ; Sal. Reinach,
Esquisses Archiol. 1888; Chron. d' Orient (1888), p. 465 sqq., 6ggseq. ; 1893, p. 35 ; SchreiberinZzta;?-. Central-
Matt, 1888, 1687; Valentin, Kunst, Kit nstler, und Kunstwerke (1889), p. 219, 313; Haberlin, Studien zur
Aphr. von Melos, 1889 ; Kroker in Festschrift fiir Overbeck, 1893, p. 45 (' Die linke Hand der A. v. M.') Finally,
my own short discussion of the question written in 1882 for Roscher's Lexikon, i. 414 seq., embodying part of the
material of the present chapter, but requiring correction in some points. The theory recently put forward by
Mironoff, that the Venus of Milo was a Victory, has been thoroughly disproved by Petersen, Rdm. Mitth. 1894,
p. 91 seq.

368 THE VENUS OF MILO
it is unfavourable to his theory that the figure formed part of a group.1 The left upper
arm must belong to the statue, for it contains a dowel-hole exactly corresponding to
one in the torso. The hand with the apple matches the fragment of arm in scale,
material, and technique, in the traces of weathering, and in the marks of breakage : 2
evidently it must belong to the same figure as the arm. These facts may be regarded
as settled on technical grounds apart from any aesthetic considerations. It is true
that the two fragments show poorer workmanship than the main parts of the statue,3
but this must not be taken as contrary evidence, for in ancient marble statues the
extremities are not unfrequently neglected, especially if they were to be as distant
from the spectator's eye as the left hand of this statue.
The fragment of plinth with the artist's inscription offers a much more difficult
problem, because, as is well known, the original has disappeared. It is not, however,
by ignoring or by concealing it that we can make good this loss, the sin of an earlier
generation. Indeed, the evidence, if fairly weighed, leads inevitably to the conclusion
that the lost inscribed fragment originally belonged to the plinth — a view recently
maintained by Overbeck, by Geskel Saloman, and by H. Heydemann. One ascertained
fact must be borne in mind : when the statue and the separate pieces were brought
to the Louvre and the first attempts (quite unbiassed by preconceived theories) were
made to put them together, it was at once noticed that the inscribed fragment exactly
fitted the breakage on the right side of the plinth, nor did any of the witnesses present
— savants like Clarac, Quatremere de Quincy, and Saint- Victor among them — ever
express the smallest doubt as to this, or suggest that the inscribed fragment did
not fit. It must therefore have appeared quite obvious that the inscribed piece
belonged to the statue. Another question, however, was immediately raised : Did
it form part of the original work, or was it an antique restoration ? They
decided for the latter, but were not agreed as to how this restoration had come
about. Clarac believed quite simply that the inscription gave the name of the real
artist, while Quatremere de Quincy conjectured that the plinth had been mended
with a chance piece of marble which happened to be inscribed.4 Quatremere could
not bring himself to believe that the inscription was that of the real artist, and
therefore invented this far-fetched hypothesis. He would never have done this had
he been able to discover the least imperfection in the fit of the two pieces.
Therefore the theory lately urged, that if the inscription had really fitted it would
have been fastened on and would be there to this day, is absolutely unwarrantable.6
But it is very easy to see why the piece was not fastened on, and why it disappeared.
In the first place, since it was believed to be merely a later addition, it was neglected
as of no consequence for the original composition. Secondly, the late obscure artist
named in the inscription could not possibly, it was supposed, have made the statue.
Thirdly, since the statue was to be presented to the king as a work of Praxiteles (see
1 Ravaisson makes the hand rest on the shoulder of a male figure grouped with the Venus. If this were so,
the short muscle in the palm of the hand would appear flattened by the contact, which is not the case.
2 The restorer Lange's remark (quoted by Ravaisson, V. de Milo, 1892, p. 55) is specially important. He
notices that the traces of weathering on the upper side of the hand are a continuation of those on the upper side
of the arm.
3 Besides this, Henke (loc. cit. ) has shown that there are parts of the torso which are not worked any better
than the hand. So Ravaisson and Kroker (loc. cit.)
4 F. Ravaisson has lately repeated this conjecture ( Vinus de Milo, 1892, p. 52), therefore he too must
believe that the piece belongs to the statue. He does not, however, deign to explain how it happened that the
Mars or Theseus grouped with the Venus came to disappear (without however the hand with the apple that
rested upon his shoulder also disappearing 1), and to be replaced by a little terminal figure with an inscribed basis
picked up anywhere. 5 Sai_ Reinach, Gaz. des Beaux-Arts, 1S90, i. 384.

THE LOST FRAGMENT 369
Adr. de Longperier's famous letter on the subject in Friederichs, Bausteine), it would
naturally be inconvenient to have to affix to it the name of an unknown sculptor.
The disappearance of the inscription, in my opinion, is only a proof of its
genuineness.1 It was an awkward witness, and had to be quietly got out of the way.
A letter from Clarac to Forbin, published by Ravaisson {Vtfnus de Milo, 1871, p. 22),
throws a startling light on the state of affairs in the Louvre at that time. Clarac,
although ' Conservateur du Musee Royal des Antiques,' was excluded from the con
sultation that took place concerning the setting up of the figure, his ' notice ' on the
statue never reached the king, the statue was set up clandestinely and hastily 2  all
this doubtless to suppress Clarac's unwelcome opinion that the inscription, although
added later, gave the real name of the artist.
Now this theory of an antique restoration, which alone prevented Quatremere
and Clarac from seeing that the inscription had belonged to the statue from the
beginning, was quite arbitrary, and undoubtedly false. It arose solely from ignorance
of the technique of ancient marble statuary. Nowadays everybody knows that Greek
originals may often be distinguished from Roman copies by the very fact that they
are pieced together. In the best period artists were very economical with their marble.
The practice of accurately piecing marble together was already very general among
the artists of the second half of the sixth century B.C. At a later date it may be
observed in the Hermes of Praxiteles, and it was still more widespread in the period
after Alexander ; instances are the Nike of Samothrake, some of the Pergamene figures,
the Belvedere torso, and other works of the first and second centuries B.C., men
tioned by Tarral {apud Goler von Ravensburg, Venus v. Milo, p. 37).3 It is among
the last-named works that the technique of the Aphrodite of Melos finds its nearest
parallel.4 Not only were the upper and lower parts of the torso made separately
and then put together, but also the left arm, the left foot, and a piece of the right hip.5
To reject any one of these pieces because it does not suit our theories or taste is
purely arbitrary, and if we carried the method to its logical conclusion we should
accept only one piece as original, while all the rest would be put down as later
additions, a manifest absurdity that no one has yet ventured upon.6 Knowing, then,
that the original statue was made of different pieces, we shall require definite proof
before allowing that any one of the pieces is a restoration.
It is true that the var ous pieces of marble are not all of the same quality, but
this is no proof that they did not originally belong together. In other statues
made up of several blocks it is not unusual to detect strong differences in the marble,
especially between head and torso. Therefore Clarac's assertion that the inscribed
fragment is of a marble ' un peu different ' — ' d'un grain un peu plus gros ' than
the rest of the basis is no argument against its genuineness, especially as it has been
observed that the upper part of the body is made of a different marble from the
lower.7 1 This is rightly emphasized by E. Robinson, Catalogue of the Casts from Greek and Roman Sculpture, Boston
1887, p. 92 ; 1891, p. 253.
2 Ravaisson, loc. cit.
3 Cf. also Saloman, Plinthe, p. 17.
4 Cf. Arch. Ztg. 1881, p. 306, and Collignon in Rev. Archiol. 1888, i. p. 294.
5 The joined piece of the right hip can be clearly seen on the new casts, and in Ravaisson 's Vinus de Milo,
Plates I. and II. 1871. Clarac (Vinus Viclrix, p. 23) and Ravaisson (loc. cit. p. 10) are wrong in supposing it a
restoration. 6 Clarac ( Vinus Viclrix, p. 13) was at first inclined to consider the lower half of the body as a restoration,
but on closer examination he changed his view.
7 Des Cloiseaux apud Ravaisson, Vinus de Milo, 1871, p. 67. 3 B

37o

THE VENUS OF MILO

On the other hand, although the inscribed fragment is lost, the part of the
plinth still extant shows that a piece is missing to which the lost inscribed block
exactly corresponds. There is therefore not the smallest ground for doubting the
testimony of those first eyewitnesses who, in spite of their preconceived theories,
accepted the inscribed block as part of the plinth.
Since casts have recently been taken of the ' Venus ' without the restorations
(Fig. 158), anyone can easily convince themselves that the plinth, like the whole
statue, was put together out of different pieces.1 The block of marble was insufficient,
and the plinth had to be continued
to the right by joining on a piece.
To the right of the existing plinth
there is a slanting contact-surface
worked smooth at the upper edge,
and it is plain also that the piece
adjusted to the right must have been
higher than the rest of the plinth.
Now those first eyewitnesses, and
Debay's drawing, which we still pos
sess (Fig. 159), prove that the lost
piece of marble had a slanting sur
face to its left, exactly corresponding
to the surface on the existing plinth,
and that it was of the desired height.
Geskel Saloman, in his treatise
on the Plinth of the Venus of Milo
(p. 30), described carefully and in
detail the marks on the existing por
tion of the plinth, and my own ex
amination confirms his results in all
essential particulars. Yet his exposi
tion does not quite account for certain
difficulties which arise on closer con
sideration. I therefore consulted the
sculptors who have been working at
-MM. Possenti and Freres — in Berlin,

Fig. 158. — Venus uf Milo (with plinth unrestored).

the reconstruction of the Pergamene frieze-
and with their help arrived at the following results.
Needless to say that Overbeck's restoration, made before the real shape of the
plinth was known, must be wrong,2 because it gives a false distorted turn to the left
foot ; this foot must have projected beyond the edge of the plinth. Tarral's restoration
(Goler von Ravensburg, pi. 4) is also wrong, because it contradicts Debay's drawing
(Fig. 159) by placing the inscribed piece too far to the right. Saloman, Overbeck,
and Ravaisson3 maintained, but incorrectly, that the existing plinth was let into a
second plinth. This was often done in the case of marble statues, but not in this
particular instance. There could not have been a plinth at the back, because the
perpendicular surface of the existing plinth is neatly finished off, and the folds of drapery

1 The supposition of Valentin (Kunst, Kiinstler, und Kunslwerke, p. 321), that the plinth was put together for
the first time in modern times — actually in the Louvre — is refuted by the appearnnce of the plinth.
2 Gesch. d. Plastik, 3rd ed. vol. ii. p. 331.
3 Saloman, p. 36; Ravaisson, V. d. M. 1892, p. 54.

THE LOST FRAGMENT

371

on the left (of the spectator) are worked right down to the ground. The left perpen
dicular face (looked at from the front), like the back of the plinth, is smoothly
finished, and therefore also intended to be visible. For it is well known that the faces
of plinths intended for insertion were left rough. But the front face appears not to
correspond ; it cannot possibly have been visible, nor can the front of the inscribed
block have formed a continuation to it. Something must have been fastened on
in front of it. The manner in which it is prepared with a smooth edge, top and
bottom, is precisely that of a tooled contact-surface, and excludes every other inter
pretation. L To complete the rectangular plan, as Possenti showed me, the lost piece
must have been wedge-shaped (Fig. 1 60).
The inscribed block need not have stood more than a very little farther forward
(£ cm., according to Possenti, meets the technical requirements) than the existing
plinth. Clarac's statement
{Venus Victrix, p. 49), that the
fragment lay ' bien juste dans
I'alignement de la surface an-
terieure de l'ancienne plinthe,'
is therefore not quite correct,
but the mistake is easily under
stood. The deviation of plane
is so slight as to be easily passed
over ; indeed, the left-hand edge
of the inscribed block on which
the thin wedge-shaped piece
joined on in the front abutted,
seems, as Debay's drawing
shows, to have been injured.
Besides, Clarac's remark is not
made in the course of a scien
tific description, but in a rhe
torical period in which he is
attempting to prove, not that the inscribed piece fitted, which nobody doubted, but that
the inscription was there not by accident but by intention, and that it really gave the
artist's name. Clarac's inexactness of wording is shown by another undoubted in
stance in the very same sentence. What he called ' fractures ' are really, as Quatre
mere and Saint- Victor rightly maintained, contact-surfaces. Debay's drawing does
not show the front of the plinth as a junction-surface. It has been drawn smooth,
possibly because it looked better so ; hence it is not surprising that Debay also
overlooked the divergence in plane between the two front surfaces.
If, as seems likely, the wedge-shaped piece put on in front was of the same
marble as the inscribed block, then the whole front of the plinth was originally
uniform in material. Therefore there is nothing in the condition of the extant
fragments against the genuineness of the inscribed block. On the contrary, several
circumstances afford proof positive that the missing piece of plinth must have been

'ANAPOX.-.HNIAOY
X EYZATOMAIANAPOT
ETToiHZCN

Fig. 159. — Drawing by Debay, showing the inscribed block adjusted
to the plinth.

1 Overbeck's assumption (Leipz. Renuntiationsprogr. , loc, cit,), that the plinth was sunk in another up to the
smoothed edge of the front surface, is contradicted by the fact that the back and left side are smoothly finished.
Ravaisson ( V. d. M. 1 892, p. 54) recognized the original existence of a separate piece of marble in front of the
plinth, but by his hypothesis of a large ' fausse plinthe ' he completely ignored the finished state of the back and
left side.

372

THE VENUS OF MILO

just such a piece as Debay's drawing and the evidence of the eyewitnesses attest the
lost inscribed block to have been.
We have already pointed out that the missing piece had a slanting contact-
surface, and must have been higher than the existing plinth. This extra height may
be calculated from the remaining portion. On the outer side of the left foot of the
figure are to be seen {a on Figs. 161, 162) the remains of a small horizontal plane
surface hitherto unnoticed, which must have been continued in the piece of plinth
added on. Now this plane is exactly level with the height of the inscribed block in

ArHI-ANAPoZ.-.HNIAOY
AfcTX'ltJX EYIATOMAIANAPc-Y
E To I HZ E N

so&m.

Fig. 160.— Ground-plan and projection of the plinth, with

restorations indicated.

Debay's drawing ; it is considerably lower than a horizontal plane at the level of the
heel of the left foot would be, as may be easily calculated from the remains of the
ankle visible through the drapery. Hence the foot, the front and larger portion of
which was, as is well known, made of a separate piece, must have sloped down
wards in a slanting direction.1 The small fragment of flat surface a gives the level
where the foot rested, and this is the level of the inscribed block. The ends of the
folds of drapery behind the foot turn up just at the same level, showing that at that
1 Cf. Saloman, Plinthe, 37 ; Overbeck, loc. cit,

RESTORATION OF THE PLINTH

373

point they touched the ground. The actual edge of the drapery is not preserved,
but the smoothly worked contact-s,urface b shows that the missing edge of drapery
was made of a small piece of marble inserted between the torso and the additional
piece of basis. This was done, Possenti informs me, because the edge of the drapery
worked in one piece with the torso would not have afforded a solid enough termination
to the block.
The lost portion of the plinth cannot possibly have formed the support for a male
figure grouped with Aphrodite, as Ravaisson has again recently suggested — first,
because it was higher than the part which supports the female figure, and the man could
not possibly be placed on a higher basis than the woman ; and, secondly, because
it did not reach to the back of the existing plinth, and was therefore only large
enough for a smaller object. This is evident from the appearance of the contact-
surface, which comes to an end before it reaches the back of the plinth (cf. the
side view of the plinth, Fig. 1 6 1, where the slanting hatchings mark the smoothly
worked portion of the contact-surface).
Again, the drapery at the back does not turn
up as in front, where it met the higher ad
ditional piece of plinth, but falls right down
to the lower edge of the plinth. The in
scribed piece drawn by Debay satisfies
these conditions in so far that the rect
angular hole visible on the upper surface
shows distinctly that it was only destined
to support some small pillar-shaped object.
On the other hand, Debay has placed the
line terminating the plinth at the back too
high, probably with the intention of im
proving the appearance of the whole plinth :
evidently for the same reason he also, as is
well known, drew the existing plinth con
siderably higher than it really is.1 In our
Fig. 1 60 the added piece is made to end
at the back where the smoothed edge of
the existing plinth ceases completely ; the end must have sloped down somewhat
at the back, as shown in Fig. 161.
Finally, the ends of folds turning up behind the foot, the remains of the horizontal
surface a, and the vertical surface on the inner side of the left foot2 make it absolutely
certain that the left foot was placed not on a round object but on a flat tectonic
surface like the fragment sketched by Debay.
Since, then, the extant remains of the basis point quite positively to a con
tinuation shaped precisely like the inscribed fragment drawn by Debay, it becomes
impossible even for the most sceptical to doubt the original authenticity of this frag
ment, whose exact correspondence, at the time when it was still extant, was not
disputed even by those who would have welcomed every opportunity for doubt.
The front of the plinth thus consisted of two parts — the longer and lower to the
left, the shorter and higher to the right. On either of these the artist could have cut
1 The correction of this error naturally brings the inscription somewhat nearer the lower edge.
2 Saloman, Plinthe, p. 37, and Overbeck, loc. cit. I should like to add further that the rounded hollow (c on
our sketch) above this vertical edge, which must necessarily have belonged to a tectonic member, is not part of the
horizontal surface of that member, but merely belongs to a fold of drapery.

Fig. 161. — Side view of the extant plinth.

374

THE VENUS OF MILO

his inscription. He chose the right-hand surface, probably because it was of a more
convenient shape, and because it formed the front of a solid block instead of beino-
merely a thin added piece like the left-hand wedge. As a rule, inscriptions which are
cut not on the pedestal but on the work of art itself (as in this case) are found to be
written in a small space in several lines, and are not conspicuously placed. In this
instance the inscription is under the foot, and yet is easily read from below if the
statue be placed at the height that was customary in the third and second centuries B.C.
The Belvedere torso is inscribed between the legs of the figure on the marble seat.
How completely our restoration harmonizes on the one hand with the existing
remains, and on the other with Debay's drawing, can be so easily ascertained by a
comparison of Figs. 158 and 159 — 161 that it need not be further emphasized.
We cannot follow the fortunes of the inscribed block from the time it left Melos
till its arrival in the Louvre, but this is immaterial. The fact is enough that it was
brought to the Louvre with the other pieces of the statue ostensibly from the same
provenance, and that it was at once adjusted to the plinth. It must not, however, be

Fig. 162.— Side view of the plinth. (Restoration indicated by dotted lines.)

forgotten that the suggestion (afterwards so often reiterated as a fact) put forward by
the Vicomte de Marcellus, to the effect that the inscribed fragment was only brought
from Melos by the Marquis de Riviere in November 1820, remains a pure conjecture.
Although years later M. de Marcellus professed to have no recollection of receiving
the inscribed block in May 1820, this is no proof that he did not take it away.
Supposing the lettering to have been much effaced, the Vicomte might very
well mistake the block for a mere shapeless fragment ; and he himself asserts that
he brought away everything that was found with the Venus except the large exedra
inscription. The block in question may very well have been among the ' quelques
socles 'which he received. That the Marquis de Riviere brought it later is a pure
supposition. Dumont d'Urville immediately after the discovery stated, as is well
known, that an illegibly inscribed fragment was found with the statue, and describes it
as the pedestal of one of the terminal figures discovered at the same time. This
means that it had a hollow as if for the support of a term, thus corresponding to the
piece we are now examining. We can hardly doubt, therefore, that the fragment
mentioned by Dumont d'Urville is identical with ours. Again. Voutier's drawing

PLACE OF DISCOVERY 375
published by Ravaisson,1 and said to have been made in Melos before the statue was
shipped, i.e. before May 24, 1820, contains the inscribed piece under discussion. It
is drawn as ' piedestal d'un des hermes.' It is evident that Voutier's drawing repre
sents the same fragment as that described by Dumont, and that this fragment is the
inscribed block drawn by Debay.2 To connect the inscription with a terminal figure
is, of course, a purely arbitrary hypothesis, suggested at the time of the discovery, and
adopted by Dumont and Voutier. We shall attach but little value to Voutier's
evidence if we bear in mind that he restored the plinth of the Venus as a complete
rectangle. He also drew an inscribed basis 3 under the second term, but this basis
plainly shows by its section of profile that it could have had nothing to do with the
term. It was simply a stray fragment of pedestal with a hole for insertion. Voutier's
drawing, then, is valuable only as evidence that the inscription was found with the
statue, but his combinations of inscriptions and terminal figures are quite arbitrary.
A few words must here be added about the character of the spot where the
statue was found, though this is really a minor consideration. According to Salomon
Reinach,4 this was a limestone-burner's kiln or workshop {magasin de chauxfournier),
where bits of refuse marble of all kinds were collected for burning. If this statement
were correct, it would still be no evidence against the genuineness of the inscribed block.
But I cannot admit that it is correct. Dumont d'Urville, an eyewitness, speaks of ' une
espece de niche,' 5 and of hewn stones found first by a peasant. He also says that an
inscribed marble block was found over the opening to the niche {entre'e de la niche).
Clarac speaks of a ' niche carree d'environ quatre pieds de large.' In Olympia I have
often helped to excavate limestone kilns where had been burnt precious works of
statuary, which however had always been previously broken up into small pieces. But
not even the most ignorant person could ever mistake such a limekiln for an archi
tectural ' niche.' Reinach's supposition has no foundation in fact. The contents of
the ' niche ' consisted, not of a mass of odd fragments of a number of different statues,
but simply of the pieces of the Venus and of the small terms (such terminal figures
being, as is well known, a favourite decoration of niches) and of an odd hand. A foot
' chausse d'un cothurne ' was found, according to Dumont, ' en meme temps,' but,
according to Marcellus, not in the same place, but lower down near tbe sepulchral
caves. This foot is identical with the left foot with a sandal, of smaller proportions
than the Venus mentioned by Clarac on p. 24 of his work.6 In November two
clumsy arms were brought to the Marquis de Riviere, as coming from the spot
where the statue lay — a statement which is naturally not very reliable. It is
evident that these premises are insufficient to warrant the theory of ' un magasin de
chauxfournier.' My own opinion is that the statue was found in situ. I base it solely on
the evidence of the earliest informants as given above, leaving entirely out
of the question the statement made long after the discovery, and already dis
proved by Ravaisson, that the statue was found intact.7 A square niche with an
inscribed block over it was evidently what the discoverers saw. The niche
1 V. de M. 1892, PI. 2 ; p. 10, 51. Cf. Berl. Phil. Wochenschr. 1893, 1108.
° Lowy's objections (loc.cit. p. 112) are not serious. Debay may have drawn the inscription from a transcript,
and the original may very well have been illegible to the unpractised.
3 Mentioned nowhere else. 4 Gaz. d. Beaux- Arts, loc. cit. p. 382.
5 Brest (20 Nov. 1820) also calls the spot a ' niche.'
6 The foot must still be in the Louvre, though Ravaisson ( V. de M. p. 54) says it cannot be found there.
' Cf. Ravaisson, V. d. M. 1892, p. 26 sqq. It is needless to discuss the foolish nonsense contribute d by
H. Rochefort to the Art Franc ais of Jan. 21, 1893, and quoted in extenso in Arch. Anz. 1893, p. 27 (cf.
Sal. Reinach, ChroniquecT Orient, 1893, p. 35 ; Berl. Phil. Wochenschr., loc. cit.)

376 THE VENUS OF MILO
contained the statue and two terminal figures,1 and, as far as is known, no other
fragments of sculpture. Clarac says (p. 25) that the block, from its dimensions, seemed
to have been placed above the niche. He evidently considered Dumont's statement
about the position of the block an inference, but a correct inference, and we have no
reason for holding a different opinion. Morey's assertion that the statue was found in a
sepulchral grotto partly hewn out of the rock 2 found great favour, but had no foundation
other than statements of the Melos peasants, who were only too ready to show the
English traveller (who visited the island eighteen years after the discovery) any
cave which they thought would satisfy his curiosity. The real niche in question,
bein°- in their arable land, was probably destroyed very soon after its contents were
removed. It is not the late statement of Morey, but the first authentic records,
which can serve as evidence, for we know how quickly ancient remains disappear if
the o-round where they lie is cultivated. Now the place where the statue was found
was in cultivated soil, above the caves. These are hewn in the rocks of the valley
called Klima, which slopes down towards the sea. Ross has given a vivid description
of these in his fnselreisen (3, p. 9). No trustworthy record exists of any remains of
sculpture which the caves may have contained. The higher ground, on the other
hand, especially the site of the ancient town of Melos, was peculiarly rich in remains.
In the immediate neighbourhood of the place where the Venus was discovered there
were found some years later a statue of Hermes bearing the artist's inscription, seven
fragments of another statue, and the inscription C. I. G. (Bockh) 2431, to which I shall
have to refer a°-ain.3 Recently the statue of a boxer has been discovered at the same
spot as the Venus.4 In a ' grotte ' within the town ruins, but not localized more
accurately, has been found, along with innumerable votive gifts to Asklepios and
Hygieia, the superb Asklepios head now in the British Museum.5 In the lower
part of the city near the sea, ' dans la localite dite Klima,' 6 several well-preserved
statues now in the Central Museum, Athens, recently came to light.7 These probably
stood in a sanctuary of Poseidon, to whom one of them bears a dedicatory inscrip
tion. Another is a colossal figure of the god himself,8 of peculiar interest as affording
the closest analogy to the Venus. It is nude above and draped below, and consists
of two blocks of slightly different marble joined just within the upper roll of drapery,
exactly as in the Venus. The treatment of the folds, especially on the right leg, which
supports the weight of the body, is strikingly similar. No one has ever maintained
that the Poseidon is earlier than the Hellenistic period.
The inscription on the block above the niche reads ° : —
Ba/rwo? %aTiov10 VTroyv[pvaaiap^r]ij]a<i Tav ts e^eBpav teal to / ...'Epfiut, 'Hpatckei.
1 Tw o according to Dumont, Brest, and Voutier, yet three came to the Louvre, and Marcellus also speaks of
three ; the third was probably found later. Cf. infra, p. 377, and Ravaisson, V. de M. p. 38.
2 The assumption so often repeated latterly, that the statue was found in a hiding-place, is purely arbitrary.
Even Ravaisson (loc. cit.) speaks of a ' cachette.' It is true that Le Blant (in Milanges d'Archiologie et d'Hist. x.
(1890) p. 389^^.) had shown from literary sources that in the fifth century A.D. statues of the gods were sometimes
concealed from the Christians. That the Hercules Mastai and the Capitoline Venus were found in such hiding-
places seems certainly correct, but it is quite false to assert this of the Venus of Milo. Le Blant himself (p. 394)
seems to doubt it. The original report on the find of the Melian statue makes it quite impossible ; a niche with
a dedicatory inscription above the entrance, and adorned with terms, is scarcely a ' hiding-place.'
3 Brest in Ann. d. Inst. 1830, 195. The Hermes is now in Berlin. For the inscription belonging to it see
Ldwy, /. G. B. No. 354. 4 Rev. des Etudes Grecques, 1891, 192.
5 Annali, 1829, p. 341. Cf. Wolters, Ath. Mitth. xvii. p. 7. 6 Bull, de Corr. Hell. 1889, p. 498.
7 Kabbadias, "/Xvirrd rov iBv. povtretov, Nos. 235 — 238.
3 Bull, de Corr. Hell. 1889, PI. 3 ; Kabbadias, No. 235.
0 Clarac, Inscript. du Louvre, PI. 54, 441 (No. 802), p. 853 ; Bockh, C. L. G. 2430.
10 Clarac's reading is ~i.(e\rov) 'Artov.

PLACE OF DISCOVERY 377
The 'exedra' is doubtless the niche itself, dedicated by a sub-gymnasiarch to the
special divinities of the palaestra, Hermes and Herakles. Kal to j ... (the slanting
stroke belongs to an A or A) is most naturally restored as teal to d[ya\ua — ,
and this can only refer to the Venus. The name of the goddess may have followed
the word aya\p.a. It would be quite in accordance with usage 1 for a gymnasiarch
on leaving office to dedicate a niche to Herakles and Hermes, the two patrons
of gymnasia, but there was no reason why he should not include among his votive
gifts images of other divinities. We have already noted that in the immediate neigh
bourhood of the niche were found the inscription {C. f. G. 2431) — again a dedica
tion to Hermes and Herakles — a statue of Hermes, and the fragments of another
figure of which nothing is known : the recently discovered boxer also comes from
the same spot. All this points to the conclusion that we have here remains of the
decorations of a gymnasium, and not a chance heap of marble fragments.2 The small
terminal figures found with the Venus, and which, according to Clarac ( Venus Victrix,
p. 6), were leaning against the posterior wall of the niche, are quite in place here.
At first only two were found, one bearded and one beardless. These were drawn by
Voutier, and described by Frohner {Notice de la Sculpture Antique, Nos. 194 and 209).
The shafts are in good preservation. A third (Frohner, 195) was found later, and was
also brought to the Louvre. The shaft was broken and is restored. All three are of
somewhat different proportions ; 3 but in general decorative effect,4 in marble, and in
style, they harmonize so closely that it seems reasonable to suppose that they orna
mented one and the same enclosure. No. 209 represents the old bearded type of
Hermes, No. 194 presumably Hermes as a youthful athlete, No. 195 the young
Herakles. Probably there were originally more of these decorative terms, but their
importance as votive gifts must have been small compared to that of the large statue.
The workmanship is poorer than that of the Venus, but may easily be of the same
date : it is certainly pre-Imperial.5
We are, however, justified in asking whether the dedicatory inscription may
not belong to a later date than the artist's signature. The block with the dedi
catory inscription has likewise disappeared (another inconvenient witness hushed
up !), but we fortunately have a fairly exact copy by Clarac. Both inscriptions are
written in the same characters, except that the dedicatory inscription has Pi with
the second stroke long, a form which we know from the Pergamon inscriptions to
have been in use at the end of the third century B.C. along with the older form with
the shorter second stroke.6 We therefore have no reason for dating the dedicatory
inscription later than the artist's. The character of the epigraphy allows us to place
either inscription anywhere between 200 B.C. and the Christian era, but their nearest
1 Cf. Pergamon, viii. Inscr. No. 9, dedication to Hermes by a gymnasiarch, C. I. A. iii. 105. Do. 123 to
Hermes and Herakles. Also Bull. Corr. Hell. 1891, p. 251 sqq. for several similar inscriptions from the
gymnasium in Delos. The usual votive offering seems to have been a terminal figure.
2 Cf. Saloman, La Statue de Milo, p. 20 seq.
3 The exact measurements, communicated to me by M. Michon, are as follows : No. 194 : whole height
(exclusive of the modern basis) = ri53. Height of bust = o-30, of head = o'i7 ; breadth of shaft = o-i7, depth =
0-14. — No. 209 : whole height (exclusive of the modern basis) = I -26 ; height of bust = 0-31, of the head down to
the (modem) point of the beard = o-24; breadth of the shaft = 0'i9S ; depth = o-i6.—No. 195 : whole height of
ancient portion = o745 ; height of bust = o-32 ; of head = o-i85 ; breadth of shaft = o-i75, depth = o-i45. The
phallos was attached somewhat differently in each.
4 The height of the arm-holes (7 cm. ) and the distance from the head to the terminal shaft is the same in all three.
5 The edge of the eyelids is smoothed off, the eyeball is almost concave instead of convex— peculiarities which
no longer occur in Imperial times.
6 See Pergamon, viii. Inscr. Nos. 33—37 (Attalos I.), 47, 51, 53 (about 200 B.C.), 58, 169, 236, 239, 240-2
(end of empire). For the different forms of pi in Attic records see Dittenberger, Arch. Ztg. 1876, p. 139.
3 c

378 THE VENUS OF MILO
parallels are found from 150 to 50 B.C. For more exact dating we must rely on a
comparison of styles.
It is interesting to compare the quadrangular niche in which the Venus was dis
covered with the remains of niches found ranged round a court which served as the
place of assembly for the Italic colony on the Island of Delos. These niches x con
tained either benches — in which case they were genuine egiSpai, — or statues offered
by various persons at different times, and the whole building was dedicated to Hermes-
Mercury, the patron of the colony. The building with its sculpture may be dated
about the end of the second century B.C., the very period to which we have assigned
the exedra of the Gymnasium of Melos.
That court in Delos had probably been copied from a gymnasium. Several
constructions of this kind serving as gymnasia or palaestrae, and dating from
Hellenistic or Roman times, have been preserved. The characteristic portion of the
design is a square, pillared courtyard surrounded by a wall containing rectangular or
rounded niches {exedrai) once adorned by statues. Such are the Gymnasium of
Hadrian in Athens {Upa/cTitcd tt}? dp%. eTaipla<;, 1885, inv. i.), the Gymnasium in
Delos {Bull, de Corr. Hell. 1891, p. 238), and the Palaestra at Olympia. For others
in Perge, Side, and elsewhere see Petersen in Lanckoronski, Stddte in Pamphylien
und Pisidien, i. p. 41, 134. It is obvious how excellently the Melian finds fit into
this series ; the further fact that the theatre is close to the spot accords admirably
with our theory (cf. Petersen, loc. cit. p. 1 34 seq.)
It has been a somewhat difficult task to work our way back to the actual cir
cumstances of the discovery of the ' Venus of Milo.' The facts themselves are so
simple that it is astonishing they should ever have been misunderstood. It was the
blind and prejudiced notion of the time that this statue was a unique and unrivalled
treasure, the work of Pheidias or of Praxiteles, something quite isolated from, and
independent of, historical limitations. Ignorance of marble technique fostered this
fancy, for the separate pieces which did not tally with a preconceived ideal were
ruthlessly rejected as later additions. When an opinion has once taken hold it is
extraordinarily difficult to eradicate it, however flatly an unprejudiced examination
of fact may contradict it. II. Restoration of the Statue.
Having ascertained what parts of the statue are in existence, wc must now pro
ceed to restoration of the missing parts. The square hole in the inscribed piece of
plinth evidently held something like a pillar. This object must have stood fairly high,
for the left side of the figure is less carefully worked than the right, evidently because
it was to be partially concealed. Dumont d'Urville and Voutier, as we have seen,
supposed the inscribed block to be the pedestal of one of the terms. Quatremere de
Quincy says that the hole in the block corresponded in width to the socket of one of
the three terms, and Clarac remarks ( Venus Victifx, p. 38) that possibly this figure
may have been set in the block on the occasion of a restoration. Nothing is said to
show that it fitted exactly, hence there is no evidence for the view, upheld chiefly
by Tarral and G. Saloman, that one of the beardless terms belonged to the in
scribed block. The approximate correspondence in size might be purely accidental,
and there are various circumstances that tell directly against the Tarral-Saloman
theory; for instance, the workmanship of the terminal figure is much poorer than that
1 Homolle in Bull, de Corr. Hell. v. 390 sqq. viii. 113.

RESTORATION 379
of the statue, and the material, a common bluish marble, is inferior. The artistic
effect produced by a conjunction of statue and term is distinctly unpleasing, as has
sufficiently been pointed out by others,1 and as Tarral's restoration shows at a glance.
It would moreover be not only a hideous but an unparalleled device to place
a term as support to a figure with which it had no definite connexion.2 Overbeck's
notion of a pillar on which the shield stands must be rejected, because we know
that the left hand held an apple. The same fact speaks against Heydemann's hypo
thesis,3 that the goddess is decking a trophy. For such an occupation she must have
both hands free, and not be negligently holding the apple with her left : besides, the
ancient tropaion was always round, not rectangular, as must have been the shaft inserted
in this hole.
Let us for once simply follow the clue given by the shape of the hole : we shall
immediately see that a pillar must be restored here. We must, however, find a motive
for this pillar. This motive could only be to serve as a support : Aphrodite was
resting the elbow of her left arm on a pillar. This is, in fact, the solution of all the
difficulties. Others4 have already observed that the whole attitude demands a
support on the left side. The biceps of the left arm is not tense, as it would be were
the arm held freely. For the rest, it can easily be gathered from the restorations
that show the arm raised free into the air how constrained and unnatural, how ugly
and angular, this position would be.
Owing to the fact that in both torso and arm there are distinct traces of
the hole for the large dowel which once connected the two, the position of the
upper arm can be settled beyond dispute.5 The arm ought to be moderately raised
to the side ; this is proved by the edge of the fracture on the left armpit. The
convexity below prevents the arm being lifted higher. The direction of the forearm
is absolutely certain from the direction of the dowel-hole in the upper arm. Possenti's
attempt at restoration proves that if the dowel-hole in the upper arm be rightly
connected with that in the body, the biceps of the upper arm would be turned not
upwards but forwards, and that the forearm, which, as the muscles of the upper arm
show, was bent nearly at a right angle,6 would also be directed not upwards but for
wards. Earlier attempts at restoration have all been incorrect, because they did not
take into account the direction of the dowel-hole. Lastly, the position of the left hand
must be fixed. It did not hang down, but was turned upwards, as Possenti rightly
remarks, because the swelling on the upper arm at the elbow points to this position of
the lower arm. This also explains the position of the apple, which was held by the
thumb, the third and the fourth finger, while the first and second fingers were gracefully
extended.7 Now, if the hand had hung down, the apple must have been grasped more
firmly. This position of the apple is natural only if it rested on the upturned palm
of the hand. It is easy to see why the back of the hand is so roughly worked. It
1 Cf. Heydemann, Pariser Antiken, p. 7.
2 Certain terra-cottas derived from pictures (e.g. Samm. Sabouroff, Taf. 84) fall under a different heading from
large statuary compositions.
3 Other archaeologists take no account of the hole in the basis, though Fried. Kiel, to be sure (Die Venus von
Milo, Hanover 1882, p. 32) imagines a lance fixed in the hole and held by the goddess with both hands — i.e.
Venus as giantess with a spear as thick as a tree 1
4 Cf. Ravaisson, Rev. Arch. 1890, xv. 148 ; Vinus de Milo, p. 56.
5 A portion of the smooth surface of junction may still be seen on the fragment of arm. On the torso the
corresponding surface has been broken away, but the piece may be restored with absolute certainty by aid of the
dowel-holes. 6 Cf. Henke in Liitzow's Zeitschr. f. Bild. Kunst, 1886, p. 198.
7 Cf. Kroker, loc. cit. Kroker's notion that the first and second finger must have held something — and that
this was the edge of the shield — is in itself sufficiently unsatisfactory. It will be seen to be quite impossible, from
the proofs adduced above as to the real position of the arm.

38o

THE VENUS OF MILO

was turned downwards, and was therefore not much seen. Fig. 163 gives a sketch
of a restoration proposed by Possenti, which is carried out
in plaster, and illustrates what I have just said.
The arm thus raised and supported under the elbow
forms a not ungraceful motive — at any rate, it has many
analogies. A figure of Hermes on a vase (Millin, Vases
Peints, ii. 20) may first be mentioned ; it is evidently
influenced by sculpture, and the arm is supported on a
tall slender pillar in the way supposed for the Venus.1
The vase appears to be of fourth-century Attic manufac
ture. Still more instructive is a series of terra-cotta statuettes
of the fourth century B.C. representing Aphrodite half-
draped like the Melian statue, the elbow supported on
a pillar and the left arm raised. Further, Aphrodite in
a similar attitude often occurs on gems of the Hellenistic
period (Fig. 164).2 A beautiful bronze statuette of the
nude Aphrodite (Dresden, Fig. 165) 3 reproduces a motive
which is still commoner in terra-cottas. The forearm is
not raised,4 but bent forward in the position proved for
the Aphrodite of Melos. The left hand— and this is
specially interesting in the present connexion — hangs loosely
from the wrist holding an apple. The support was in
a separate piece arid has disappeared, but the whole atti
tude makes it quite certain that it was originally there.5 Pillar supports arc
of very common occurrence in ancient statuary. Pheidias himself had given
one to the Parthenos, though it was purely technical, and not, as here, part of the
composition. In further confirmation of our restoration we may note two late
Roman marble variants of the motive of the Melian statue, in both of
which are to be seen traces of a support under the left arm. These are
a statuette in Treves, and another in Dresden. The traces are less
distinct in the Treves statuette,6 but in the Dresden example 7 (Fig.
166) is still to be seen the end of a rectangular pillar on a low base.
This statuette is, however, in no sense a replica of the Venus of Milo.
Probably the pillar supported an attribute — possibly a lyre s — with
which the left hand was occupied, and towards which the right hand
also was directed.

Fig. 1

Proposed restoration

of the Venus of Milo.

Fig. 164.— Aphrodite
on a gem (Berlin).

1 Published examples : Dumont-Chaplain, Ciramique Gr. ii. 28, 2, apparently identical with Arch. Anz. 1S91,
p. 121, 10. See also Arch. Anz., loc. cit. p. 22, 6, and Samm. Sabouroff, Taf. 132 below, to the left.
2 Brown convex paste (Berlin, Inv. p. 581). The gem belongs to a distinct class, which is most certainly
Hellenistic. In the right hand is a leaf-fan, the left hand seems to have held the end of the cloak (Fig. 164).
For an analogous Hellenistic gem cf. King, Ancient Gems and Rings, ii. PI. 23, B. 5.
3 Basis and pillar modern. Pre- Roman.
4 Berlin Antiq. TC. 7794. Half-nude Aphrodite ; left upper arm raised and supported on pillar, forearm
hanging down.
5 Cf. the Apollo bronze, Arch. Anz. 1889, p. 105, where the left arm is held as in the Melian statue, and
presupposes a similar support.
6 Flasch, Jahrb. d. Ver. v. Alterthumsfr. im Rheinl. Heft 62, Taf. 2, p. 74 sqq. ; Hettner, Die Rdm. Stein-
denkmaler in Trier, No. 684. Flasch's interpretation of the figure as a Hygieia seems very uncertain to Hettner,
who inclines to think it represents a Victory setting up a trophy.
7 Half life-size figure, Hettner, No. 174 ; published with the old restorations (now removed), Le Plat, 124,
and Clarac, 595, 1301. Poor, late workmanship.
8 Hettner suggests Terpsichore. Cf. Clarau, 481, 959 B.

TYCHE OF MELOS 301
I had carried my inquiry to this point, when I first noticed what is the strongest
proof of my restoration, and what at the same time seems to place it beyond all
uncertainty — namely, that the highly revered cultus-statue of Melos, the Tyche of the
island, was represented in the same attitude as the one which we are compelled to
suppose was that of the Melian statue. This type is known to us from Imperial
bronze coins of the island, of which there are three examples in Berlin (see Fig. 167,
with the legend TTXH),1 and from a relief on the drum of a column found, along
with its counterpart which reproduces an archaic cultus-image of Athena, in Melos,

Mm
is^HI
"i"
m H
.; MB Me.
¦
¦ .- rmf!i \\j
y.ffwu ¦f KS
¦¦-^aKair
Fig. 165. — Bronze Aphrodite (Dresden).
near the theatre (Fig. 168).2 The goddess supports her right arm on a simple pillar
or column in the position ascertained for the left arm of the Aphrodite, except that
in the Tyche the forearm too is raised, in keeping with the solemn attitude of the
goddess. It is not impossible that, like our Venus, the original statue held the apple,
which as the symbol of the island would be a singularly appropriate attribute.
It is at once obvious that the motive of the supported arm must have been
transferred from the Tyche to the Venus, and that the reverse was not the case,
1 The three examples in Berlin all have the pillar as support and the child raising its arms to the right. The
description in Imhoof-Blumer, Griech. Miinzen (Abh. Bayr. Akad. i. Cl. xviii. 3), p. 547, should be rectified
accordingly. " Cf. Wolters in Ath. Mitth. xv. 248.
3-S2

THE VENUS OF MILO

for the position is quite in harmony with the calm dignity of Tyche; she holds
the infant Ploutos on her left arm, and with a festal, solemn air rests her left elbow
upon the pillar. On the other hand, the motive is so
little suited to the animated pose of the Aphrodite that
it must have been borrowed. The statue of Tyche,
although, to judge by the drapery, not earlier than
the middle of the fourth century B.C., is most pro
bably older than the Aphrodite. Hence we may con
clude that the motive belonged originally to Tyche,
and was transferred to Aphrodite, although the artist
altered the forearm and did not represent it raised.
This transference seems only natural when we remember
the close connexion of the two divinities in Hellenistic
times, especially among the Asiatic Greeks.1 In my
view, the artist wished to characterize the Aphrodite
of the Gymnasium as goddess of Good Luck, and
therefore gave her the pose of the Tyche of the
city. If, as I suppose, Tyche held the apple, the trans
ference was still more fitting, for the apple was the
peculiar attribute of Aphrodite.2 A remarkable coin
of Melos of Imperial date proves that the Tyche type
was used even for a male figure personifying the Good
Luck of the city.3
The preceding considerations not only justify our restoration of the left arm
with the support, but have given us fresh material for deciding the question of
' originality.' Our statue cannot be an ' original ' in the usual sense of the word, since
the artist who designed it borrowed, for purely external and
not artistic reasons, a motive foreign to the central conception.
It is necessary to bear this in mind in our attempts to restore
the right arm. We shall no longer demand (as those critics
have done who took it for granted that they were dealing with
an absolutely original work) a perfect correspondence between
the motive of right and left hand. The discovery of the left
hand holding the apple was in itself enough to exclude the
idea of such correspondence.
The strong tension of the upper part of the right arm
can be explained only by supposing that the hand was extended
downwards across the body towards some point a little distance from it. This point
can only be the drapery on the left thigh. Most restorations make the arm too
short.4 The hand could easily reach the lower edge of the mantle falling about

Fig. 166. — Statue in Dresden.

Fig. 167. — Bronze coin of the
island of Melos (Berlin).

1 Cf. Samm. Sabouroff, Text to Taf. 25, p. 3 seq.
2 In Magnesia accordingly she was called 'Aipp. MriXeta, as we know from a coin (Imhoof-Blumer, Monnaies
Grecques, p. 292 ; cf. Herakles M^Aeiot). This Aphrodite has naturally nothing whatever to do with Melos (in
which case the adjective would be MriXta). Ravaisson, however ( Vinus de Milo, p. 109), quoted this coin
(without giving the reference), and translates ' Venus des Meliens ! '
3 The coin is published by Imhoof-Blumer, Griech. Miinz. Taf. 2, 9. The figure, which is bearded, is,
regardless of sex, designated as Tuxi- I cannot agree with Imhoof 's view that this is due to mere carelessness ;
it seems to me more likely that the image refers to a benefactor of the island who was represented as her Tyche,
holding the infant Ploutos.
4 So in Tarral and Hasse. The correct length is given in Valentin's restoration (Kunst und Kiinstler, plate
to p. 240).

TYCHE OF MELOS

3§3

the thigh. All this portion is unfortunately much
worn away, yet the lie of the folds is still sufficiently
clear to show that the hand was busied here.
Possenti, who is of the same opinion, has also
pointed out to me that the right arm, like the
left, must have been joined on, because the block of
marble of which the upper torso was made was not
large enough to include the whole of the arm. The
existing arm-stump shows evident marks of break
age, hence the contact-surface was probably lower
down, as indeed we should expect from the size of
the block, which is large enough to have included
the upper arm.
The right arm was not fastened on in the same
way as the left arm. It was not supported, but hung
down, therefore it needed a more solid attachment.
It was connected with the body of the statue by a
strong side dowel. Under the right breast is a large
square

Fig. 168. — Relief from a column in Melos.

Fig. 169.— Aphrodite with the apple. Terra-cotta from Myrina
(Eerlin).

hole two inches deep, for
the dowel which supported the arm.1
In the front view, this dowel was, as
Possenti's restoration shows, not visible,
and in the side view which our figure
gives it is covered by the arm. The
explanation of the right arm already
given by Dumont d'Urville — namely,
that it was feeling for the drapery —
turns out, therefore, to be correct, as
must strike any careful observer un
prejudiced by aesthetic theories. The
mantle is but loosely thrown round the
body, and the action of pulling it closer
by the end which falls over the left
thigh is a very natural one, requiring
no special intention to account for it.
The folds over the left thigh are some
what displaced and drawn up, and this
can only be due to the action of the
right hand. We should note, however,
that the motive of the left hand, which
was supported and held an apple, is
comparatively independent.
It is clear that the two arms thus
restored lend neither unity nor harmony
to the composition ; in short, their loss
is one less to be deplored than might
at first appear. But since we are now

I am now convinced by Signor Possenti's arguments that the hole belongs not to a later restoration (as I
formerly believed), but to the original fastening of the arm, which was very firm and solid.

384 THE VENUS OF MILO
trying to find out, not what the statue ought to have been or how it would answer
best to our preconceived notions, but what it actually was, we shall not allow this
discovery to affect our results.
We may reasonably ask at this point : If the goddess is leaning on her left arm
and calmly holding her attribute, whence comes the agitated position of the upper
part of the torso, and why is the left foot raised ? Why has she put on the mantle
in such a way that she must grasp it with the right hand ? What is the meaning
of the motive ?
An exquisite terra-cotta statuette from Myrina, of Hellenistic date (Berlin,
Fig. 169),1 gives a variant showing the harmonious grace of which the subject was
capable. In this figure we have again an Aphrodite with the upper part of the body
undraped, the left arm leaning on a pillar, the right leg supporting the weight of the
body, and the head turned to the left. But the motive is vigorous and concentrated.
The figure seems to be one of the few terra-cottas which really reflect the spirit of
statuary. It is an old type effectually made new by the upward turn of the head
and the spread of drapery arranged to form a background to the nude torso. A
specially interesting feature of the conception is the animated manner in which the
right hand holds up the attributive apple.2
How was it that the artist of the Melian statue failed to round off and harmonize
his work in some such way ? The answer to this question demands a separate
chapter. III. fnfluences that affected the Artist of the ' Venus' — Skopas. — Historical
Position of the Venus.
The artist of the Venus of Milo took two entirely independent traditional types,
and tried to combine them by means of partial modification. The result, as might
have been imagined, was not altogether happy. We have already seen that the
motive of the left arm was taken over from the Tyche of Melos. But it can also
be ascertained whence the artist borrowed the main design : it can be traced back
to an original, the best-known copy of which is the well-known Venus of Capita
(Fig. 170).3
The usual theory, that this figure is a weak Roman variant of the Venus of Milo,
is singularly perverse, and can be easily disproved by external evidence alone. The
Venus of Milo is unique, and no replicas of it exist.4 Of the Capua Venus there are
not only several exact replicas 5 but sundry Roman variants, showing that this statue,
and not the Venus of Milo, was the one renowned in antiquity.
The Capuan type several times appears grouped with Ares,6 but that this was
1 In the Berlin Antiquarium, TC. Inv. No. 8151. Height 0-29.
' Cf. another statuette from Myrina in Karlsruhe. (Sal. Reinach, Chronique d' Orient, p. 325.) The right
hand holds the apple, while the left draws up the cloak.
3 For literature see Friederichs-Wolters, Gipsab. 1452. Both arms are restored from just below the shoulder.
The basis in the original is about 35 cm. broader to the right of the helmet. This surface (left out in the cast,
cf. Fig. 170) would not be large enough for an Ares, and probably supported an Eros, a favourite addition of the
copyists. Good illustration in Brunn-Bruckmann, Denkmdler, No. 297.
4 The supposed replicas are really copies of the Venus of Capua.
5 e.g. (a) ' Venus Torlonia ' in Villa Albani, No. 733 ( Valentin, Hohe Frau von Milo, Taf. 4, 10 ; cf. Helbig,
Museums, 838). The head does not belong to the statue, (b) Torso from Smyrna in Vienna (Gdler von Ravens-
burg, Venus v. Milo, p. 173 ; Benndorf, in Arch. Epigr. Mitth. aus Oesterr. 1880, p. 72, note).
6 In the Florentine group (Clarac, 634, 1430, phot. Alinari) the Venus, both in body and drapery, is an
exact copy of the Capuan statue. The small poor group in the Villa Borghese agrees in this respect with the
Florentine ; the goddess sets her foot on a pair of greaves (group given by Nibby in Mon. Scelti d. Villa

VENUS OF CAPUA

3§5

hot the original intention is plain from the circumstance that the figure of Ares
is now of one type now of another. If these groups could all be referred to one
original, the type of Ares would always be the same. The composition of the groups
is so poor and clumsy that it is impossible to refer it to an earlier period than that in
which the extant monuments were actually executed.
The Capuan type occurs with the addition of an under-garment, not only as a
single figure,1 but also grouped with Ares2 or Asklepios.3 The chiton differs so much

Fig. 170. — Venus of Capua (Naples).

in the various examples that it cannot have formed part of the original design. It
seems clear that it was added because the type had to be used for portraits of

Borghese, Tav. 44 ; more lately by Ravaisson, loc. cit. PI. 7, I. Ravaisson is unaware of the former publication, and
has left out the head of Ares, which although broken and wrongly set is yet genuine). In the Capitoline group
(Clarac, 634, 1428 ; Helbig, Museums, 502), the Venus, being turned into a portrait-statue, has been given a chiton.
The same is the case in the Paris group (Clarac, 326, 1431), where the arrangement of the mantle is likewise
altered. Cf. the sarcophagi, and the relief from Side in Lanckoronski, Pamphyl. u. Pisid. i. p. 147 ; Reinach,
Chron. d'Or. p. 701 ; Ravaisson, loc. cit. PI. 7, 3 ; the fragment of a terra-cotta relief from Eocri published by
R. ibid. PI. 7, 4, and on which he lays great stress, has absolutely nothing whatever to do with the group. It is
the remains of a fifth-century group, similar to the one which I described in the 50th Berl. Winckelmannsprogr.
(see also Herzog, Stud. i>. Gesch. d. Gr. Kunst, p. 12).
1 (a) ' Venus Falerone' in the Louvre (No. 1737), Mon. d. Inst. iii. 2, 1 ; Golerv. Ravensburg, V. v. M. 178.
Left foot on helmet, (b) Statue in Madrid, cast in Paris (Goler, p. 1 79 ; Ravaisson, V. de M. PI. 6, 3, 4). Chiton
added after the model of the Venus of Kos of Praxiteles. Replica in the Vatican, Giardino della Pigna.
(Ravaisson, loc. cit. PI. 6, 1, 2.) (c) Statue in the Louvre (No. 1733) restored as a Muse, (d) Cast of a torso
said to have been found in the Giardino Boboli, Ecole des Beaux- Arts, Paris, (e) Coll. Jacobsen, Copenhagen.
A good copy without head or arms. (/) Statuettes in Treves and Dresden (see supra, p. 380).
2 Cf. p. 384, note 6.
3 Group in Turin, Diitschke, iv. 312 ; pub. by Ravaisson ( V. de M. PI. 7, 2).
3 D

386 THE VENUS OF MILO
distinguished Roman ladies,1 represented as Venus Victory,2 Hygieia, or a Muse, and
in all such cases a fuller costume was naturally indispensable.
In Imperial times, as I have already said, the type is adopted for a figure of
Victory writing a list of victories on a shield. This idea seems to have been first
conceived by some artist of Vespasian's reign, for the type — in various trifling variants
— suddenly appears on the bronze coins of Vespasian ; 3 we next find it on the silver
money of Trajan,4 on his triumphal column,5 and on later coins,6 but there is no
instance earlier than Vespasian.7 The famous statue of Brescia8 is known from its
provenance to have belonged to the same reign. Its artist did not care to make an
exact copy of the Aphrodite. Not only did he add the chiton, but he altered the
folds of the upper garment to suit the petty taste of his day. The nature of the
alterations, however, plainly shows that the original on which he founded his work
was — not the Venus of Milo, but — the Venus of Capua. The hypothesis that the
Brescia statue is the original type is completely excluded not only by the presence of
the chiton but by the fact that it stands alone, while the Capua type is represented by
numerous replicas. A very small alteration was enough to transform the Capuan
Venus into a Victory, for the Venus, as we shall learn presently from Corinthian coins,
was herself originally represented with a shield.
By the restitution of the shield in the Capuan type, the unusual motive, which in
the Melian statue remained obscure, becomes clear and intelligible throughout. The
goddess has placed her left foot on a support 9 in order to rest the shield on her sloping
left thigh.10 The shield was placed sideways so as not to cover the front of the statue,
consequently the body is turned towards the left. The right shoulder is lowered and
the right arm pressed close to the breast, because the right hand held the lower edge
of the shield. The left arm is raised horizontally, because the left hand held the upper
edge of the shield. The head is bent to look into the reflecting surface, and the upper
torso is nude, because it is just this part of the body that the goddess wishes to see
reflected. The mantle is high on the left hip, because the weight of the shield held it
firm on that side. On the right side it slips lower down.
A comparison with the Melian statue shows at once that its motive is derived
from the Capuan type. The main lines of the composition — the raised foot, the turn
of the nude torso to the left, the gesture of the arms— are all meaningless when the
shield is removed, and are adopted here only because they form a graceful pose.
Yet the artist was no slavish imitator, like those Roman copyists who grouped together
at random and without alteration traditional types of Ares and Aphrodite. He
was one who knew how to subject the composition to a thorough remodelling for a
definite purpose. All the movements which had the shield for centre might be made
less pronounced now that the shield was removed. Thus the inclination of the body
1 All copies with chiton where the heads are preserved are Roman portraits. To the same series belong
also the sarcophagi where the dead woman appears as Venus.
2 The late Roman figure on a cinerary urn from Lykia is half Venus half Victory (Ath. Mitth. ii.
Taf. io).
¦' Shield supported on the thigh or hanging on a palm. Upper part of the body nude.
4 Shield on a pedestal. Upper part of the body nude.
5 Shield on a pedestal. Upper part of the body draped.
6 Cf. Frohner, Mid. de I' Empire Romain, p. 163 (Caracalla).
7 The Nike writing on a shield on coins of the gens Egnatuleia and on some pre-Imperial gems is a different
type altogether.
8 Friederichs-Wolters, Gipsabg. 1453.
9 The helmet under the left foot, which appears in several replicas, was most likely arfeature of the original.
10 The shield, which was most certainly made of metal, has naturally left no traces on the drapery.

APHRODITE WITH THE SHIELD 387
and head to the left and forward being lessened, the whole figure becomes more erect,
and the eyes look straight into space. The right shoulder droops less and the right
arm falls more perpendicularly — all this evidently because the goddess is no longer
looking at her image in the bright surface. These variations alone should have
sufficed to warn those who wished to restore the Venus of Milo with a shield. But
the drapery too has been altered. As the shield is not there to keep up the left side
of the cloak, and there is accordingly no reason why one side should be higher than
the other, both sides have been allowed to slip down as far as they can without falling
off. The torso, especially from a back view, gains in sensuous charm by the change,
but the drapery would always produce an impression of insecurity even were the right
hand still intact to keep it in place.
With the rejection of the shield as mirror the meaning of the drapery and
the naturalness of the pose are lost. In the original type the goddess, who is
wont to be draped, has partly disrobed for a definite reason, in order to contem
plate her own beauty undisturbed. In the Melian type she displays her charms
apparently with no object at all, and one does not see why she does not disrobe
altogether. When we remember that our modern notions of the 'ideal nudity'
appropriate to Aphrodite were quite foreign to Greek artists of the best period,1 and
that in the fourth century, when Aphrodite began to appear undraped, there was at
first always some definite motive to account for her doing so, it must be still more
plain to us that the Capuan is the older motive, the Melian the later variant. Not
till after the period of Alexander did it become customary to represent Aphrodite
nude without any special reason, as in the well-known Medici and Capitoline types.
Praxiteles made his Venus of Knidos nude because she was just stepping into the
bath, and the Venus of Aries, which we have traced to Praxiteles, is partly undraped
because she is at her toilet and holds a mirror. The Aphrodite looking at herself in
a shield is analogous to these, and would admirably suit the period in which Praxiteles
created a pseliumene . A well-known passage in the Argonautica of Apollonios (i.742)
proves that the motive was invented before the middle of the third century B.C. The
poet is describing a piece of embroidery, part of which represents the goddess looking
at her own reflection in the shield of Ares : —
e|fe»7? h'fjo-KrjTo /3a6vTr\6/cap,os K.v0epeia
"Aj0e<>9 d%p,d%ovaa dobv crd/cos. etc Be ol &pov
irrjj/yv etro crtcaibv ^vvo^rj tce^dXacrTO ^jtwi'o?
vepOev viretc p,atplo- to B'dvTiov aTpeices avTa><;
yaXKeiri Beltcrfkov ev dairihi (paJveT IBeo-ffai.
The poet seems to have the Capuan type before his mind's eye. He too thinks
of the goddess as fully draped and loosening her chiton on one side in order to see
herself in the mirror. We have here literary evidence for our view that the disrobing
required a motive to account for it. The chiton, which takes the place of the mantle
of the statuary composition, is clearly only an effective poetic invention, and must
not be referred to another definite artistic prototype.
1 The few examples of the nude Aphrodite in archaic art (cf. Korte in Arch. Studien H. Brunn dargebr. 1893,
p. 24) are influenced by foreign types ; the nude figures supporting mirrors seem to have been borrowed direct
from Egypt. Other nude female figures (not to be confused with Aphrodite) represent attendants or priestesses :
e.g. archaic bronzes with cymbals (Korte, p. 28), flute-player (Rdm. Mitth. 1892, p. 54), and probably also the
original of the Esquiline Venus (Helbig, Museums, 566). The latter statue has been most improbably interpreted as
Atalanta, for the resemblance of the motive naturally proves nothing.

388 THE VENUS OF MILO
It would only be natural to suppose that it was the shield of Ares which
Aphrodite was choosing as her mirror, seeing that the two divinities are often closely
united in one cultus,1 were it not that the evidence of Corinthian Imperial coins seems
to point another way. The cultus-image reproduced by these coins stood in a small
temple 2 on the Akrokorinthos, and corresponded in attitude to the Capuan type, except
that the sides were reversed {supra, Plate VI. 38).3 Now in Corinth Aphrodite was wor
shipped with Helios, not with Ares. Pausanias calls the image of Aphrodite on the
Akrokorinthos ' armed ' {«ynrKio-p.evr)). Of late it has been assumed that the goddess
on the coins is the armed Aphrodite of Pausanias.4 But the goddess on the coins
holds the shield alone, which she is using as a mirror, and no other weapons of any
kind. To call such a figure ' armed ' would surely be far-fetched. It is much more
likely that the image seen by Pausanias was an ancient idol. If this be so, there
must have been two images in the sanctuary of the Akrokorinthos — one ancient and
armed, mentioned by Pausanias, and a later statue, reproduced on the coins, but
ignored by Pausanias. Similar instances arc not unfrequent. The old image of
Artemis Laphria seen by the periegete at Patrai is not reproduced on coins,6 while
the later cultus-statue appears on coins but is passed over by Pausanias. An old
idol of this kind, if we may assume its existence at Corinth, would find an analogy
in the armed figures of Aphrodite in Cyprus, Kythera, and Sparta,6 especially as
the Corinthian worship was derived from Kythera 7 and closely connected with the
Peloponnesian pre-Dorian cultus of the armed goddess in Sparta. We conclude,
then, that the shield is not derived from Ares, since the goddess is not worshipped
with Ares in Corinth, but that it is a survival of the old tradition of an armed
image. The notion of an armed Goddess of Love being unintelligible to a younger
generation, it had to be modified and humanized as in our statue.
The same change is to be noticed in other representations of divinities in the
Praxitelean period. Apollo Sauroktonos plays with the lizard which in some older
conception we may suppose to have been his attribute. The Artemis Brauronia, as
she puts on her upper robe, delicately reminds the spectator of the ritual presentation
of garments. The Apollo Smintheus of Skopas, which replaced an older image,
sports with the mouse which in older days was sacred to him. So our Aphrodite
lays aside her heavy armour, but keeps the shield to mirror her loveliness.
We shall not be far wrong if we assume that the original of the Capuan type was
made for Corinth in the fourth century, that it disappeared when the city was sacked
(while the old idol, valuable only in a religious and not in an artistic sense, may have
been spared),8 and that at the time of the new colonization its place was supplied by
a copy which, to judge by the coins, reproduced the original, only with reversed sides.
It is just possible, but less likely, that the original existed in Roman times in some
locality where Aphrodite was worshipped in conjunction with Ares, and that the
1 Cf. Preller-Robert, Gr. Myth. i. 340, and Paus. iii. 17, 5.
vaiSiov, according to Strabo, viii. p. 379. The coins differ greatly in the reproduction of the temple.
3 Imhoof-Blumer and P. Gardner, Num. Comm. PI. G, 121 sqq., p. 25 seq ; PI. D, 63 (with Poseidon). AU
the coins with one exception have the sides reversed as compared with the Capuan statue, hence they probably
reproduce the attitude of the Corinthian cultus-image. In the one exception, I mhoof- Gardner, PI. FF xiii. (coin
of Commodus in Berlin), the change is probably necessitated by the grouping of the figure with Herakles.
4 Imhoof-Blumer and Gardner, loc. cit. ; Head, Hist. Num. p. 340.
5 Studniczka, Rdm. Mitth. iii. (1888), 297 seq.
c Cf. Preller-Rober , Gr. Myth. i. 357. Aphrodite is called dpela and ii>6nXws in Sparta, (yxeios in Cyprus,.
a-nd irrpareia in Caria.
7 Cf. Alkiphron apud Preller-Robert, i. 350, A, 2.
8 We learn from Pausanias that several archaic idols had been preserved in Corinth. Cf. Imhoof-Gardner,
loc. cit. p. 10.

APHRODITE WITH THE SHIELD

389

Corinthians of the time had a copy made from it without reference to their own local
cultus. The older armed image, to which, as I believe, Pausanias refers, is decided
evidence in favour of the first view.
However that may be, it is certain that the original of the Capuan statue belonged
to the fourth century. This is plain from the reasons already adduced, and from the
style, notably of the head (Fig. 171).
The Capuan statue, which was found in the amphitheatre renewed under Hadrian,
had been correctly assigned to this Emperor's period,1 and bears every mark of the
time in its style. The cold
elegance of the workmanship,
the dry mannerism with which
the hair is treated, the plastic
indication of the pupils, the
dead smoothness of the flesh,
even the profile of the plinth
are characteristic, and for
some time prevented due ap
preciation of the value of the
figure. Fortunately however
we possess in a marble bust in
the Palazzo Caetani at Rome
an earlier and more lifelike
copy at any rate of the head
(Fig. 172).2 Careful compari
son will show that this head
must be referred to the same
original as the Capuan statue.
The only variations are a fillet
instead of a diadem, and the
absence of the small ends of
hair on the back of the neck.
The diadem is a usual Roman
adornment of Venus, and is
probably added by the copyist. The simple fillet, on the other hand, belongs
to Aphrodite in Praxitelean art, and is therefore likely to be an original feature.
The Victory of Brescia affords another proof that the fillet is correct ; she
wears a similar one (except that a laurel-wreath is engraved on it), and it
can be derived only from the original, which is identical with the original of
the Capuan statue. That the Caetani head is not a copy of the Victory of Brescia
(as I stated in Roscher's Lexikon, i. 414), and, in fact, does not belong to a Nike
statue at all, is plain not only from an accurate comparison of the two, but also
from the measurements. The Caetani head corresponds in proportions to the Capua
type, the Brescia statue stands alone. The latter is an independent Roman modifi
cation of the Aphrodite, as we have seen from the drapery. The artist kept fairly
near his original in the workmanship of the head, but did not attempt to make an
1 Bernouilli, Aphrod. p. 161.
Matz-Duhn (Zerstr. Bildw. in Rdm. 797). Von Duhn was the first to call special attention to the head ;
Friederichs-Wolters, Gipsabg. 1454. Said to be intact ; I have not seen the original. From the shape, probably
a bust, and not intended for insertion into a statue (in which case the piece to be let in would be longer in
front, and narrower on the shoulders).

Fig. 171. — Head of Venus of Capua.

Fig. 172— Head in Palazzo Caetani (Rome).

SKOPASIAN AND PRAXITELEAN APHRODITES 391
exact copy. The face is shorter and broader, and by making the lips thinner he has
omitted a distinctly Aphrodisian trait. He probably added the little curls in front
of the ear. The empty soulless expression of the face shows how totally unable this
Roman artist was to impart to his work any quality corresponding to the spirituality
of the Greek original. In studying the Capuan type, then, we had better leave the
Victory out of the argument, and keep to authentic copies.1 Of these the Caetani
bust seems to be the most faithful ; although a coarser version of the original, it
yet is uniform in style, and nothing extraneous has been added. This style, as need
scarcely be remarked, is that of the circle of Praxiteles and of Skopas. This is especially
marked in the simply parted hair, which forms an angle over the forehead, is con
fined by a smooth band, and worked without deep shadows.2 In the facial forms,
moreover, it is impossible not to recognize the character of that period.
In judging of details it is necessary to inquire whether we have to imagine an
original of marble or of bronze. The question of material is more important in con
nexion with the fourth than with the fifth century B.C. In the fifth century marble
and bronze were similarly worked (cf. p. 7) ; in the fourth century, with Skopas and
Praxiteles, was developed a technique peculiar to marble. The wonderful freedom in
the treatment of masses of hair and the delicate indication of roots, to cite only one
point, which we see in extant originals such as the Hermes, the Eubouleus, the
Petworth Aphrodite, the head from the south slope of the Akropolis, and the Demeter
of Knidos, and which we can divine from copies such as the Aphrodite of Knidos, would
be impossible in bronze. Copies of the Sauroktonos show how Praxiteles treated
bronze hair. Our Aphrodite is evidently derived from bronze, not from marble, and is
analogous to the Sauroktonos, not to the Aphrodite of Knidos and other works whose
originals are to be imagined as marble. The hair is not loosely massed, but defined
from root to point in separate locks that show great variety of form. On the knot of
hair at the back and on other parts of the head are marks of incision, evidently an
imitation of bronze technique. The motive of the statue, an extended arm holding a
shield, is much more appropriate to bronze than to marble.
It is instructive to compare with the Caetani head similar conceptions as treated by
Praxiteles. The Aphrodite of Knidos (p. 322) is a good instance for our purpose, and the
Venus of Aries (p. 320) a still better one, because the pose of the head is similar to that
of the Caetani Aphrodite, and the motive, a mirror, is probably the same. The Caetani
Aphrodite differs materially from both these types. At the first glance we notice the
absence of the long regular oval face so characteristic of the Praxitelean school, we
miss the marvellous harmony and calm conveyed by the gentle and even balance of
the face-curves from centre to sides and from sides to centre. Here the lines are
restless and broken, and seem to crowd to the front. The hair forms a less regular
angle on the forehead, which is wider, especially at its base, the eyebrows are thicker,
the cheek-bones broader, the setting of the eyes more marked. The inner corners of
the eyes are deeply sunk — almost drawn into the head — and at a higher level than the
outer corners. The nose projects more, making a less perpendicular profile. The
nostrils swell as if breathing. But the mouth shows the most decided difference of all.
The lips are fuller and more animated, the raised upper lip is strongly curved, the
lower one has a strong dent in the middle. The ear, too, has its own peculiar form,
1 The marble head worked into a bust in the possession of Count Stroganoff in St. Petersburg is a replica
of the Capuan statue. Another is in the Coll. Barracco, Rome ; Coll. Barracco, p. 60, 60 a. Helbig (text, p. 46)
does not recognize it. A much-restored head in the Louvre known to me only from Ravaisson, V. de M. PI. 5,
3, is possibly a copy. 2 Cf. Samm, Sabouroff, i. Skulpt. Introd. p. 1 1 sqq.

392 THE VENUS OF MILO
unlike that observed in those two Praxitelean heads. Instead of standing upright it
lies over towards the back. The hollow is small, the cartilaginous part broader with
decided protuberances, the lobe is small and delicate, and completely joined to the
head in front. What we lose in grace, harmony, and repose by comparison with
Praxiteles we gain in a more individual expression, in a more vigorous life, and in the
fiery energy of mouth and eye.
Now the contrasts noted above are practically the very same that mark off the
works of Skopas from those of Praxiteles.1 The Caetani head possesses, in common
with Skopasian heads, not only expression and structure, i.e. the short broad
face, but even details such as the shape of the ears, the breathing nostrils,
and the raised upper lip. The few points of difference are not noticeable enough
to warrant us in ascribing the work to any other artistic individuality. We must
remember, too, that this is only a copy, and that the subject is a new one in the
Skopasian school. The head is looking down instead of up ; hence the upper eyelids
are plainly seen, instead of nearly disappearing as they do in those other heads to
which an upward gaze has been given.
The Skopasian type represented by the Caetani head is not isolated. Several
other monuments are closely analogous to it. The head of a Leda2 in Florence
shows great similarity in hair, fillet, forehead, and eyes, and although of poor
workmanship it must be referred to an original by the artist of the Caetani head
or by one of his imitators. More significant is a head of rich, strong beauty, adjusted
to a statue called ' Giunone,' in the Capitoline Museum ; it bears unmistakable
analogies to fourth-century types (Fig. 173) ;3 in the forehead, eyes, and breathing
nostrils it recalls the Caetani head ; at any rate, it comes nearer to the Skopasian than
to the Praxitelean school. We must of course beware of imagining that all Skopasian
heads must have the intense upward look to be seen in many works of the school.4
Further, in the Jacobsen collection, in Copenhagen, there is a head of Aphrodite
which, though a poor and much-defaced copy, shows the distinguishing marks
of the style of Skopas.5 A colossal head in the Hermitage may, on the other
hand, be a Hellenistic adaptation 6 of the Caetani type.
We have seen that the head of the shield-bearing Aphrodite should be ascribed
to the school of Skopas. Turning now to the body, and comparing it with that of the
Knidian Aphrodite for instance, we shall be struck by the elasticity and vigour of our
statue, due to the greater spareness and slenderness of hips and breasts, while the
heavier build of the Knidian goddess seems to betoken easy negligent repose. This
distinction is in keeping with what we know of the different tendencies of Praxiteles
and of Skopas. Specially peculiar to the Capuan type 7 are the sloping shoulders and
1 Botho Graf, Rdm. Mitth. 4, 1889, p. 189 sqq. Cf. Von Sybel in Liitzow's Zeitschrift fiir bild. Kunst,
N. F. ii. 249 sqq.
2 Diitschke, Uffizien, 192 ; Overbeck, Zeus, p. 514; Miiller- Wieseler, Denkm. 2, 44. Cast in Dresden.
A replica in the courtyard of the Naples Museum.
¦3 For literature cf. Helbig, Museums, i. 532. The head is inserted into the statue, not broken. It
is of different marble, still it might belong to the statue, were not the style so different. It is of fourth-
century style, while the body is Hellenistic and Pergamene. Cast of the head in Dresden, from the Mengs
collection. 4 Graf (loc. cit. p. 218) admits that the female head of the Capitol (cf. also Helbig, Museums, i. 445), which
he classifies as Skopasian, has been worked over and intensified. I should go further, and say that it is a decidedly
Hellenistic development, a long way removed from Skopas.
5 Ny Carlsberg Glypt. 1073. Mouth, nose, and other portions are restored. The hair is simply arranged
in a knot, without fillet. Already Helbig had with fine insight recognized an Aphrodite in this work.
G D'Escamps, Marbres Campana, PI. 38 ; Guedeonow, Ermit. 175.
7 In this respect the replica grouped with Ares in the Florentine copy agrees with the Capuan statue.

3

Fig. 173. — Head in the Capitol. (From the cast.)

394 THE VENUS OF MILO
the rounded breasts, which are more globe-shaped than in the Praxitelean
Aphrodites. In the drapery, again, fourth-century taste is conspicuous.1 The rolled drapery
about the hips is peculiarly characteristic, and is found on countless Attic grave-reliefs
of the period. The Ludovisi Ares — an invention, as we have seen, of Skopas — wears
drapery similarly treated, and so do the Niobids. In the latter connexion, the
Florentine Niobids should be specially studied,2 for, though poor copies, they faithfully
reproduce fourth-century style, while the famous ' Chiaramonti ' Niobid 3 is really not
a copy at all, but a free translation of one of the original figures into Hellenistic
forms. But to resume— the drapery of our Aphrodite well illustrates the tendency
of Attic art in the Praxitelean period to despise decorative effect and aim straight
at truth of nature.4
In the agitated pose and slim proportions of our type scholars thought to
detect the Lysippian manner. But we have everywhere recognized in Skopas the
precursor of Lysippos, the artist whom Lysippos most closely followed. Of those
works which have been attributed to Lysippos the one most similar to the Aphrodite
is the 'Eros stringing a bow';5 the movement— except that it is directed to the
other side — is almost identical, a correspondence which makes it probable that the
two compositions are related. The position of the Aphrodite, however, although full
of energy, is yet tranquil and firm, while the Eros has that sway of the torso which
is so distinctive a mark of the Apoxyomenos. Now, as the Palatine Hermes {supra,
Fig. 129) which we referred to Skopas had precisely the same motive, we are bound to
consider whether the Eros also may not be traceable to Skopas. This, however, seems
to me improbable, owing to the insignificance of the type of head.
We have assumed that, like that of the ' Eros stringing a bow,' 6 the original
of the Capuan Aphrodite was of bronze. This does not invalidate the Skopasian
attribution, for not only does Pliny mention Skopas among the bronze-workers, but
we know for certain that his Aphrodite Pandemos at Elis riding the goat was of bronze,
and there may have been many other bronzes among the works by him whose names
have been handed down to us without mention of the material in which they were
executed. We have no reason to suppose that Skopas, any more than Praxiteles,
worked solely in marble. Among the works which we have already brought into
connexion with Skopas, some may have been of bronze — for instance, the youthful
Athena {supra, Fig. 130).
There are still a few more statues which we venture to refer to Skopas because
of their resemblance to the Aphrodite ; chief among them is the Leda in Florence
(for discussion of the head-type see p. 392). The torso and drapery are allied to the
Aphrodite, the shoulders and the rounded but less developed breasts are similar.7
The motive, which has been often misunderstood, is free from any voluptuous sugges
tion.8 The frightened bird has flown to the girl for protection ; she is bending over
it and wrapping it in the folds of her garment. The swan is small in scale, as in
1 Much restored in the Capuan statue. Better preserved in the Florence group.
2 Cf. the hanging drapery of the ' Son of Niobe,' Overbeck, C.
3 Helbig, Museums, 73.
4 Cf. Samm. Sabouroff, i. Skulbt. Introd. p. 14.
5 Cf. Roscher's Lex. i. 1362.
6 Cf. Helbig, Museums, i. 429. The Thespian Eros of Lysippos was of bronze.
7 Note also the bracelet on the right upper arm corresponding to the Florentine replica of the Capuan type.
The Praxitelean Aphrodite wears it on the left arm.
8 Such as Overbeck, Zeus, p. 514. O. Jahn (Arch. Beitr. p. 4) appreciated the group much more
correctly.

PSYCHE OF CAPUA. 395
the earlier ' Leda ' monuments, and the whole conception is that of the Greek terra
cottas of the fifth and fourth centuries B.C., and of a statue which I have discussed
elsewhere.1 In this case too the figure should be interpreted rather as Nemesis than
as Leda. It is interesting to compare this Skopasian 'Leda' with that older oft-
copied Leda, which, like the Barberini ' Suppliant,' to which it is allied, must be
referred to the fifth century, and to an artist nearly connected with Pheidias.2 It
is more dramatic and effective than our Skopasian statue, which is simpler, more
natural, and deeper in feeling.
A second not less interesting work of the same series is the Psyche of
Capua (Naples),3 which has so often been called Praxitelean,4 but is really a
very good copy after a purely Skopasian work, with the characteristic sloping
shoulders and globe-shaped breasts. The latter have, it is true, been somewhat
worked over, but the essentials of the form remain unchanged : it is the same
flat, youthful breast that we noticed in the Nemesis-Leda.5 There is a bracelet
on the right arm.6 Marks of the Skopasian Aphrodite type in the face are seen
in the peculiar mouth, the dilated nostrils, the ear set slanting back with small
opening, the strongly marked ear cartilage uneven in outline, and in the lobe
growing close to the head. Such cumulative evidence leaves no room for doubt
that we have here a work by the artist of the Caetani head. But the interpretation
of the figure is not so easy. It is certain, however, that it does not represent Psyche,
as was formerly supposed. Aphrodite would be nearer the truth. The figure obviously
leaned its weight on the right leg, and the left hand drew over the left shoulder the
drapery which shrouded the lower part of the body.7 The head was turned to the right
and bent down. Perhaps the goddess was grouped with an Eros holding a mirror.
The 'melancholy character' of the work cannot be urged against this interpretation,
for it is easy to imagine melancholy in a bowed head when we know nothing of the
motive. The head is posed as in the Capuan Venus, only more inclined. This inclina
tion doubtless formed the principal charm of the figure in the eyes of the artist, who
in this instance has with characteristic boldness gone beyond his prototype. One step
more, and we should have the ' Kallipygos ' motive of the succeeding generation.
A third beautiful plastic conception, which we may refer to a bronze original
by Skopas, is the Hypnos, represented by a marble copy in Madrid and a bronze
head in the British Museum.8 It has long been acknowledged that the original must
have been in bronze. This is proved not only by the extant remains of a bronze
copy, but by the vigour of the action, which, when represented in marble, necessitates
the presence of a meaningless and disfiguring support. The style of this work,
1 Samm. Sabouroff, i. Vases, Introd. p. 8 sqq. 2 Ibid. p. 9 seq. 12.
3 Friederichs-Wolters, 1471. The piecings are original, and not, as was formerly believed, due to restora
tions. Date first century B.C. The figure is analogous to the Ares of the Conservatori Palace (Helbig, Museums,
568) (see supra, p. 271), which is also made up of several pieces. Neither figure had wings. The working over
is by no means as bad as had been supposed : it can be detected on the breast in the region of the collar-bone,
but the head has been left absolutely untouched.
4 So lastly by Benndorf, Bull, della Comm. Arch, di Roma, 1886, p. 73.
6 Cf. for the breasts of the Psyche, E. Briicke, Schdnheit und Fehler der Menschlichen Gestalt, p. 67.
6 Cf. p. 394, note 7.
7 E. Wolff recognized this (Bull. d. Inst. 1833, 134) ; also Stark (Sachs. Ber. i860, 90). An examination 01
the original will show Kekule's supposition (Annali, 1864, 145), that the left upper arm was wrongly set on,
and had originally been drawn back, to be quite false. His further hypothesis, that Psyche was tied to the tree,
naturally becomes untenable.
8 Friederichs-Wolters, Gipsabg. 1287, 1288. For list of small bronze replicas see Winnefeld, Hypnos, p. 8.
Of these, the Turin figure is published in Atti della Soc. di Archeol. e Belle Arti di Torino, iv. 1883, Tav. 15,
p. 113, and the two Lyons copies in Gaz. Archiol. 1888, PI. 6. See also Sauer in Roscher's Lex. i. 2849, and
Brunn, Gdtterideale, p. 26, Taf. 3.

396 THE VENUS OF MILO
rightly attributed to the Praxitelean epoch, has been likened to that of the Sauroktonos,1
but the more closely we look at the head the more clearly we shall see that its
resemblance to the Sauroktonos is merely superficial, while it is closely analogous to
the Skopasian Aphrodite — in a word, it offers the same contrast to the Sauroktonos
that the Aphrodite of Skopas does to the Aphrodite of Praxiteles. We find in the
Hypnos a broad type of face and a pronounced bony structure, quite different from
the graceful elongated oval of the Sauroktonos. The forehead of the latter, and
the wavy hair that frames it, correspond closely to the Venus of Aries and the
Knidos Aphrodite, while in the Hypnos the forehead and the hair growing back from
it so strongly resemble the Caetani Aphrodite that we are justified in referring the
two conceptions to the same artist. The dilated nostrils of the Hypnos are character
istically Skopasian, but the protuberance in the middle of the forehead, which, as
Brunn showed, is definitely necessitated by the wings, is individual to the divinity
represented. Finally, the swinging stride, recalling the Palatine Apollo,2 and the
chiastic disposition of the limbs are much more appropriate to Skopas than to
Praxiteles, while, to pass to the treatment of the nude, the modelling of the abdomen
and the navel with its little furrow are details more nearly related to the Skopasian
Meleager than to the Sauroktonos or the Hermes.
This magnificent invention, then — the work of an artist who went straight to the
heart of his subject — must be assigned to the later period of Skopas's activity. Unlike
the Kairos of Lysippos, a composition carefully built up by an effort of activity, this
conception flowed warm and living from the soul of the artist. With gliding gait
and quiet pulse of wings the god of sleep sinks down upon a weary world. It was
a happy inspiration of Skopas to place the wings on the head instead of on the
shoulders, as in the earlier representations of Hypnos.
The original of the statue must surely have been intended for one of the few
shrines where Hypnos was worshipped. We only know of . three such in Greece.
There was an ancient altar in Troezene which Hypnos shared with the Muses (Paus. ii.
31, 3); statues of Hypnos and Thanatos (to which a cult was probably attached)
stood near the Chalkioikos in Sparta (Paus. iii. 18, 1) ; in the Hieron of Asklepios
at Epidauros Hypnos was worshipped, at least in later times ;3 and finally the front
chamber of an oi/ajpa in the peribolos of the sanctuary of Asklepios at Sikyon
(Paus. ii. 10, 2) was consecrated to him.4 As Asklepios was wont to manifest himself
to his suppliants in sleep, and Sleep might easily be personified as a daemon favourable
to recovery, the connexion of the two divinities seems natural enough. Pausanias
states that only the head of the figure was in existence in the cella.6 Possibly the
original statue had been stolen, and if so the head Pausanias saw must have been
a copy, for whoever took the statue would not leave the head behind. We know
that Skopas worked in Sikyon, for Pausanias refers to a Herakles there made by
him (probably reproduced on Imperial coins).6 Hence it is not impossible that the
statue in the cella was also a work by Skopas.
Other coins of the Imperial times seem to show that a third famous work of the
same artist, not assigned by tradition to any special place, may have belonged to
1 Cf. Murray, Hist, of Gr. Sculpture, ii. 259, where Benndorf is quoted.
2 Cf. p. 305, note 6.
J Cf. Blinkenberg in Nord Tid-Skriftf. Filol. ny Rdnke, *. 270, 8 ; 273, 20.
4 The back cella was consecrated to Apollo Karneios.
6 In the, Stoa which was naturally attached to the Asklepieion, Pausanias saw another statue of Hypnos—
KUraKoiv-lCiov Xiovra ; Pausanias may be here inaccurately referring to the type (known by later copies) of Hypnos
sleeping on a lion. (Bull. d. Inst. 1877, 122 seq.) e cfi p> ^ n0(e 2

SKOPASIAN MAENAD. 397
Sikyon. I refer to the Maenad, known to us through two epigrams and a passage of
Kallistratbs.1 On the coins of Julia Domna in Sikyon there occurs the type of
a frenzied Bacchante which agrees with the description in the epigrams,2 and which
Imhoof-Blumer and Gardner3 have pointed to as reproducing one of the marble
Pdicxai seen by Pausanias (ii. 7, 5) in the temple of Dionysos at Sikyon beside
the gold and ivory image of the god. Pausanias, as often, has omitted to name
the artist, but it seems to me highly probable that this Bacchante, celebrated enough

Fie. 174.— Head from Tralles, in Smyrna. (From the cast at Bonn.)
to be reproduced on coins, was actually the Bacchante of Skopas. The flying drapery,
floating gait, and rapturous attitude show that the original statue must have borne
a strong resemblance to the Apollo Kithairoidos.
But we must return from this digression to our special subject, the Venus of Milo.
Enough has been said to prove that it was based on a work presumably by Skopas ;
1 Kallistratos himself is probably dependent on the epigrams (cf. Jahrb. d. Ver. d. Alterthums-Freunde im
Rheinland, Heft 90, 1891, p. 66).
2 As xw'po^vos she must have held a sword ; no such weapon can however be distinguished on the coins.
3 Num. Comm. p. 29, PI. H. vi. v.

398

THE VENUS OF MILO

thus it was a right instinct which induced scholars who could not bring proofs to
confirm their impression to bring it into connexion with this artist.1 In spite of the
numerous modifications, the Skopasian groundwork is manifest not only in external
details, such as the arrangement of the fillet and of the hair (which is however
effectively modernized by the addition of the loosened strands falling on the neck),
but in the attempt to impart fire and vitality to
the expression. The artist sought to heighten
this Skopasian quality. In his desire to bring into
the conception something of dignity and eleva
tion, he lengthened the lower part of the face,
thus returning to the older manner. The suc
cessors of Praxiteles, in the same way, tried to
outdo him in grace. The result in each case is a
slight exaggeration of the master's peculiarities.
The Aphrodite of Skopas becomes the Venus
of Milo, the Aphrodite of Praxiteles becomes
the Venus of Medici.
It is characteristic of a later development
that the firm, well-knit forms of the original
Skopasian head have undergone in the Melian
statue a certain relaxation, threatening to become
almost fluid in their lack of compactness and
definition. The same tendency may be observed
in other heads of Hellenistic date, notably in
the much-admired head from Pergamon,2 so
often compared with the Melian Aphrodite.
This head, though doubtless considerably older
than our Venus, shows the laxity of the
forms carried to such a point that an ex
pression of weakness, almost amounting to
vacancy, is the result. On the other hand, the
beautiful head from Tralles, now in Smyrna
(Fig. 174),3 belongs to the beginning of the
series, and is not far removed from Skopas in
general character, betraying Hellenistic taste
only in the treatment of the hair.
Mention must also be made here of a smaller and more widely known head from
Tralles which has been brought into close connexion with the Melian statue, and even
referred to the same original,4 although it has in common with it only quite general
traits. Its artist evidently imitated Praxiteles, from whom he may have been
separated only by a generation or two. It cannot be said that he made a definite
copy, but he followed that master's characteristic rendering of form, such as the delicate

Fig. 175. — Statue in Pal. Valentini (Rome)

1 e.g. Waagen, Welcker (Ant. Denkm. i. 445), Stark (Sachs. Berichte, i860, p. 51), Urlichs (Skopas,
p. 122), and B. Graf (Rom. Mitth. iv. 217) have all in a more or less degree pointed to the Skopasian
characteristics. 2 Liitzow's Zeitschrift f. Bild. Kunst, 1880, xv. 161.
3 In the Mus. of the Evang. School. Badly published in Bull. Corr. Hell. 1882, PI. III. Cast at
Bonn (Fig. 174).
4 Benndorf in Oesterr. Mitth. 1880, Taf. 1, 2 ; p. 66 sqq. ; Overbeck, Gesch. d. Plastik, 3rd ed. ii. 342;
Friederichs-Wolters, Gipsabg. 1451 ; R. v. Schneider, Uebersicht der Wiener Kunsthistor. Samm. 1891, p. 78,
interprets the head as an Artemis.

ATTIC PROTOTYPES.

399

oval of the face, the dimple on the chin, the sweet expression ; but the whole conception
has become small, not to say trifling.
We have seen (p. 377) that the epigraphy of the inscription allows us to date
the Venus of Milo as early as the end of the third century B.C., though the period
between 150 and 50 B.C. was more probably that in which the Venus was made.
The stylistic peculiarities now confirm this later
date. The hair is arranged, as we have already
said, in the fourth-century manner, but the treatment
is in marked contrast to fourth-century work. What
hard rendering of the roots of the hair on the fore
head, what lifeless parallel grooves in the strands that
are combed back ! The above-mentioned head from
Pergamon, which belongs probably to the third cen
tury, illustrates the exact reverse of this procedure.
The artist has arranged the hair according to the
fashion of his time, but in the delicate indication of
the growth of the hair and in the light treatment of
the region round the eye he preserves the best tradi
tions of fourth-century work — traditions which seem
to have been unknown to the artist of the Melian
statue. In determining the style of our Aphrodite, the
drapery is also of very great importance. Here the
artist has definitely altered his Skopasian model.
He has rejected the treatment which aims at simple
truth of nature, and has chosen instead — not the
usual Hellenistic manner, as shown in the marbles
of Pergamon — but, strangely enough, the manner
of the Parthenon pediment figures.1 The stuff lies
in large unbroken masses, with sharp edges and
no cross folds, and it clings to the nude parts as if
moistened. Only on the right thigh does it show
a touch of Hellenistic taste. To this large massive
treatment of drapery we owe the theory, not with
out adherents even at the present day, that the
statue is by a successor of Pheidias. In my view
this imitation is precisely what shows the statue to belong to that Renascence which,
about the middle of the second century B.C., embodied a reaction against the extravagance
of the Hellenistic school. The Belvedere Herakles torso is a work of kindred ten
dency. Here the nude is treated according to Pheidian tradition,2 and without any of
the realism in which the art of Pergamon excelled. A comparison of the Venus of
Milo with the Venus of Medici will convince us that the Melian figure shows a return
to Pheidias not only in the draped but also in the nude parts.3 The impression of
grandeur of style which the statue has always made was intentionally aimed at by the
artist. It is probable also that the statues of the Pheidian school which gradually
lead up to this motive were not unknown to him. Perhaps the statue of Aphrodite

Fig. 176. — Statue in Pal. Valentini.
(Restorations omitted.)

1 Cf. Overbeck, Gesch. d. Plastik, 3rd ed. ii. 341.
' Roscher's Lex. i. 2182, 20.
3 Cf. Waagen, Kunslwerke und Kunstler in Paris, p. 108,

400 THE VENUS OF MILO
Ourania, in Elis, by Pheidias, with the left foot on a tortoise, was the origin of the
whole series.1
An excellent work of art belonging to the generation after Pheidias, probably an
Aphrodite, is preserved to us in two copies in the Palazzo Valentini 2 and the Palazzo
Odescalchi at Rome (Figs. 175, 176). This figure shows so many points of analogy to
the Venus of Milo, more especially in those particulars where the latter differs from
its Capuan original, that we may very well imagine that the artist in his adaptation
of the Aphrodite of Skopas consciously emulated the style of some such Pheidian
prototype. Aphrodite here, as always in the Pheidian period, wears the chiton under the
mantle. It clings closely to the body, thus forming an almost transparent covering,
in the manner of the sculptures on the Nike balustrade and of the Aphrodite of
Alkamenes. The left foot is advanced but not raised, and the form of the left leg is
distinctly seen through the mantle, just as in the case of the Aphrodite of Melos. The
sharply defined curves of the mantle wrapped round the body and enveloping the
right leg, on which the weight of the body is supported, show the breadth of concep
tion, the combined wealth and animation of forms which characterize the school of Phei
dias from the epoch of the Parthenon pediments onwards. The Hera of Ephesos (p. 84,
note 4) and the so-called Barberini Hera (p. 82) are similar works, but the Aphrodite
Valentini far surpasses them in beauty and richness of drapery. One of the fragments
of the reliefs by Agorakritos 3 is very closely related to the Valentini figure. Again,
the head of the figure, with its large full forms, its forehead high in the middle, and
its delicate roots of hair, finds its nearest analogies in two statues which we have
already ascribed to Agorakritos (p. 88), i.e. the Barberini Apollo and the Athena of
the Capitol, and this circumstance gives some weight to the belief that the Valentini
Aphrodite is to be referred to Agorakritos. There is the same dignity and majesty
of pose, the same expression of frank victorious pride, that we saw in other works by
the same hand.
The artist of the Melian Aphrodite has tried to catch at least a ray of inspiration
from Agorakritos, but his work, as we have shown, must be dated not earlier than the
Renascence of the second century B.C. This chronology is confirmed by a purely
external circumstance — the appearance of the inscription on an integral part of the
statue. This is without any analogy before the end of the second century B.C.4
Previous to that time, with the exception of course of the archaic period, it was the
invariable custom of artists to sign their statues on the pedestal ; after this date, prob
ably in order to insure permanence in case the statue should be removed, signatures
were placed on some block — such, for instance, as the supports — which was intimately
connected with the statue.
The Venus of Milo, then, belongs to a series of works executed in the latter half of
1 Cf. p. 72 seq. In a recently acquired statue in Berlin, see p. 73, note (Arch. Anz. 1893, p. 74, and
Kekule, Weibliche Gewandfigur, etc.), the left foot is placed higher. In Stackelberg's terra-cotta (Grdber,
Taf. 69) the foot is only slightly raised, and in the terra-cotta (Rev. ArCh. 1891, i. PI. 6, p. 289 ; Th. Reinach)
from the Troad, probably a copy of a statue, it is raised extraordinarily high.
2 Pal. Valentini: Guattani, Mon. Ant. ined., 1788, Tav. 2; Clarac, 698, 1655; Matz-Duhn, Zerstr. Bild-
¦werke, No. 606 ; Roscher's Lex. i. 414. The left arm and lower right arm and the cymbals are restored. The
copyist has neglected to give the Ionic sleeve on the left arm. The head is broken, but belongs to the figure.
The left hand evidently drew the mantle up over the shoulder. Thick sandals on the feet. " Large eyes.
Good, intelligent workmanship.— Poorer replica in Pal. Odescalchi. Head, lower right arm, and left arm are
restored (Matz-Duhn, 605).
3 'E^ij/i. apX. 1 89 1, Taf. 9, 1. Cf. p. 85.
4 The earliest examples are the Borghese Gladiator by-Agasias (Ldwy, /. G. B. 292 ; ibid. 293, 343) and the
Belvedere torso (cf. Ldwy, p. 350).

DATE AND AESTHETIC POSITION 401
the second and first century B.C., and many of which are of considerable excellence.
The artist took his motive from a creation of Skopas, which he modified con
siderably, and contaminated, not altogether happily, with the type of the Tyche of
Melos. At the same time he strove to impart to it something of Pheidian grandeur.
So far then he showed independence in his modification of the style, inasmuch as
he drew his inspiration from older Attic art, and sought to emulate it. Call him
' eclectic ' if you will, he was at least a man who could make a traditional type his
own, and reproduce it with all the freshness of a new conception. The pleasure
arising from this quality in his work should not be lessened by the scientific analysis
to which we have subjected the statue.

THE APOLLO OF THE BELVEDERE

THE APOLLO OF THE BELVEDERE

S'

ELDOM has any archaeological hypothesis met
with such enthusiastic and universal approval
as did the famous contention published by
Stephani in i860 concerning the Apollo of the Belve
dere. Ever since that time it has been considered
practically certain that the god bore an aegis in his
left hand. This view soon made its way into every
popular handbook as an ascertained result of modern
archaeology,1 although the only evidence for it was a
doubtful fragment in the hand of a bronze replica of
the Apollo belonging to Count Stroganoff in St. Peters
burg. Not only was Stephani's identification of this
fragment accepted almost without dissent, but con
fidence was carried to such a pitch that his further
vague and entirely unfounded supposition, that the
statuette was part of the great bronze find made at
Paramythia in 1792 — a statement evidently intended to increase the value of the
statuette — was accepted as a proven fact even in scientific circles.2 In reality, the
statuette had been traced back to Italy, and no further.
When for the first time I saw the Stroganoff bronze in 1882, I was to my
astonishment at once convinced that, whatever might be the meaning of the frag
ment in the left hand, it had never formed part of an aegis.3 In 1883, however,
Kieseritzky defended Stephani's view,4 which henceforth passed almost unchallenged.
I made no answer at the time, as I wished first to examine the bronze again ;
accordingly, in the article 'Apollo' written in 1883 for Roscher's Lexikon, I stated
circumspectly that the fragment of the attribute had not yet been fully explained.5
This reserve was, to my regret, misunderstood in some quarters : I was supposed
to be incapable of acknowledging an error,6 and faith in Stephani remained unshaken.
1 See Helbig, Fiihrer, No. 158 ; Friederichs-Wolters, Gipsabg. 1523 ; Preller-Robert, Griech. Myth.
i. 295. " Overbeck, Apollo, p. 248 ; Helbig, Fiihrer, i. 106. The figure resembling the Apollo Belvedere— of
which we know neither the material nor the provenance — given by Veli Pasha to Dr. Frank cannot have been
one of the exquisite bronzes found at Paramythia in 1792, and may easily have been a forgery. This is again a
perfectly arbitrary and improbable conjecture of Stephani. The Pasha was not such an ignorant fool as to
present an object of such value, even supposing him to have had it to give, to a person like Dr. Frank.
3 Arch. Ztg. 1882, 247. •> Arch. Ztg. 1883, 27. 5 Roscher's Lex. i. 465.
6 Studniczka, Wiener Zeitschrift fiir Gymnasialwesen, 1886, p. 682.

406 THE APOLLO OF THE BELVEDERE
Opposition was silent till 1887, when almost simultaneously two scholars, A. Gercke
and 0. A. Hoffmann, showed by a number of incontrovertible proofs derived from
ancient literature that the notion of Apollo bearing the aegis and victorious over
the Gauls was simply an erroneous modern fancy.1 In 1889 O. A. Hoffmann
proceeded to an investigation of the Stroganoff bronze, and maintained that
Kieseritzky's defence of the aegis theory was self-contradictory and out of harmony
with facts.2 Gherardo Ghirardini followed him on the same lines, and in an elaborate
treatise emphatically denied the existence of the aegis.3 He deplored that the argu
ments adduced by me in 1882 had hitherto met with so little notice, and hoped for a
speedy change of opinion. Nevertheless Helbig, in his Fiihrer durch die Sammlungen
Roms,4- 1 891, set aside as unimportant every objection to the prevailing theory. On
the other hand, F. Winter, in his admirable treatise on the Belvedere statue, published
in 1892,6 decided to leave the bronze statuette and its would-be aegis entirely out of
the discussion. In the spring of 1891 I carefully examined for the second time the Stroganoff
bronze, which was then accessible to the public on certain days. All at once the
scales fell from my eyes : the solution of the problem was infinitely simpler than had
ever been imagined — the bronze is nothing but a poor and absolutely valueless modern
forgery.® Since the year 1882, when I first saw the statuette, I had had every opportunity
of training my eye to appreciate the distinctions between real and false bronzes, in the
course of my work at the Berlin Museum, where a great number of forged bronzes
from different collections have been specially brought together for purposes of study.
In 1882 I failed to detect the forgery, partly from lack of practice, and partly because
the trammels of authority were too strong to allow of doubt where a figure so famous
was in question.
The statuette — which however is not small, but one third life-size — shows not a
trace of the genuine patina which appears on all antique bronzes that have really been
excavated.7 This is alone sufficient to prove the modern origin. There is another point
which any one can appreciate without having seen the original. A support under the
raised left heel exactly corresponds to a similar support in the Belvedere statue. Now
this detail is throughout antiquity specially characteristic of marble work. When one
foot of a statue was set back with raised heel, it was usual to leave under the foot the
piece of marble connecting it with the plinth. This was done to afford support to
the foot and to prevent it from breaking. In a metal figure, which is cast separate
from the base, such a support is absolutely meaningless. It is therefore never found
in real antique bronzes, and its presence here satisfactorily shows that the statuette is
a copy of the Vatican figure.
Again, the statuette is very badly cast and full of small holes.8 These are not
repaired in the usual antique way, i.e. with small rectangular patches, but roughly filled
1 A. Gercke, Jahrb. d. Inst. 1887, 260; O. A. Hoffmann, 'Aegis oder Bogen?' Jahresber. d. Lyceums zu
Melz, 1887. 2 Herrn- Apollo Stroganoff, Marburg 1889.
3 Bulla della Commiss. Arch. Com. di Roma, 1889, 407, 451.
4 Vol. i. p. no. [For Helbig's latest views see Museums, 160. — E.S.]
' Jahrb. d. Arch. Inst. 1892, p. 164.
6 I at once communicated this discovery to Kieseritzky, who later passed it on to Conze. In a note
in Jahrb. d. Inst. 1892, p. 164, Conze pronounced my view to be worthy of consideration, but did not express
any decisive opinion of his own.
7 Had the bronze been found at Paramythia, as Stephani contended, one would expect to find on it the
specially beautiful patina of those bronzes. Its patina is not only not antique, it is merely counterfeit modern.
8 Cf. the forged Amazon bronze in Verona. See p. 137, note 1.

THE STROGANOFF BRONZE 407
up from inside with molten lead, according to a favourite method of modern forgers.
Not long ago there was offered to the Berlin Museum a forged Herakles bronze,
showing similar holes filled up with lead. The Stroganoff statuette was apparently
broken and carelessly put together with lead, but the parts — including the left arm —
are of the same date, and evidently belong together. The mantle was never better
finished than it is now ; on the arm there is only a small fissure patched up with
lead, but no traces of solder to show that the mantle once covered it. The original
intention probably was to cast the mantle with the forearm separately and fasten them
on, but this intention was never carried out, no doubt because the crowded thin folds
of the mantle were difficult to render. In the sixteenth-century bronze copy in the
Louvre of the Belvedere statue (by Durant), the left arm with the whole front part of the
drapery is separately cast.1 The so-called ' aegis ' in the left hand of the statuette is
cast to imitate a breakage, and the patina is false. The piece is slightly scolloped out
at the edge, but otherwise there are no traces of chisel work. The head, also unchiselled,
is poor and common, and like the body bears every trace of being, only a slavish copy
by a mediocre workman. What he meant by the fragment in the left hand we do not
know, and need hardly inquire ; it is evident that by this divergence from the original
the forger, who knew quite well the hand of the Apollo to be restored, intended to
make the statuette more interesting. And in this he has doubtless succeeded, since
for over thirty years scholars have been mystified by his bronze.
Another forged copy, or rather variant, of the Vatican statue is to be seen at
Saragossa.2 I know it only from illustrations. . The left arm is raised with a dramatic
gesture ; the drapery on the arm is omitted, but hangs down over the body as in the
Stroganoff figure. The chlamys (here with a fringed edge) is. drawn across so as to
cover the pubes — a trait which alone suffices to betray the modern origin. The fingers
are cast to imitate breakages.
Having now disposed of the Stroganoff .figure, we need have no doubt whatever
that the Belvedere Apollo with the quiver on his back held the bow in his left hand,
possibly also the end of an arrow grasped in one finger. But neither was his right
hand empty : it. bore an attribute, a portion of which is still to be seen at the upper
end of the antique part of the tree-stem, although oddly enough this has not been
noticed.3 The right arm is broken, once in the upper arm and once at the elbow.
The upper arm is antique, the forearm, as marble and workmanship show, is restored.4
To the same restorer is to be traced the piece which unites the tree-stem with the hand
(the fingers of which are restored in plaster). A large rectangular puntello on the
thigh shows that the lower part of the right arm was originally somewhat more raised
and more to the front. The connexion between the tree-trunk and the hand was
effected by the attribute held in the hand ; the end of this attribute may be seen worked
in relief on the tree-trunk ; it consists of a few laurel-leaves 5 pointing downwards, and
four ends of the peculiar knotted woollen fillets which formed the usual adornment of
1 Here, as in the Stroganoff bronze, the support under the left foot is cast with the figure.
2 Gascon de Gotor, Zaragoza (1891), pi. to p. 56.
3 Except by K. Bdtticher (Erklar. Verzeichniss der Abgiisse in Berlin, 1872, p. 323) who was, however,
mistaken in supposing the right hand genuine. My own observations were made on the original, and are quite
independent of his. The tree-trunk, although broken, was in one piece with the round plinth and the right foot.
It has been fastened on again, not quite accurately. Freerichs (Apoll. v. Belvedere, Paderborn 1894) is mistaken ill
believing that the plinth, the tree-stem, and the feet do not belong to the statue.
* Proved by Petersen, Jahrb. d. Inst. 1890, 51. In the small catalogue of the Vatican, ed. 1875, p. 236, the
lower arm is already noted as a restoration.
5 In the Apollo statue, Berlin, No. 51, a laurel-branch likewise formed the connexion between hand and
tree-trunk.

408 THE APOLLO OF THE BELVEDERE
Apollo's Omphalos and Tripod.1 Doubtless, then, the right hand bore a laurel-branch
with stemmata, the well-known Apolline symbol which the archaic Apollo on early
coins of Kaulonia raises in his hand, and which, together with the bow in the left hand,
continued to be frequently given to the god during the severe period, and right
down into late antiquity.2 The knotted fillets also occur frequently on coins. Apollo
holds them in his hand, sometimes with the laurel-branch, sometimes without.3 The
meaning of this attribute is always the same ; it symbolizes the cleansing, healing, and
expiating function of the god, just as the bow designates him as ' Far-darter.' The
two attributes in conjunction characterize him as the god who wards off darkness and
evil disease. So Herakles, in the Orphic hymn,4 scares away6 with his branch
and his arrows evil spirits — the /ea/ca? aro? and the icfjpas ^aXe-n-a?. The motion
of the god indicates his passage through his realm, and the turn of his head
signifies that he is ever on the alert, ready to bring protection and succour
wherever the oppressed call him. Thus on Persian coins the Great King appears
armed with the arrow and striding swiftly like a divinity bringing help to his
subjects. In a previous treatise 6 I tried to show that no definite ' situation ' is expressed in
the statue (a view now further substantiated by the discovery of the laurel-twig), and
that an older, possibly a fifth-century, type underlies this conception. Since then
Winter has published his theory — in my eyes absolutely convincing — that the
statue is a copy of a work by Leochares.7 It is in the Ganymede of this artist
that the unusual gliding gait of the Apollo, the rhythm in the movement of
the limbs, as well as the bodily forms, have their nearest analogy; the resem
blance between the two is so strong and so personal that we are justified in
referring them to the same artist.3 Further, there is also a type of Alexander, which
has been recently shown to be probably derived from Leochares,9 and which
manifests a decided likeness to the Apollo Belvedere in the form of the neck,
the curls behind the ear, and above all, as it seems to me, in the shape of the
eye and the pressure of the eyeball against the outer corner.
In any case I agree with Winter in separating the Apollo from Hellenistic work
and in placing it in the Attic school of the fourth century. This is a conclusion which
cannot be disputed. Even details like the hair tied in a bow,10 the feet and the sandals,
1 Cf. tripod on coins of Kroton. The kerykeion is also found with these knotted fillets. See Wiener
Vorlegeb. 1890 — 91, Taf. 10, also the holy tunny-fish from Kyzikos, Brit. Mus. Cat. of Coins, Mysia, PI. 3, 20,
and other sacrificial animals and utensils.
2 Cf. in particular the fine coin of Side, of severe style, Gardner, Types, PI. 10, 6 ; Oberbeck, Apollo, Coin-
plate 3, 52 ; coins of Metapontum (Overbeck, loc. cit. 3, 9) and of Athens (ibid. 3, 29, 33), Samml. Sabouroff,
Text to Taf. 8 — 11, p. 3, and the Berlin statue, 51.
3 See Overbeck, Apollo, Coin-plate 3, 48 (Magnesia), 49 (Aegae), and 51 (Myrina). Eros on a gem (motive
similar to the statue of Parion) carries in his right hand a twig with knotted fillets. (Cades, Amore, 55. Cf.
Roscher's Lex. i. 1359.)
4 Orph. Hymn, 12, 5. Cf. Rohde, Psyche, i. 219, note.
5 An interesting R.F. vase from Greece (in Berlin, recently acquired) shows Herakles seizing a ker, repre
sented as a small winged daemon, by the throat and threatening it with his club.
6 Arch. Ztg. 1882, p. 251.
7 Jahrb. d. Arch. Inst. 1892, p. 164. Overbeck's answer in Sachs. Berichte, 1892, p. 34, is not of much
consequence, seeing that O. considers the Apollo to be standing much in the same way ('nicht viel anders') as
the Doryphoros of Polykleitos.
8 The head of the Ganymede (quoted as an additional proof by Winter, p. 176) must be left out of the
argument, as its genuineness is by no means certain.
9 Kopp, Bildniss Alexanders des Grossen, p. 18, Taf. ii. Head of Munich statue.
10 For undoubted examples on originals of the fourth century, see Samml. Sab. Text to Taf. 22 ; Ath. Mitth.
1885, Taf. 8 and 9. This fashion of hair will be seen from those instances to be peculiarly characteristic of the
Praxitelean epoch. Also on a head from Aegion in the Nat. Mus., Athens (Kabbadias, No. 192), on an Attic

THE STEINHAUSER HEAD 409
resembling those of the Olympian Hermes,1 and the conformation of the body,2 all
point to the later period of Skopas and Praxiteles or to the earlier period of their suc
cessors (Leochares was a somewhat younger contemporary of those artists, seeing that
he worked with Lysippos on the ' Lion-hunt ' of Alexander). The strong resemblance
between this Apollo and the Diana of Versailles (p. 328), which once led scholars to the
false assumption that the two statues formed a pair,3 is now explained by the fact
that they belong to the same period — circa 330 B.C. — and to the same artistic school.
We have already shown that the ' Diane ' has many features in common with the later
Praxitelean statues of Artemis. The rhythm of the figure and the gliding gait are
so closely related to the Apollo that she too may not impossibly be a creation of
Leochares. The Belvedere statue is a careful and evidently a faithful Roman copy ; the
presence of the tree-stump, which jars on the eye, makes it reasonable to suppose that
the original was of bronze.4 The copyist seems to have done his best to reproduce
the details of this original, and after the experience we have gained of ancient
copying there is no reason to doubt his trustworthiness. A second replica exists
in the so-called ' Steinhauser ' head,5 one of those hasty and inexact marble copies
from bronze which only reproduce the essential features. The copyist has in this
case treated the hair very cavalierly, merely inadequately indicating it on the top
of the head and leaving it as a solid mass behind.6 So too he has omitted all the
finer modelling of the features, and confined his chisel to the principal forms ; and
yet, like many copyists of the good, i.e. the Augustan, period, he has succeeded in
reproducing the expression of the face and the characteristic light and vigour in the
eye, so that, before the technical methods of antique copying were as well understood
as they are at present, this head was supposed, especially from the characteristic
although sketchy 7 treatment of the eye, to represent an older variant of the Belvedere
Apollo type, while in reality it is simply another copy of the same original, executed
in a different manner.8 We must do our best to imagine the original with the help
of both replicas. For the eyes we should rely on the Steinhauser head, at the
same time neglecting its lifeless mouth, empty forms, and clumsy hair in favour
of the more exact rendering to be found in the Belvedere statue.
So far we possess no good second copy of the torso, though I hold it not
impossible that some day a replica will turn up without the chlamys. I cannot
help suspecting that this drapery is an addition made by the copyist, like the chlamys
grave-relief, seated female figure (Coll. Jacobsen, Ny Carlsberg Glypt. 1060 a), and on two of the Muses on the
Chigi reliefs, Petersen (Rdm. Mitth. 1893, Taf. 2, 3 ; p. 73).
1 Cf. Winter, loc. cit. The formation of the toes should also be noticed. The sandals do not, as I formerly
quite erroneously supposed (Arch. Ztg. 1882, p. 251), indicate Hellenistic origin.
2 I also associated the bodily forms quite incorrectly (loc. cit. ) with the Pergamene Apollo. On a closer
comparison one recognizes the strong difference between them ; the Pergamene work bears traces everywhere of a
common external naturalism. 3 A mistake in which I also shared (loc. cit.)
4 The whole polish of the flesh, however, seems modern ; on the right side the original surface is better
preserved ; hair and drapery are not polished.
5 For literature see Friederichs-Wolters, Gipsabg. 1525.
6 This kind of careless copying must not be confounded with that broad touch of genius which often ignores
details, and which may be observed on many original works. The difference in the hair, however, had formerly
made me suppose, though quite incorrectly (Arch. Zeilg. 1882), that the Steinhauser head must be the copy of a
different original.
7 The left tear-gland, for instance, has come out abnormally large ; at the same time, the part below the left
eye is very good.
8 The difference in scale, so much emphasized by Kekule (Arch. Ztg. 1878, p. 9), is slight. The ' Steinhauser '
copyist worked hastily, and may have cut too deep into the marble, thus making a reduction of all the measure
ments necessary. Winter still overestimates the Steinhauser head. 3 G

410 THE APOLLO OF THE BELVEDERE
of the Meleager, which it resembles in style and treatment : it neither harmonizes
with the quiver on the back, nor do its petty folds and its artificial stillness suit the
splendid swing of the whole conception. In turning from the question of copies
back to the original, we next have to consider whether this Apollo descending like
the storm and vigorously turning his head to the side was a new creation now for
the first time finding expression in art.
It was, as I have already shown,1 from a pictorial composition that Leochares
derived the idea which he worked out so boldly and skilfully in his Ganymede. True,
he had to give up the beautiful and natural motive in which the boy's arms are
clasped round the eagle's neck, and, though the motive he chose was less appropriate,
it yet afforded a fine rhythmic motion, essentially the same as that of the Apollo,
and still more similar to the dancing Satyr, which is possibly by Leochares, and
certainly of his school.2
Now coin-types, and notably those of Amphipolis,3 make it evident that,
like his Ganymede, the Apollo of Leochares also was inspired by an older model.
The beautiful silver coinage of Amphipolis with the Apollo head facing is com
prised within the period between the foundation of the city in B.C. 437 and its
capture by Philip in B.C. 358. Among these coins the latest and softest in style
(they are further characterized by the inscription 'Ap.cpnroXiTobp and the symbol
of a Boeotian shield) have been assigned, with much probability, to the period of the
Theban hegemony (371 — 362).4 On the other hand, the coins of severe style bearing
the shortened inscription 'Ap,<pt, should be dated during the first years of the city's
existence.6 Here the head of Apollo is in a tranquil attitude, and there is no indication
of any chlamys. A third series of coins, still more beautiful and varied than the
other two, are inscribed 'Ap^cpLiroXoTecov,6 and probably belong to the period between
370 and 430 — 420 (Plate VI. 24, 25). Here the Apollo head, also facing, is turned
decidedly to the left, and on some examples the chlamys is indicated. Practically the
same type, with the head turned to the side, the floating curls, and the chlamys,
appears on some magnificent coins of Klazomenai (Plate VI. 27, 28, 29) 7 of the
same date, and again on coins of Miletos (Plate VI. 26).8 This was the type which
influenced the heads of Helios on the earliest coins of Rhodes, struck, as we must
suppose, soon after the founding of the colony in 480 B.C.9
The head of Apollo on the coins seems based upon a statuary type representing
the god in a vigorous, striding attitude, and with head turned to the left — a figure
corresponding in all essentials to the statue of Leochares. Now there was a proto
type for this statue, actually among the works of the later Pheidian epoch ; the Apollo
of the coins belongs to that series of heads, also represented facing and with streaming
1 Samm. Sabouroff, Text to Taf. 147, p. 5.
I have dwelt on the connexion with the Ganymede, loc. cit. For replicas (a statue in Naples and two
statuettes from Thessaly and Pompeii respectively), see Satyr von Pergamon, Berliner Winckelmannsprogram,
1880, p. 14, Taf. 3, 2. s Arch. Ztg. 1882, p. 252.
4 De Witte, Rev. Numism. 1864, P- 100. Von Sallet (Beschr. d. Ant. Miinzen in Berlin, ii. 34) disputes
this chronology, and dates the coins among the earliest. But the recent style speaks for De Witte's view.
5 In this case also the style is quite decisive. The shortened inscription corresponds to the older custom.
Von Sallet (loc. cit. p. 36) recognizes the severity of the style, without however drawing the right conclusion.
Specially severe are the small coins with the head in profile and the legend Ajufii. Nor can I agree with Head's
supposition (Hist. Num. p. 190), that the coinage only began after the town was taken by Brasidas : like the
similar colony of Thurii, the city must have had the right of striking its own coinage from the first.
0 Good reproductions in Percy Gardner, Types, PI. 7, 11 ; Head, Guide, PI. 21, 7, 8; Berliner Cat. ii.
Taf. 3, 22—25. A beautiful example, newly acquired, in London, Num. Chron. 1892, PI. 2, 6.
7 Head, Guide, PI. 19, 24—26 ; Hist. Num. p. 491.
8 Brit. Mus. Catal. Ionia, PI. 21, 8. Chlamys indicated. 9 Head, Guide, PI. 20, 37, 38.

FIFTH-CENTURY PROTOTYPE

411

hair, in which we have already recognized dependence upon the style of Pheidias
and his contemporaries {supra, p. 105 sqq.) Since Amphipolis was founded under the
auspices of Perikles, it is not surprising to find Athenian artists of the school of
Pheidias cutting Athenian types for her coinage.
But we also possess marble copies of a work, or at any rate of the head of a work,
which was a prototype of Leochares's conception. Two copies of this earlier work
exist, one in the British Museum (Fig. 177) x and one in Naples, where it is wrongly
placed on the torso of a wounded hero and misnamed a gladiatore2 The work is

Fig. 177.— Head of Apollo (Brit. Mus.)

in reality a youthful Apollo, well characterized by the full hair and the forehead
broadening at the base. The analogy to the Belvedere statue consists not only in
these general features but specially in the animated turn of the head to the left and
the bright steady gaze of the eyes, which almost seem to emit rays of light. But
a long interval of time separates the two conceptions. The London and Naples
heads have the firm simple planes, the large and sharply defined forms of the older
style ; the Belvedere head shows richer modelling and smoother transitions. In the
1 I noticed the resemblance of this head to the Belvedere Apollo on my first visit to London in 1881 ;
Fig. 177 is from a photograph taken at that time. Kopp has since published (Bildniss Alexanders des Grossen,
p. 24) two other views of the head.
2 Museo Borbon. v. 7 ; Clarac, 865, 2203 ; Photograph Sommer, 1582. A portion of the neck is also
preserved.

412 THE APOLLO OF THE BELVEDERE
earlier head the mouth seems to be put on from outside, in the later it is well knit
with the rest of the face, and its edges melt impalpably into the surrounding surfaces.
Again, the eyes of the earlier head show no trace of the careful modelling of the
region around the eye, so telling as a vehicle of expression ; there is no indication
of the lower edge of the eye-socket, but the cheek continues unbroken up to the
groove below the eyeball ; the lower lid is broad and hard, and the part between
the upper lid and the upper edge of the eye-socket is modelled according to old
convention. Instead of swelling out gradually towards the temple, it is represented
as a compact roll of flesh separated by a slight depression from the eyebrow, which
continues in a hard sharp ridge as far as the temple. All these are marks which
distinguish Pheidian from Praxitelean work.
But the type of this British Museum Apollo is not unfamiliar to us, for it is
closely connected with a series of heads which we have previously discussed (p. 99
sqq.) ; the whole expression of the head and its turn, the form of the open mouth
with the powerful curve of the lips, recall the Dioscuri of Monte Cavallo and the
works related to them. Above all, the analogy which the hair presents in its forms
and its peculiar asymmetria to the heads ascribed to the Elder Praxiteles {supra,
Figs. 42, 1 ; 43, 44) seems to me so personal and individual as to warrant us in
attributing the London head to the same artist.
Doubtless the whole movement of the statue had nothing of the gliding
grace of the Belvedere figure. We may imagine a motive more in the manner of
the Monte Cavallo Dioscuri, though not so violent. The right knee was probably
more bent, and a real stride represented. An Athena of the Pheidian school, copied
in a statuette of Epidauros,1 very well shows how artists of that time conceived of a
divinity hastening to give succour and aid. The head is turned with an animated
gesture, the knee of the striding leg is bent and well advanced ; it was the singular
merit of Leochares to impart to the attitude grace, lightness, and elasticity.
In this Apollo of the Pheidian circle — presumably by the Elder Praxiteles —
the god is represented as Healer and Saviour, striding like the Athena Hygieia of
Epidauros,2 or like that earlier Apollo of Kaulonia who waves the laurel-twig in token
of expiation. On the Akropolis of Athens stood a bronze statue of Apollo, called
' Parnopios ' because the god had saved the land from a plague of grasshoppers. The
figure was 'said to be' by Pheidias {to aya\p,a Xeyovo-tv QeiStav iroir\o-ai, Paus. i. 24, 8).
That it was certainly not signed we may conclude from Lucian's remarks on the
Athena Lemnia.3 Like other works attributed by Pausanias to the master, it may
have been by one of his pupils or colleagues ; for instance, like the Promachos, which
Pausanias also attributed to Pheidias, it may possibly have been a work of the elder
Praxiteles, and be identical with the statue of ' Apollo the Saviour ' by that master.
Some hundred years later, when Leochares had a similar commission to execute, he
adopted the Pheidian type ; an Apollo by him at Athens formed, we know, a pendant
(of like significance, probably) to the Alexikakos, the warder of evil, by Kalamis, in
front of the temple of the Patroos.
1 'Eip-nfi. dpx- 1886, Taf. 12 (left hand) ; Ath. Mitth. 1886, p. 309 (Petersen). Not only the helmet
but the hair and type of face are Pheidian.
2 Petersen's explanation (loc. cit. ) seems far-fetched. The goddess is simply hastening to help in need. For
the torch on the basis of the relief, see Friederichs-Wolters, 11 76.
3 Cf. supra, pp. 10, 32, 62. The Xeyovaiv of Pausanias (on the strength of which Michaelis in Ath.
Mitth. ii. p. 1, note 2, reckons the Parnopios as doubtful) does not in itself necessarily imply anything at all,
as Gurlitt (Pausanias, p. 193) has shown, for even when he is quoting an inscription Pausanias not unfrequently
introduces it by fcurlv. — e.g. in the ease of the Aphrodite by Kalamis, i. 23, 2.

APPENDIX
THE TEMPLES OF ATHENA ON THE AKROPOLIS

APPENDIX
THE TEMPLES OF ATHENA ON THE AKROPOLIS
IN order thoroughly to appreciate the sculptured de
corations of the temples of the Akropolis, or — to go
no further — of the east frieze of the Parthenon,
which, in a preceding inquiry (see page 47), we have
confidently ascribed to Pheidias, it is necessary to attain to
a clear conception of the significance of the temples and of
their mutual relations. The solution of this problem has
been the object of specially active endeavour in the last ten
years, a period which has given us the discovery by Dorpfeld
of the oldest of those temples. There still remains, how
ever, a sufficient number of obscure and puzzling points, to
the elucidation of which I shall endeavour to contribute in
the sequel. I. The 'Old Temple' of Athena.
Even since the finding of Dorpfeld's ' Old Temple ' the
predominant view is that the Erechtheion occupies the site
of a temple of the oldest times. (See annexed plan of
temples, Fig. 178.) Dorpfeld himself inferred at first from
his discovery merely that this supposed ' older Erechtheion ' was somewhat smaller than the one
actually in existence, since the Porch of the Caryatids projected over part of the substructions of the
newly discovered temple.1 Later, he so far modified these views as simply to suppose that a small
Temple of Erechtheus had once stood on the site of the Erechtheion, immediately to the north
of the old Athena Temple.2 Wachsmuth conjectures that the Erechtheion and its assumed older
predecessor occupied exactly the site of the Megaron in the old Royal Palace on the Citadel.3
Michaelis supposes an old double shrine on the site of the Erechtheion.4
These suppositions are all without external support from facts, since no trace exists of
an older building on the site of the Erechtheion, with the exception of the piece of the south
side where the great ' Old Temple ' lies under the Erechtheion, and of a few blocks on the
west -side which belonged to a smaller, older, though not pre-Persian building or enclosure.
Although it is true that other walls of an older date might have disappeared, yet Wachsmuth's
conjecture is highly improbable, on account of the great unevenness of the ground on
which the Erectheion stands. The ancient Megaron certainly did not lie in this place. The
altar of Zeus Herkeios in the Pandroseion is no doubt, as Wachsmuth supposes, the same
that in the old kingly times stood in the palace courtyard, but the royal abode did not necessarily
stand so near the altar as does the present Erechtheion : it should be placed on the more level plat
form of rock farther to the east. In this place excavations have, in fact, brought to light remains of
cyclopean walls so stately as fully to justify us in supposing that we have here the actual palace. Nor
does our literary evidence lend support to the predominant supposition that in old times there
1 Ath. Mitth. xi. 350. 2 ^id. xii. 199.
3 Berichte der Sdchsisch. Gesellsch. 1887, p. 399 sqq.
4 Altattische Kunst (Speech, 1893), p. 16. E. Szanto (in the Arch. ep. Mitth. aus Oesterr. xiv. 116), starting
from this unproved assumption of an older Erechtheion beside Dorpfeld's temple, even attempts to separate the
Athenas of the two temples : the one, he says, is Polias, the City-Goddess, the other is the goddess tav A%i\vav
ineliovaa, She that cares for the City. In the year 480 Themistokles persuaded the Athenians to believe that the
' City-Goddess ' had fled, and that therefore they must flee, but that ' She who cares for the City ' had remained
in the other temple (!)

416 THE TEMPLES OF ATHENA ON THE AKROPOLIS
stood beside the great Athena temple a second temple, corresponding in the main to the later
Erechtheion, although of smaller size. The two well-known passages of the Iliad and the Odyssey
(which, however, can almost certainly not be dated earlier than the sixth century1) attest the
existence of only one shrine upon the Citadel ; this is held by Erechtheus and Athena in common.
For, according to II. ii. 549 sqq., the cult of Erechtheus takes place in the temple of Athena,2 where the
goddess has assigned him a place. When, in Od. vii. 81, Athena betakes herself to the house of
Erechtheus as to her home, this can only be that same shrine, which accordingly belonged to
Erechtheus as much as to Athena, and could be called after either of the two. By the ' house of
Erechtheus' is meant no royal palace, but a place for worship : this is shown by that very passage of
the Iliad, in which Erechtheus has his place of abode assigned to him not as the king but as a person
who is an object of worship. There can be no doubt that in both passages the sixth-century poet is
alluding to a double shrine of Erechtheus and Athena on the Citadel, which shrine was well known to
his listeners.3 Now this double shrine of Erechtheus and Athena, attested for the sixth century by the
passages from Homer, can be no other than the great ' Old Temple ' discovered by Dorpfeld.
This is no ' Temple of the Tyrants,' as Petersen 4 and Curtius 5 suppose, beside which
it wduld be necessary to imagine the real shrine standing on the site of the later
Erechtheion ; only the peristyle is from its construction evidently not earlier than the
'Peisistratan' period; but the naos itself is so distinct and so much older in the technique and
material of the foundations, that in all probability it is considerably more ancient.6 Moreover, there
lie below it remains of a yet earlier building,7 so that, even if we concede the possibility that the naos
dates from no further back than the age of Peisistratos, yet we have to suppose a predecessor to it on
the same site. The oldest witnesses, the passages from Homer, in spite of their antiquity, have in
view at any rate the stately naos Hekatompedos with its double cella. The peristyle, judging
by its architectural forms and its decoration, can hardly have been built under Peisistratos himself :
it was built at the earliest under Hippias, or perhaps not till the period immediately following his
death.8 Now this temple is quite clearly a double temple : it had two separate sections, the
one approached from the east, the other, consisting of three chambers, from the west,
an arrangement exactly suitable to the double worship of goddess and hero.9 But, it may be
objected, this western section is merely a treasury. This certainly has been unanimously asserted
since the discovery of the temple,10 though I fail to see that any one has produced a single proof.
The plan of the temple is, as is well known, quite unusual, the only analogy to it being the
double temples in which two deities are worshipped back to back,11 and of which the Erechtheion is
a notable example. That there ever existed temples which, in addition to the usual Opisthodomos
1 Cf. especially von Wilamowitz, Homerische Untersuch. p. 247 sqq.; Aristoteles und Athen, i. 239.
2 filv in II. ii. 550 can refer only to Erechtheus ; cf. for a recent discussion Rohde, Psyche, p. 127, note I.
3 Cf. von Wilamowitz, loc. cit. p. 248 ; also Walter Millerin Amer. Journ. of Arch. 1893, pp. 478, 525, and
J. G. Frazer in/. H. S. xiii. p. 167. 4 Ath. Mitth. xii. 64 sqq.
5 Stadtgeschichte, p. 71 sqq. So also Michaelis, Altattische Kunst (Speech, 1893), p. 16.
0 Cf. Ath. Mitth. xi. 344, xii. 209. The foundations of the naos, constructed of limestone from the Citadel
hill, show a style closely resembling that of the ' Cyclopean ' walls of the earliest period.
7 Ath. Mitth. xii. 61.
8 The architectural forms especially resemble those of the Aeginetan temple, which belongs to the period
500—480 (cf. supra, p. 19, n. 2), and those of the Temple of Zeus at Olympia (cf. Dorpfeld, Olympia, vol. ii.
Baudenkm. p. 20). The style of the pediment sculptures, and especially the ornamentation of the Sima, corre
spond to the vases of the fully developed R.F. style. It may accordingly be asked whether we ought not to put
the peristyle later, after the time of Hippias, perhaps in the period about 507, in which the Athenians won their
great successes over Thebans, Chalkidians, and Peloponnesians, and set up the quadriga in honour of their victory
over Chalkis. Cf. with regard to the great development of Attic art after the time of the Peisistratidai, my
remarks m the Berl. Phil. Wochenschr. 1894, p. 1278. Michaelis (Altattische Kunst, p. 26 ; cf. p. 16) dates the
peristyle in the time of Hippias, the naos in that of Peisistratos. He conjectures that the two great sculptured
pediments of poros stone belonged to the latter : it is possible that they did, but in that case they would clearly
point to a period earlier than Peisistratos.
9 Herodot. viii. 55 shows that the space lying to the north below the great temple, and containing the Olive
and the Salt Spring, was specially dedicated to Erechtheus. This does not necessarily clash with our view that
Erechtheus was worshipped in the great temple, as a house-mate of Athena. If we are correct in our identification
of the double shrine alluded to in Homer with the ' Old Temple ' now extant, we are justified in introducing a
slight emendation in the manuscript text of Herodotos : for 'EpexBeos vr)As read rather 'Ep. o-tjkos. The fact that
the small enclosure below the temple which contained the 'tokens' was sacred to Erechtheus proves nothing
against his having been worshipped in the west half of the great temple.
Cf. as recent instances, Michaelis, Altattische Kunst (Speech, 1893), p. 16 ; Fowler, Americ. fount, of
Arch. 1893, 7 sqq.
n Cf. e.g. Paus. ii. 25, 1, Ares and Aphrodite, the former looking west, the latter east ; also Venus and
Roma at Rome.

THE 'OLD TEMPLE'

417

corresponding to the Pronaos, had as a treasure-chamber a second cella approached from the west,
is a modern supposition,1 supported, so far as I can see, by no building either now extant or known
from literary evidence.2 The Parthenon is the one and only temple on which the conception was based ;
but, as we shall see presently, it was owing to special circumstances that the second cella in this case

GROUND PLAN
OF THE FOUR
GREAT TEMPLES
ON THE
AKROPOLIS
[from antike denkmaeler]

OLD TEMPLE OF ATHENA
[DESTROYED BY THE PERSIANS]

OLD TEMPLE OF ATHENA
[DESTROYED BY THE PERSIANS]
FIRST PARTHENON
PARTHENON OF PERIKLES
ERECHTHEION

PARTHENON OF PERIKLES

Fig. 178. — The four great temples of Athena on the Akropolis.

contrary to its original destination, became a treasury. At any rate, unless the opposite is distinctly
proved, it is impossible to assume of a temple cella that it was used for any but religious purposes.
1 Thus, Curtius says (Stadtgesch. p. 132, note): 'The temples, in accordance with established usage,
always had treasure-chambers behind the cella ; ' but he gives no proof whatever. The 'back chamber' of the
temple at Corinth, to which he refers, seems more likely to be the open porch of the western cultus-cella (cf. the
following note). The 6intT0ASop.oi trrrjXai of Polybios (xii. 11,2) are records of state, which it was the custom to
keep in the usual open back porches, the lmiaB6Sofi.oi of the temples. They have of course no more connexion
with treasure-chambers than have the decrees of Proxenia on the pillars of the temple which Polybios mentions
at the same time. Wilamowitz (Aristot. und Athen, i. 306) suggests rds biriirdoypdipovs or^Xas-
2 The old temple at Corinth was, as Dorpfeld has proved, a double temple, with two separate cellas
facing east and west ; and it just happens that here the character of the latter as a place for worship is put
beyond doubt by the preservation of the basis for the cultus-image, as Dorpfeld has shown (Ath. Mitth. xi. 302,
Taf. vii.) The Temple of the Kabeiroi near Thebes had a spacious back chamber, not accessible from the principal
cella, but the character of this also as a place for worship is made quite certain by the sacrificial trenches (Ath.
Mitth. xiii. Taf. 2, p. 91). 3 II

4l8 THE TEMPLES OF ATHENA ON THE AKROPOLIS
Accordingly, the rooms in the ' Old Temple ' of the Akropolis were presumably, like the rest,
simply places for worship.
But there is the Hekatompedon inscription. It is supposed by its discoverer, Lolling, and
equally by Dorpfeld,1 to afford proof that the western compartments of the temple served only for the
purposes of a treasury ; according to Lolling and Dorpfeld, the large western cella was called to
o'Urifxa rapiewv, and the two adjacent rooms ra oiKijp.ara. But ro o'Urjfia tapielov was not in the
inscription, and is an impossible expression, as has been already pointed out by Dittenberger.2 He
restores thus : ras ^anopovs [prj i'xctv o'i]icqpa rapielov, and translates : ' They are not to set up any
store-room.' Still more correct seems to be Kirchhoff's restoration,3 which has the additional
advantage of corresponding exactly to the number of letters wanting (9) : pfj iroielv o'i]Krjpa ra/iielov
ipTroXei, i.e. the priestesses and other women concerned in the service of the temple were not to
convert any o'Urjpa on the Citadel — the temple in particular is clearly not intended — into a rajxielov, a
private store-chamber. The other passage in the inscription, ta olK^para [ra iv ra 'E«aT]o^7rc'8(B avolyeiv
{rolls'] rapias, does, without doubt, refer expressly to the temple, but proves nothing for the assumed
treasure-chambers. The expression ra. oIk. ra iv rm 'Ex. designates all compartments of the temple
that are of the nature of rooms ; and of these there are four, since the two larger are clearly just as
much oln.rijj.ara as the two smaller. Or if the eastern section, being distinguished by two interior ranges
of columns, is not to be considered a room, there remain at any rate three ohrjpara. That the
term o'Uijpa can be applied to a chamber serving for worship is well known ; we need only call to mind
that Pausanias speaks of the west part of the Erechtheion as an o'lK-qfia 'Epe'y&eov icdKovpevov.1 The
oi<r)jj.ara of the Hekatompedon we can only regard as chambers for worship. Even if we limit the
expression ra oiKijpara to the two smaller rooms only — and we are in no way forced to do this — yet
nothing whatever can be inferred with regard to their purpose as treasure-chambers. The raplai of
the inscription are by no means merely officials of the treasury : they are the supreme board for the
supervision of the Citadel in general ; they have power there in matters of poiice ; in particular the
priests and other persons attached to the temples are subjected to them ; they are responsible for
seeing that nothing irregular occurs. It is therefore a matter of course that they keep the keys to
the chambers of the temple, seeing that it is doubtless they, not the I 'petai and {duopoi, that are answer
able for the objects there deposited — votive offerings, precious images of the gods, furniture, utensils,
and the like. These are the rapiai rot Ipov who, according to Herodotos, in 480 — Kirchhoff places
the inscription a little before that date, in 485 — 484 — were alone sitting on the Citadel when the assault
ing Persians climbed the hill, and of whom some fled into the peyapov, of which they held the keys.
Where they kept the treasure we do not know. The inscription, at any rate, contains no indication.
It was not necessarily in the temple ; the pre-Persian Citadel contained so many smaller buildings —
as is proved by the numerous ancient simae of marble and terra- cotta — that there can have been no
lack of suitable closed places within the sacred precinct.5
We have found, then, not the smallest reason for departing from the rule and looking upon the
chambers of the ' Old Temple ' as anything other than places for worship. The eastern cella, which
is marked out by its size and its interior columns as the most important, seems from the inscription
to have been called 6 veas in a special sense ; the front division is designated to irpovrjwv, the temple
as a whole awav rb 'EKarapireSov. The last expression is certainly of a very ancient character, as may
be conjectured from the nvpr) eKaropnoSos of the Iliad (xxiii. 164). The name refers to the length of the
temple exclusive of the peristyle, and is manifestly older than this, coming down from the days when
the whole building was a hundred feet long. Only at a quite early period could the erection of a
temple a hundred feet in length be a subject for special pride.0 Even the Heraion at Olympia
is considerably bigger, and the mighty temples of ancient Ionia were on a very different scale.
Hekatompedon, then, was the official name for the whole edifice. The eastern half belonged in
a special sense to Athena ; the western half, approached from the west, to Erechtheus. We shall speak
1 Lolling, 'AB-qva, 1890, 627 ; AeXriov, 1890, 92 ; Dorpfeld, Ath. Mitth. xv. 420 sqq. So also Fowler, Am.
Journ. of Arch. p. 9.
2 Hermes, vol. xxvi. (1891) p. 472. 3 C. I. A. iv. 139.
4 Cf. also Paus. x. 4, 4. Therefore I cannot assent to Dittenberger's assertion (loc. cit.) that an essential
part of a temple could not be called ofjcrj/u*.
5 It should be noted here that the existence of separate buildings to contain temple treasures is attested in
several instances ; so the Brjiravpos with door and roof in Eleusis ('Etprjp.. apx. 1883, p. 126, 1. 4 ; p. 118, 1. 73;
p. 115, 1. 13); cf. Rubensohn (Die Mysterieuheiligthiimer, p. 65) and the stone structure in Epidauros ('E0rj,u. apx-
1886, p. 147, 1. 230, 243, 300); cf. Swoboda in Wiener Studien, x. 299, note 78.
c There is absolutely no ground for the assumption that the name Hekatompedon in itself points to a ' splendid
innovation,' a 'display of magnificence,' belonging to the period of the Tyrants (Curtius, Stadtgesch. p. 72;
Petersen, Ath. Mitth. xii. 66 seq.) On the contrary, the Olympieion at Athens shows us excellently what was
meant by a temple built for display in the time of the Peisistratids.

THE 'OLD TEMPLE' 419
farther on of the subdivisions of this latter half. The popular conception pictured Erechtheus as
actually dwelling in the temple,1 and his ritual, described in the passage from Homer, may in old
times have been more important in the eyes of the people even than that of Athena.
The 'tokens/ the Olive and the Salt Well, were quite close to the temple. The difficult
problem of bringing even the Salt Well into the architecture of a temple must have been far removed
from the ideas of those old times. The fencing in of the sacred spot, which moreover lay in the
most immediate and most sacred precinct of the temple, was doubtless quite sufficient. We may
compare the custom of fencing round spots struck by lightning, as sacred places of Zeus Kataibates.2
The Olive was, naturally, left in the open air even after the Salt Well was built over : similarly, at
Olympia, the eXaia KaWio-ricpavos stood outside near the temple. The assumption that we must
suppose an older temple on the site of the Erechtheion, if only on account of the 'tokens ' in the
rock, is unfounded. The Olive was certainly always outside. The ' tokens ' were of course
immovable, but the temple was not. In contrast to those cases in which a new temple has been
erected on the foundations of an old one, there are, as is well known, several instances of the opposite
proceeding, in which another spot near by, within the sacred precinct, was chosen. The shrine of
Dionysos at the foot of the Citadel was one instance ; so was that of Hera at Argos ; and the same
was done at Samothrake and elsewhere. Now each point within the lepov was equally dedicated
and appropriated to the deity ; it was therefore a matter of indifference whether his image and
temple were set up on this or on that part of the precinct within which he dwelt. Only it was
customary on the occasion of any change affecting the possessions of a deity, whether his temple, his*
image, or other votive offering, to avert the possible displeasure of the deity by means of a kind of
propitiatory sacrifice, called dpecrrrjpiov.3
The old Hekatompedon, then, was until 480 the only temple of the City Goddess, and her
companion Erechtheus, upon the Citadel. In its east cella stood the olive-wood image of the
goddess ; this was not necessarily older than the temple itself, since the fables of later writers, who
represent it as fallen from heaven, as set up by Kekrops, or Erichthonios, as made by Endoios, the
pupil of Daidalos,4 have about as much value as the Christian legends ascribing a host of pictures to
St. Luke. They prove only that the image appeared to those of later days to be very old. The date
of the temple exclusive of the peristyle is to be placed in the long interval between the Mykenaian
and the Peisistratan periods : in thus placing it we are guided by the style of the foundations. The
artistic forms of the building are unfortunately quite unknown.

II. The First Parthenon.
The second great temple of the Citadel is the older Parthenon, the substructions of which lie
under the building of Perikles. Dorpfeld has lately shown 5 that this older Parthenon was not erected
till after the burning of the Citadel in 480. The most natural supposition is that it was to replace the
'Old Temple' burnt by the Persians. Unfortunately we cannot now know with certainty the
ground-plan of the building, as it has been covered by the later Parthenon. But Dorpfeld supposes,
with great probability, that it had the same number of columns on the short sides, namely eight, while
on each of the long sides it had two more than its successor, i.e. nineteen instead of seventeen. The
older edifice was thus, he thinks, about six metres longer than the later, but on the other hand
somewhat narrower. The idea which we find in the Periklean Parthenon, of making the eastern cella
by itself a eKaropireSos veins, as long as the entire old 'EKaropne dov, was no doubt embodied in the
old construction also.0 The greater length of this last must therefore have been bestowed upon the
western cella. Dorpfeld has pointed out that the difference in length between the two temples just
corresponds to the depth of the small rooms behind the west cella of the old Hekatompedon.7 We
1 Cf. Rohde, Psyche, p. 127.
2 In the Altis at Olympia, Paus. v. 14, 10. On the Akropolis at Athens, tseXriov apx- 1890, 144. In the
Olympieion at Athens, 'Eit>np. apx- 1889, 61.
3 Cf. on this point Kdhler, in Hermes, xxvi. 1891, p. 44 seq., and the inscriptions he cites.
O. Jahn, De Antiquiss. Minervae Simulacris, p. 9 seq.
5 The recent attempt of Penrose to revive the earlier view of the pre-Persian origin of the substructions
seemed finally refuted by Dorpfeld (Ath. Mitth. xvii. 1892, p. 158 sqq.) In/. H. S. xiii. 32, however, he has
tried once more to defend his views against Dorpfeld, but without bringing forward any new material that could
alter his opponent's results. The decisive argument is afforded by the character of the filling of debris on the
south side of the Parthenon. Any one who had opportunity to observe this filling during the excavations must
fully agree with Dorpfeld : the layers of building debris are beyond uncertainty. Now, if the debris is contem
porary with the Parthenon, Penrose's view is thereby refuted.
" Cf. Dorpfeld, Ath. Mitth. xvii. 174. 7 Cf. ibid.

420 THE TEMPLES OF ATHENA ON THE AKROPOLIS
may therefore conjecture that the older Parthenon was intended to have similar rooms. Moreover,
the whole cella was certainly, as Dorpfeld supposes, considerably narrower than the Periklean
building, which needed so wide a cella simply for the sake of the gigantic cultus-image. It follows
from this that the ground-plan of the older Parthenon was essentially only an enlargement of the old
Hekatompedon, which it was intended to replace ; therefore the worship of Erechtheus, and all that
this involved, was to be transferred to it. If we further recollect that the site of the temple was so
chosen as to necessitate enormous substructions on the southern side, we shall realize that this plan
represented no ordinary degree of audacity. Without regard to the spot consecrated by antiquity, and
the proximity of the divine ' tokens,' the cult was to be transferred in its entirety — of course after the
offering of the proper dpecrrrjpia — to a site on the Citadel which had first of all to be in great part
artificially produced.
No evidence has come down to us as to the date of the temple, or the person who erected
it We know, however, that it must belong to the time between 479 and 447, and that the work
was abandoned before it was finished ; only the lowest drums of the marble columns had been
prepared, and on these the fluting had not been cut. Moreover, it is highly probable that the date
of erection comes very much nearer to the upper limit (479) than to the lower (447) ; the Periklean
edifice is so widely different from the older one, that a comparatively long interval between the two is
probable. The difference is especially clear in the substructions added on the north side. The steps,
moreover, are of marble in the Periklean building, of poros stone in the older edifice. A third and
topmost step of the latter was entirely taken away,1 and a part of the second removed. The marble
drums of columns prepared for the older building were thrown aside, and used in levelling up the
ground about the Periklean building. Some of them had been, probably at an earlier date, built into
the north wall.2 The plan was altered, and the foundations carried farther north. This whole
revolutionary proceeding is certainly easier to explain if the work had lain for some length of time
untouched when Perikles began to build.
The older Parthenon has been ascribed to Kimon, the only difference of opinion being as to the
time of his life to which it belongs. Thus Kopp considets that the work did not begin till after the
return of Kimon from exile, about 4S4.3 But on this theory there would be an interval of only a year
or two between it and the Periklean Parthenon, and that is practically impossible. Dorpfeld,
therefore, conjectured that the building was begun some time earlier, soon after the banishment of
The r, stokles (471). My belief is that we ought to go still further back, and place it in the years
immediately following 479.
The only fact that appears to point to Kimon is that the south wall of the Citadel is closely
connected with Kimon's name, and it is assumed that the substructions of the Parthenon are
contemporary with the south wall, the two being parts of a single design. Contemporary, however,
it is not. The south wall is undoubtedly later than the foundations of the Parthenon,4 and has no
necessary connexion with them. On the contrary, the method pursued in levelling up the ground to
the south of the Parthenon seems to me to show that when the foundations of the temple were built
there was as yet no thought of the great south wall. It was clearly intended in the first place to raise
only the part between the projected temple and the old cyclopean circuit-wall, so as to form a terrace
surrounding the temple. For the support of the masses of up-piled materials it was necessary to
heighten the old circuit wall. This was done by means of the scarped 'supporting wall'6; it
starts to the south-west of the temple from the old circuit wall, but is carried on in a straight line,
cutting off the curve which the other wall makes outwards. This outer wall seems to have been
partially removed in this place ; at least stones from it were used. The supporting wall ended, no
doubt, where the museum now stands, and abutted on the circuit wall, which at that spot turns inward
again. It was probably intended subsequently to face the outer side of the supporting wall with
better stones ; but the wall would never have been erected if the south wall had been planned at the
1 Dorpfeld (foe. cit. 166) thought it more probable that the building departed from the general rule in having
only two steps ; he has now, however, changed his opinion, as he has informed me, and believes it can be proved
with certainty that there were three steps ; and this is, in any case, more probable.
Itisraicertainwlieii this was done. Dorpfeld (loc. cit. p. 189) conjectured that it was at the time of the
battle ot 1 anagra. In answer to a question from me as to whether the drums were necessarily built in before
the 1 enklean Parthenon was begun, he writes that he thinks this must have been so ; because ' if the drums had
not been built in, Perikles would certainly have used them for his new columns, since they are not damaged like
those found near the Parthenon.' 3 Jahrb_ d_ lmL v_ 'l8g0j 2yo_
,, / ; M}tth. xm. 432. Dorpfeld further corroborates this by letter : ' There can be no doubt that the
south wall of the Citadel is later than the foundations of the temple ; at most, its lowest courses maybe contemporary
with the temple. Ihe length of time that intervened between the two buildings I do not venture to estimate.'
Marked with the number 20 on the plan in AeXrlov, 1889, p. 50. Cf. Ath. Mitth. xiii. 432.

THE FIRST PARTHENON 42 1
same time. The latter is more recent, and the space between it and the supporting wall was not filled
up till later.
Now the south wall is dated by the well-known and quite trustworthy testimony given in
Plutarch, Kim. 13 ; it was erected by Kimon after the battle on the Eurymedon from the money
received as ransom for the prisoners. Plutarch here draws from a very well-informed authority, who
purports to be very complete in his account of what Kimon did for the city in the way of buildings
and the like.1 Even the trees which he planted in the Agora and the embellishment of the
Academy are mentioned. The south wall is named as an independent and important construction ;
we have seen that it is such in fact. The old Parthenon was older, and had nothing to do either
with this or with Kimon. Before the battle on the Eurymedon, it would have been scarcely possible
for Kimon to set on foot such a work, if only on account of his continual campaigns. If, however,
it had been begun by him, he would surely after that battle, which brought such abundant booty,
have applied the money to carrying on the temple, not to the erection of the south wall. From the
fact that he did not touch the unfinished temple at a time when he was at the height of his renown,
with enormous sums of money at his disposal, we may safely conclude that the erection of the temple
had been begun by the party adverse to him.
All works of importance in the Athens of those times ought certainly to be judged with reference
to the great parties which were at strife with one another, and alternately gained the predominance.
Kimon ignored the older Parthenon, because it had been begun by the party of Themistokles. The
building must belong to the time immediately after 479, when Aristeides and Themistokles were
the irpoo-rarai. rov brjpov, as Aristotle expresses it (ttoA. 'Af?. 23, 3 ; cf. 28, 2).
As early as the sixties of the fifth century there was a great new temple on the Citadel ; this
follows from the inscription C. I. A. i. 1 (iv. p. 3), if Kirchhoff and Dittenberger2 are correct in their
highly probable restoration of the passage : iv TTeptff\oka>[i ra voroff\ev rov rr/s 'Adrjvaia[s apxalov ve]a>
ipiroXet. For this inscription is, according to Kirchhoff and Dittenberger, without doubt con
siderably older than 460, and equally without doubt later than the destruction of about 480 : it
belongs, they hold, to just that time at which we consider the first Parthenon to have been begun ;
as contrasted with this new building, the temple which it was to replace was the apxdios veas.s This
inscription is also interesting in another respect, as confirming the view that the ' Old Temple '
contained no places at all for the storing of money. The Eleusinian treasure, which is here in
question, is to be kept in a peribolos, near the temple ; and this would assuredly not have been the
case if the temple had had ' treasure chambers.' 4
The plan of restoring the forsaken and destroyed city in 479 undoubtedly included the re-erection
of the shrines. The Athenians had not by any means, like the Ionians, bound themselves by an oath
to leave the ruins standing.6 When the ancient image of Polias re-entered her desolate shrine, it
can have needed no powerful persuasion to induce the Athenians to undertake a splendid restoration.
But in the method pursued we see the work of an original, unscrupulous, and revolutionary mind :
the new building was transferred from the site consecrated by antiquity to the sloping south side of
the Citadel, and in order to widen the hill at this point numbers of damaged architectural and statuary
fragments were used to level up the ground. In all this we seem to trace genuine Themistoklean
detvorrjs and ToXfia.0
1 It has been supposed, with good reason, that the authority was Theopompos. Theopompos is favourable
in his representation of Kimon, whereas he decries Themistokles. 2 Dittenberger, Syll. 384.
3 Dorpfeld (Ath. Mitth. xii. 39) reads o7ri<rflj>>' instead of vor66ev ; he interprets the place to be the Opistho
domos of the 'Old Temple.' The preceding letters, however, oA.a>, he leaves quite unregarded. Curtius
(Stadtgesch. p. 132, note I) likewise reads 8incr6]ev, and before it 4v rjj 6]6Xa. This last restoration is, however,
incorrect, since it takes, no count of the number of letters missing, but inserts a letter too few. And
how could the Opisthodomos of the ' Old Temple ' be called, as Curtius supposes it to be, 66Xos iiriaiev
•roD vaov ? It seems to me indubitable that the passage cannot refer to any part of the temple itself. KirchhofFs
restoration, exactly supplying the lacuna, is at any rate highly probable. — Petersen (Ath. Mitth. xii. 65) rightly
lays stress on the fact that the inscription is, according to Dorpfeld's estimate, older than the Parthenon ; and
from this he only draws a false inference on account of his conception of the ' Tyrants' Temple. '
4 We may conjecture that the same building, to the south of the ' Old Temple,' is intended in the old
Hekatompedon inscription, C. I. A. iv. p. 138, col. ii. 11. 8, 9 : koX vo]r68ev r[ov v]eu> evrbs rov k . . . The
following restoration would exactly supply the number of letters wanting in the lacuna : evrhs tou k[vkXov Kal
icaB'] tLirav rb 'ExartpTieSov. With this agrees the gloss in Hesychius, kvkXos' itepifio\os. Now to what deity can
have been dedicated the circular peribolos to the south of the 'Old Temple'? I am inclined to conjecture that
it was sacred to Ge, whose worship in this place is attested by the well-known rock-cut inscription of later date
(C. I. A. iii. 166 ; cf. infra, VII. p. 468). Perhaps the xbicXos yds of a recently published sacred law from Kos
was likewise a peribolos of Ge (cf. Tdpffer, Ath. Mitth. xvi. 427 seq., who leaves this possibility an open question).
5 That the oath applies only to the Ionians has been proved by Kopp (Jahrb. d. Inst. 1890, p. 272). Puchstein
calls my attention to the fact that the Ionians did actually keep the oath for a long while ; none of the restorations
or rebuildings of the great temples of Ionia belong to the fifth century. ° Plut. Kim. 5.

422 THE TEMPLES OF ATHENA ON THE AKROPOLIS
On this theory we have an excellent explanation of the fact that the ' Persian ddbris' between
the Parthenon and the supporting wall looks so fresh and is of so homogeneous a character : it had
not lain about year after year on the Citadel, but was used for this work immediately after the
destruction. We have an excellent explanation, too, for the discontinuance of the building so magnificently
begun, just as the marble superstructure was beginning to rise ; we see why the work was broken off
and allowed to lie unfinished. The reason was the fall of Themistokles, the definite union of
Aristeides with Kimon, and the beginning of the predominance of Kimon's party. Even as, after the
fall of the Peisistratids, the great Olympieion was left untouched for several centuries, so was it now
with the new temple on the Citadel. The enterprise of Themistokles was no doubt abused as too
expensive, as the scheme of a madman ; the pious and conservative were offended at the purposed
displacement of the old centre of worship, and it was probably said that too much had been begun,
that the first thing was to complete the fortification of the city and harbour. This was the spirit in
which Kimon laid the foundations of the long walls in the marshes of the Halipedon, and erected
the strong south wall of the Citadel. This was assuredly to be no mere 'adornment' of the
Akropolis,1 but a protection on the weak and accessible side. The Pyrgos at the western extremity
of the wall only lost its significance as a fortification by the erection of the Periklean Propylaia and of
the temple of Nike.2 But to the builder of the old Parthenon the Citadel was no fortress ; in the
eyes of Themistokles the Peiraieus was the true fortress of Athens ; though for him, as for others,
the Akropolis was, and must remain, the religious centre. Yet the strife of parties brought it about
that from 479 to 438 the Polias possessed upon the Citadel only a temple provisionally repaired after
conflagration and destruction.
At the time to which we ascribe the beginning of the great temple, -immediately after 479, the old
antagonists, Themistokles and Aristeides, united by the triumph of Salamis, acted in harmony for a
while at least. It was a time at which everything was still guided by the counsels of Themistokles.3
The building of the city wall he accomplished in concert with Aristeides.4 The Parthenon was no
doubt likewise planned with his approval, and probably the party of the Alkmaionidai, politically in
close connexion with that of Aristeides, was also favourable.5 In 479 Xanthippos and Aristeides
were strategoi. The Alkmaionidai must clearly have had an especial interest in the erection of a
splendid shrine to the Polias on a spot other than that desecrated by themselves ; such a work would
necessarily tend to blot out the memory of their ancient crime. And the idea of constructing such a
temple wholly of marble (this was the first instance at Athens) had been already carried into
execution by the Alkmaionidai in the facade of the temple at Delphi. On this view, then, it appears
as specially appropriate that the man who subsequently resumed the work should have been Perikles,
the kinsman of the Alkmaionidai, and participator in their curse, the ayos.
Indeed, by attributing the foundation of the Parthenon, not to Kimon, but to causes resulting from
the position of parties after 479, we place the building in quite a new historical light. We must bring
before our minds the two great parly programmes which, from the time of the city's liberation from
her tyrants, are traceable throughout the course of Athenian politics, and which, to a great extent,
are connected with the ancient families of the Philaidai and Alkmaionidai, and represent their
traditions. Then it will be clear how little appropriate the Parthenon is to the side on which are
ranged the names of Miltiades, Kimon, Thucydides, and of which the principles were solidly conserv
ative. Their tendency was antagonistic to all bold innovations. Thus, in particular, Kimon's work on
the Akropolis, the south wall, arose from a policy so short-sighted as to be almost timorous. The
valiant, open-hearted hero was certainly, in intellectual endowments, undistinguished, circumscribed,
wanting in even the ordinary Seuorrjs 'Arrticj.6 It is to the other side, the side of Kleisthenes, Themi
stokles, Perikles, which shook off the yoke of tradition and made progress its own, that the Parthenon
likewise belongs. Thus Perikles, in this case as in others, appears as the successor who carried on
the policy of Themistokles and put his ideas into execution, whereas it would be incomprehensible
that Perikles should take up a project of Kimon. The audacious conception of the temple, then,
takes its place appropriately beside the audacious erection of the city wall, and the foundation of the
Peiraieus : all these were originated by Themistokles, and the first is no less significant than the
others of his energetic thoroughness, of his abolition of ancient trammels. He embodied too, and
1 As Curtius assumes, Stadtgesch. 130.
2 Cf. Wolters in Bonner Studien, p. 92 sqq. So previously Robert, ap. von Wilamowitz, Kydathen, p. 182.
In the first years after Salamis there cannot yet have been any talk of treachery committed by Themistokles.
Cf. Bauer, Themistokles.
4 Thuc. i. 91 ; Aristotle, 'Ad. iroA. 23, 4. <s Cf. Busolt, Gr. Gesch. ii. 54.
" Stesimbrotos, ap. Plut. Kim. 4.

PARTHENON OF PERIKLES 423
expressed, the pride of Athens in her independence of Sparta. In earlier times the same tendency
which we now recognize in the building of the Parthenon is to be found expressed in the removal of
the Prytaneion from its old site, and in the founding of the Tholos by Kleisthenes.1
When Themistokles was abandoned by those with whom he had acted in concert during the first
years after Salamis, when Aristeides passed over completely to Kimon, and the coalition of the
Alkmaionidai and Philaidai, symbolized by Kimon's marriage with Isodike, thrust out the hated and
maligned Themistokles, then the Parthenon necessarily fell likewise into the background.

III. The Parthenon of Perikles.
Not till 447 was the work resumed by Perikles, with a somewhat altered ground-plan.2 It was
the time when, after Kimon's death, Perikles stood at the head of Athenian politics, in spite of the
opposition of the Kimonians, led by Thucydides. Hostilities with the Persians were given up ;
Perikles wished for peace abroad, in order to build up at home the power and splendour of
Athens. We have already spoken of the peace negotiations belonging to this period, and have
conjectured (p. 34) that Perikles had the colossal statue of Athena Promachos erected on the Citadel
as a splendid votive gift to mark the conclusion of the Persian wars. The resumption of work on the
Parthenon belongs to the same group of events. When the Persian war was given up, the payments
of the allies, which up to that time had been used for carrying on the war, became available. If
they could be successfully added to other resources, a speedy and splendid completion of the
Parthenon would become possible.
With this matter seems to me to be connected an act of Perikles's policy which has been much
discussed, and is still constantly being shifted from one period of his life to another. I refer to the
proposal of a Peace Congress, mentioned by Plutarch (Perikl. 17). All Hellenes were to be invited to
form a congress at Athens, to deliberate concerning the shrines burned by the barbarians, the
sacrifices for the deliverance of Hellas from the Persian crisis, and concerning peace and free inter
course by sea. It has been rightly surmised that this decree is to be connected with the building of
the Parthenon. But so long as a Kimonian Parthenon was assumed, this gave rise to the greatest
difficulties. The motion could, of course, not have been made while the building of the Parthenon
was in progress, and even not at a time when the work had ceased only a year or two before.
Therefore, Kopp recently proposed to date the motion at circa 459, earlier than Kimon's Parthenon, the
beginning of which he places as late as 454 : 3 he does not hesitate to suppose that in this case
Kimon carried out a project set on foot by Perikles. We escape this difficulty by dating the
beginning of the older Parthenon immediately after 479. At the death of Kimon, in the year 449,
the building had long lain an abandoned design ; and a large part of the marble drums prepared for
its columns had most probably been already built into the north wall. Thus the resumption of the
work needed to be specially urged. That motion of Perikles must belong to a time shortly after 449,
and its immediate practical result was the beginning in 447 of the splendid new Parthenon, with the
help of the payments of the allies.
Perikles, in his projected Peace Congress, did not in the least enter on ' the inheritance
bequeathed by Kimon ' :4 on the contrary, it was to inaugurate the new Periklean policy, which com
prised the conclusion of the Persian war, peace within the limits reached, and free intercourse by sea
under the hegemony of Athens. That the congress would come to nothing, Perikles, of course, fore
saw. His real purpose was to represent the restoration of the shrines as a Pan-Hellenic concern.
Then clearly no one could find fault with Athens if she appropriated for her own sacred buildings part
of the funds of the confederation hitherto devoted to the war with Persia. The Athenians could then
say : 'The religious duty arising on the conclusion of the war with the barbarians, which the others
refuse to perform, we are performing, we Athenians and allies, in memory of the Persian crisis from
which we delivered you.' Thus on the failure of that congress there no doubt followed immediately
1 Cf. Wachsmuth, Stadt Athen, i. 506.^7., ii. 315.
2. There was in the Parthenon a picture dedicated by the sons of Themistokles, and representing Themistokles
himself (Paus. i. 1, 2) : this was no doubt one of the first votive offerings in the new building of Perikles, and the
great man's sons were allowed by means of it to keep alive the recollection of the original founder of the
Parthenon. 3 Kopp in Jahrb. d. Inst. v. 1890, p. 268 sqq. Curtius (Stadtgesch. 139) rightly rejects this date as too early ;
he argues that it must be so, if only because the proposal implies a high degree of self-confidence on the part both
of the Athenians and of Perikles. Curtius's estimate, about 445, is too late, since by that time the building of
the Parthenon was in full swing. 4 Kopp, loc. cit. 271.

424 THE TEMPLES OF ATHENA ON THE AKROPOLIS
the proposal of Perikles to begin the building of the Parthenon with the gold and ivory statue of the
goddess, and to use for this purpose part of the funds of the alliance.
That some of this money was in fact expended on the building is proved by the extant fragments
of the accounts.1 The receipts of the Epistatai of the work regularly include, besides sums from the
treasure of the goddess and from various sources the designations of which are mutilated, an item napa
'EWrjvorapiav. This can refer only to payments of the allies, o-vppaxtKol (popoi, which passed into
the treasury of the Hellenotamiai; and it is not a case of the pva ano tov raXdi/rou, the sixtieth
of the tribute which belonged to the goddess ; money belonging to the confederation is meant.
In the years 447 — 445 there arose a strong opposition led by Thucydides, which passionately
opposed this application of the funds of the alliance, and loudly protested against it in the popular
assembly,2 until Thucydides was removed by ostracism and his Hetairia dissolved. Later, on the
occasion of the building of the Propylaia, Perikles seems to have proceeded with greater caution.
The contribution furnished by the Hellenotamiai, as we learn from the extant accounts belonging to
the year 434 — 433, is designated as pva awb tov rdXdvrov : it is only a sixtieth of the tribute, a second
in addition to the first which from the beginning had passed into the treasury of the goddess.3 As
Perikles succeeded by means of the proposal for a Peace Congress in finding a suitable pretext for
adorning the Citadel of Athens with the help of the resources of the confederation, so the Athenians
somewhat later, probably likewise under the guidance of Perikles, had the skill to procure themselves
by a similar proceeding the means of increasing the splendour of the shrine at Eleusis : they exacted
those contributions — a kind of Peter's Pence — which the Eleusinian decree of taxation required from
Athens and her allies, and requested from the rest of the Hellenes.4 The new Parthenon built by
Perikles differed from the older, as was noted above, mainly in the greater width of the cella, required
by the colossal image, and also in the absence of the two small chambers of the west cella, which
made the whole length less. Even the older Parthenon had in all probability departed from the
model of the ' Old Temple ' by making the east cella considerably the larger, whereas in the ' Old
Temple ' it is somewhat smaller than the west cella including the two chambers. This pre-eminence
of the east cella corresponded to the superior importance of Athena at this time compared with the
old-fashioned forms of ritual represented by Erechtheus. In the building of Perikles it is still more
emphasized, the chambers of the west cella being given up altogether. We must conclude from these
facts that those who planned the older Parthenon still contemplated the transference of the whole
ritual of Erechtheus, which, with all that pertained to it, was established in the west half of the ' Old
Temple.' The opposition, which brought about the abandonment of the projected building as a
whole, probably resisted this part of the scheme with special vehemence, since the worship of
Erechtheus seemed united by peculiarly close bonds with the neighbourhood of the old 'tokens'
and the graves. It was, we must further infer, by a skilful concession to this feeling that Perikles
relinquished the transference of the worship of Erechtheus. Accordingly Perikles left the ' Old
Temple ' standing as it had been provisionally restored after 480. It was to retain at least the worship
of Erechtheus just as before ; only Athena received a splendid new abode. That even after the
completion of the Parthenon Perikles had no intention of erecting a new temple for Erechtheus, and
therefore wished to leave the ' Old Temple ' standing, may clearly be inferred from his then taking
in hand so great a work as the building of the Propylaia.
The west cella of the Parthenon might, therefore, have been omitted altogether. Perikles,
however, only left out the chambers and retained the cella. His main reason doubtless was that he
did not wish to depart too entirely from the lines of the existing foundations. His west cella, then,
was a second chamber for the goddess, to whom the whole building belonged. The place was called
' Parthenon,'6 according to my present view, because in the women's dwelling-place that name was
given to the part which was most carefully shut off, the secret chamber into which the maidens
retired, and in which it was customary to keep precious possessions, remembrances, and relics.6 The
great east cella was called b veas 6 eKarbpweSos, the Cella of a Hundred Feet, as distinguished from
the old expression rb 'EKaropneSov, which included several cellas. This east cella was the hall of
J C. I. A. i. 300 sqq.; Michaelis, Paus. Descr. Arcis, 2nd ed. p. 40.
3 Plut. Perikl. 12 sqq. With regard to his authority, see Busolt, Gr. Gesch. ii. 500, note.
3 This is the interpretation given by Christ, De Publ. Pop. Ath. Rationibus, p. 13. Cf. Busolt, ii. 568.
( It is to be specially noted that the Srjpos t»» 'ABrivalav reserves to itself the absolute right of determining what
avaB-npara are to be offered from the money thus collected. With regard to the general question of the decree, see,
as the latest authorities, Holm, Gr. Gesch. ii. 274; Curtius, Stadtgesch. 159; Rubensohn, Mysterienheiligthiimer,
PP- 36, 5,2, 77- According to Loschcke (Enneakrunosepisode, p. 18 sqq.), it was not Perikles who originated it.
0 It is so called in the official lists drawn up by the treasurers, which lists begin in 434. The name is applied
by transference lo the whole building first in Demosthenes (xxii. 13).
" Thus the lance of Pelops is kept in Iphigeneia's Parthenon (Eurip. Iph. Taur, 826),

THE OPISTHODOMOS 425
state, the place of festivals, in which the goddess herself was manifested in her image. Her ' Par
thenon ' was the west cella, shut off by itself, which in the building of Perikles had been converted
into a second chamber of the dwelling of the maiden goddess. Now the name ' Parthenon' already
includes the meaning of a place where can be stored objects of all kinds that have to be kept shut up.
And for this purpose it was in fact used, though not, however, for storing money.
The common notion is that the treasure of the confederation, or the Empire, which in 454 was
transferred from Delos to Athens, was kept in the temple of Athena on the Akropolis ; ' it was
formerly even thought that the Opisthodomos of the Parthenon must have already stood complete in
454. All this, however, is a mistake : the treasure of the confederation, into which passed the tribute
of the allied states, was administered by the Hellenotamiai. We do not know where their official place
of business was; probably in the lower city.2 That it was not in the 'Opisthodomos' on the
Akropolis is clear from the celebrated inscription (C. I. A. i. 32) which lays down rules for the
administration of the treasures in the ' Opisthodomos ' in the year 435 — 434. It is there assumed
that the treasurers of the goddess are already installed in the Opisthodomos. The new regulations
refer to the appointment by lot of treasurers for the other gods. They are to share in the
Opisthodomos with those of Athena, the one group having the right side, the other the left, as the
scene of their official duties. Moreover, it is decreed that the Hellenotamiai should from time to
time deposit the surplus of their treasure with the Tamiai of Athena, who are to take over the
administration of these sums.3 From this it is clear that the Opisthodomos was not used by the
Hellenotamiai, who administered the funds of the confederation, but only by the Tamiai of Athena
and of the other gods.
But even these Tamiai stored their money in the ' Opisthodomos,' not in the Parthenon. Now
the Opisthodomos has hitherto been usually regarded as a part of the Periklean temple, or as a part
of the 'Old Temple' (Dorpfeld, Curtius, White4) ; but Milchhbfer 5 seems to me to have proved that
it was a separate and independent building on the side of the Citadel farthest from the approach ;
that it was, in fact, a 'back building' (dmo-dodopos) of the Akropolis.0 And with the old theory on
1 Cf. recently Curtius, Stadtgesch. 132.
2 This is the view of Kdhler, Abh. d. Berl. Akad. 1869, p. 103, note. The writers who mention the transference
of the treasure from Delos only say that it was brought to Athens, not that it was taken to the Akropolis.
3 Kdhler (Abh. d. Berl. Akad. 1869, 104) had already formed this conception of the relation between the two
boards. Kirchhoffs view (Abh. Berl. Akad. 1876) is that after the transference of the treasure to Athens the
annual receipts of the confederation, consisting of the tribute paid, were deposited by the Hellenotamiai v, ith
the treasurers of Athena, and that this arrangement was only re-enacted in 435 — 434 ; he considers that the two
boards always performed their functions in the same place. But this theory is untenable. The tribute (including
the 8e/caT7j from the Bosporos ; cf. Beloch, Rhein. Mus. 1884, 34 sqq. ) passed into the treasury of the Hellenotamiai,
and they administered it. The inscription (C. I. A. i. 32) must be understood as enacting that they should
deposit with the treasurers of Athena any surplus that might remain at the end of the year. This was shown
by J. Christ (De Publ. Pop. Athen. Ralionibus, Greifsw. Diss. 1879, p. 14) : he supplies in the inscription
rk eKa\arore irepibvra, which is probably more correct than yev6peva. Christ's interpretation was supported
by Lipsius, Busolt (Gr. Gesch. ii. 422), and Thumser (in Hermann's Griech. Staatsalterth. p. 630). The sums of
money thus deposited were of course, as Kirchhoff has seen, not intended to pass into Athena's treasure. The
administration of the treasures at Delos affords an analogy ; there the rapiai (who would answer to the Helleno
tamiai at Athens) deliver a certain smaller part of the state revenues (cf. Schdffer, De Deli Ins. Rebus, p. 120) to
the Upmroiot at the temple (these answer to the treasurers of the goddess at Athens) ; these sums are kept in a
special Srjpoiria ki&ut6s ; they are a deposit in the hands of the lepoirotoi, who alone administer the treasures in the
temple, as at Athens the rapiai trjs Beov have the management of the Opisthodomos. But did this depositing of
surplus funds on the part of the Hellenotamiai, determined upon in 435 — 434, ever actually take place ? When the
war had once begun it certainly did not, since then all moneys received by the Hellenotamiai were at once paid
away. There have as yet been found no accounts recording payments made by the rapiai ri)s Beov from funds of
the state deposited with them ; and yet this must have been the case, if Kirchhoff's views were correct. The
payments made by the r. r. Beov to the Hellenotamiai, or directly to the generals, from 01. 86, 4 onward, were,
as Kirchhoff has seen, taken from the property of the goddess, and were to be paid back subsequently. Therefore
from as early as 01. 86, 4 there was no deposit of money belonging to the state in the charge of the treasurers
of the gods. — In the fourth century also these latter held in theory the management of the reserve funds of the
state, when at any time there were such funds. But this does not in the least imply that, as Frankel asserted
(Histor. Philol. Aufsdtze, Festschr.f. E. Curtius, 1884, p. 40), the functions of the Hellenotamiai were transferred to
them ; the Hellenotamiai had always managed the current receipts only, never a reserve fund. — It is further to be
noted that Kirchhoff's view concerning the sixtieth of the <p6poi given to the goddess is of course impossible,
according to our theory. He holds that this sixtieth, the airapxtj of the tribute given to Athena, was offered as
a kind of rent for the money of the State deposited in her care. This is in itself improbable, for that votive
gift was no doubt simply intended to express thankfulness to the goddess for taking over the office of protecting
the confederation. 4 Dorpfeld, Ath. Mitth. xii. 210, xv. 437; Curtius, Stadtgesch. 132, 152 ; J. W. White, The Opisthodomos
on the Akropolis (1894). The last treatise is at present only privately printed, and I have to thank the kindness
of the author for my knowledge of it. When it has been made public, it shall be discussed in more detail.
5 Philologus, vol. liii. (1894) p. 352 sqq.
6 The whole, of which a part is designated oiritBoSopos, may be the Akropolis just as well as a temple-
building. The possibility of ' Opisthodomos ' being conceived in the former fashion is, moreover, confirmed by
3 I

426 THE TEMPLES OF ATHENA ON THE AKROPOLIS-
this point is also overthrown Dorpfeld's argument which infers from the mention of the Opistho
domos in inscriptions of the fourth century that the ' Old Temple' still existed at that period.1
In the ' Parthenon ' — the west cella of the Periklean temple — were kept only sacred objects.2 At
the time of the war the valuables of the Pronaos and Parthenon were melted down, and finally (407—406)
reduced to a scanty remnant. This remnant and some objects from the ' Opisthodomos ' were subse
quently brought into the great cella of Athena, in order, we may conjecture, to prevent its appearing
too utterly empty. The ' Parthenon' itself is named in the inscriptions after this only as the place
containing the stele which gave the data for verifying that the image by Pheidias was intact. At the
end of the fourth century the place was so empty that it could be assigned as a dwelling to Demetrios.
When the Periklean Parthenon was built, then, the intention was that the west cella given by the
existing foundations should be a closed space serving as a second chamber for the goddess. The
worship of Erechtheus was to remain where it was, attached to the ' Old Temple,' with the two sacred
' tokens ' beside it. But Athena was to move from the ' house of Erechtheus ' into a new and
splendid abode : even the Parthenon of Themistokles had destined a cella of a hundred feet to
her sole use. This arrangement corresponded to the superior importance which Athena already at
that time enjoyed in comparison with her old companion in the temple, Poseidon-Erechtheus.
Without doubt the Hekatompedos Neos of Perikles, just like that begun in the time of Themistokles,
was intended to supersede completely the cella of Athena in the ' Old Temple.' But it follows from
this that it was intended to bring into the new temple the ancient and highly revered image of Athena,
with which the most important ceremonial rites were inseparably connected.
The erection of the magnificent new statue by Pheidias in no way tells against this conclusion.
It was quite usual to display the old sacred images beside the more splendid new ones. This was
done, as we know from inscriptions, in the Brauronion on the Akropolis ;3 also, as Pausanias
testifies, in the Heraion near Argos, and in other places.4 It is known that the statue by Pheidias
the ancient explanation of the word given by the Scholiasts. They consistently regard bimrBoSopos as an inde
pendent building behind the temple. In the period before Perikles, likewise, the treasury was separate from the
temple, as we remarked supra, p. 425.
1 The other reasons adduced by Dorpfeld for his hypothesis will bear examination even less than this one. It
was attractive, but wholly erroneous, to connect together the divisions in the management of the treasure in the
Opisthodomos as established by the decree of 435 — 434, and the division into chambers in the ' Old Temple,' which
had existed long before that time. The inscription rather points to a place common to the two boards without
any division as far as building was concerned. The inferences reached by Dorpfeld from his theory are wholly
impossible. The old temple of Polias containing the lamp — the only temple named by Strabo besides the
Parthenon as standing in the lepov of Athena on the Citadel — is identified with the Erechtheion, and with no
other place, by the inscription relative to the building, which assigns the old image of Polias to the Erechtheion.
Yet Dorpfeld identifies it. with his ' Old Temple,' while on his theory the east cella of the Erechtheion must
have been absolutely useless and empty.
2 Even the Hekatompedos Neos itself, the cella of Athena, was used in the same way, with this distinction,
however, that, being the principal cella, only those utensils and votive gifts which were more pleasing to the eye
were stored in it, just as in the Pronaos, which was wholly open to view from outside, there were displayed a
multitude of handsome silver vessels. These utensils were undoubtedly brought over from the ' Old Temple ' to
the Parthenon when the latter stood complete. The aKivaicai ireplxpvcroi, for instance, are no doubt to be regarded
as belonging to the spoil of the Persians. The transfer seems not to have been completed till the first year of which
the accounts were recorded ; for the large number of eirereta which came into the Parthenon in the first year
'C' {¦ A' 1- I^1)' f°rmlng a contrast to the very small subsequent additions, are not likely to have been new
acquisitions, but had merely been brought over ; and in fact there are even fragments among them, such as feet of
couches. We must, indeed, suppose that all objects of value were at that time transferred from the ' Old Temple '
to the new one ; otherwise the records which give three headings, Pronaos, Hekatompedos Neos, and Parthenon,
would necessarily have a fourth also, the ' Old Temple. ' But it is not till the end of the fourth century that lists
appear of valuables stored in an apxaios veiss (C. I. A. ii. 733, 735) ; this is to be understood as meaning the
Erechtheion. The extant records surely represent the carrying out of the commission given by the people to the
treasurers of Athena in the decree (C. I. A. i. 32) to make an inventory of all Up* xpvpata composed of the
precious metals. Now if any of these had remained in the 'Old Temple,' they must have been entered in the
lists. The oxhaSlas, the Bwpal, and the *Kivi.K-i\s, which Pausanias saw in Athena's cella in the Erechtheion,
were at that^ time in the Parthenon, where they must be included under the various headings of the inventory
among the atavaicai BdipaKes and buXaSiai. The a/afcSiojs of Mardonios had been, according to the accusation
brought by Demosthenes in Timokr. 129, abstracted by Glauketes ; but it must have been discovered again.
The same is the case with the Siippos _ of Xerxes. That the latter was in the ' Parthenon' is expressly stated
(Harpokr. apyvpoirovs Sitppos ; Michaelis, Der Parth. 291, was wrong in trying to remove the Parthenon by conjec
tural emendation) ; Kdhler recognizes it in an official list dating from soon after 01. 95, 2 (C. I. A. ii. 646) ; it
is there entere < as (Sitppos) apyvpoirovs els. No doubt it was not till late times that one of the oicXaSiai, preserved
down to the days of Pausanias, was regarded as a work of Daidalos.
3 In the Brauronion we know that there were at least two images— that by Praxiteles, and that which the
inscriptions call rb eSos tb apxdiov. Cf. recently Robert, Arch. Mdrchen, 150 sqq.
4 Paus. ii. 17, 5. Here, in the Heraion near Argos, there were three images of Hera ; the oldest coming
from Tiryns, an old one on a pillar, and lastly the magnificent statue by Polykleitos. Other instances are given
by Petersen, Phidias, p. 87 sqq.— There is a special interest in the inscription of the Dionysiastai of the Peiraieus
(Ath. Mitth. ix. p. 290) ; this tells us (line 17) how the priest dedicated /caret pavr eiav, a new cultus-image, to the
god, m addition to the older one (irpotrtSpviraro).

THE EAST FRIEZE 427
was not originally called 'Parthenos' at all, but was just as much 'Polias' as was the old image.
The former was 'the golden image,' 'the great image,' the latter 'the ancient image,' but both
served for the same worship. The statue by Pheidias with Nike in her hand was present to the
mind of Aristophanes when he wrote (Knights, 581): & ttoXlovx* Ha\xds...8evp' dcpiKov \al3ovcra tt)v
...f/perepav £vvepybv Niki)v, and the statue of Athena Polias at Priene was a tolerably close copy of
the Pheidian Parthenos.1 But even later, when it was customary to distinguish ' Polias ' and
' Parthenos ' on the Citadel, the single worship proclaimed that they were but a single goddess ; there
was never a separate worship of the Parthenos.2
But there is a stronger proof, the strongest of all, to show that when the Parthenon was built it
was really intended to transfer the old image, and the worship connected with it, and therefore com
pletely to supersede the ' Old Temple ' as far as Athena was concerned. This proof is given by the
sculptured frieze of the temple.
In the middle of the east side of this frieze is represented the delivery of the Peplos, a fact that
should never have been disputed : the doubts which have been raised, and the new interpretations
which have been attempted,3 serve only to confirm the correctness of the old view. The group forms
the centre of the whole frieze, above the midmost intercolumniation and over the door through which
the glance fell upon the glorious statue within the temple ; it stands where an indication of the end
and aim of the whole procession was necessarily expected ; it is simply inconceivable that in such a
position nothing more should be represented than the quite uninteresting scene of a priest's
robing. It has been objected that no one is paying any attention to what is going on in the centre.
The answer to this is that the scene is clearly conceived as in the interior of the temple. The pro
cession has come up outside ; it cannot see anything of the proceedings within. Then some have
held that the action is not that of handing over. And yet it is represented in the most natural
and lifelike way : we see how the man has lifted the heavy piece of stuff with both hands to
bring it near his face and examine it critically ; the attitude of his right hand, too. is just that of
feeling a piece of stuff with a view to testing it. This great piece of heavy stuff carefully arranged
in numerous folds 4 can be nothing but the great embroidered Peplos, which the Panathenaic
procession has brought up, and which this priestly personage is now receiving.
Now the Peplos was at all times destined for the old image, and for none other. The ancient
ceremonial usages of dressing and undressing, of cleaning and washing image and raiment, were
inseparably associated with the ancient idol, the dpxdlov eSw. The contradiction between this fact
and the frieze of the Parthenon Petersen attempted to remove by supposing the procession to have
brought the Peplos first to the Parthenon and then to the Old Temple.5 In this he rightly recognizes
that the frieze necessarily implies the conception of the bringing of the Peplos to the Parthenon.
The frieze, however, can attest only the intention of the founder of the temple, and not what actually
took place. The delivery of the Peplos is represented on the frieze of the new temple, and is clearly
represented as going on within that temple ; this proves that the founder intended to transfer to the
1 The statue of the Polias of Priene is to be seen on the coins of Imperial times (Brit. Mus. Catal. Ionia, PI.
24, 13). Even the pillar supporting the right hand with the figure of Nike is copied. It confirms the accuracy of
the statuette at Athens.
2 This fact, correct in itself, was used by Botticher, as is well known, as a basis for his perverse hypothesis
that the Parthenon was intended for profane purposes in connexion with the ' Agones.'
3 Friederichs, Brunn, Flasch, Wolters ; cf. Friederichs-Wolters, Gipsabg. pp. 277, 279 seq. Michaelis s
theory (Festschrift fiir Overbeck, p. 178) that the garment worn by the boy belongs to the priest who has given it
into his charge is not tenable ; the cloak indubitably belongs to the boy, for he has distinctly put it on, arranging
it so that it passes from the left shoulder under the right arm, and is drawn up in front with a piece falling over ;
moreover, he is squeezing it against his side with his left elbow. The supposed analogies pointed to by Michaelis
are absolutely different cases, where the boy has merely thrown the garment not belonging to him over his
shoulder. [Mr. G. F. Hill has lately (Class. Rev. 1894, p. 225 seq.) made still another suggestion con
cerning the large piece of cloth held by the priest and the boy— namely, that it is the old Peplos, which,
previous to the arrival of the procession, the priest has taken off the statue, and with the help of the boy is folding
up to lay aside. Mr. Hill proposes to account in this manner for the fact that the gods are taking absolutely
no interest in the proceeding, and that, moreover, the garment is, as he thinks, being 'folded up, not
unfolded.'— E. S.] . ,
4 That it is far too large and too regularly folded to be a cloak for the man has been admitted even by
Wolters, loc. cit.
5 Petersen, Kunst d. Pheidias, pp. 35 seq., 303 seq. The passage of Lactantius adduced by Petersen as an addi
tional testimony to the bringing of the Peplos into theParthenon (Dorpfeld repeats it, Ath. Mitth. xu. 200) of course
proves nothing at all ; it is manifestly no more than a general phrase, with the names of one or two well-known
artists. If we were to understand it literally, we should have to infer also the existence of a gold and ivory
statue by Euphranor lo which a Peplos was offered ; for such a statue we have no other evidence. The size of the
Peplos as represented in the frieze would very well suit an image of a little more than life-size, such as the old idol
probably was, but would be too small for the colossal statue of Pheidias.

428 THE TEMPLES OF ATHENA ON THE AKROPOLIS
new temple the ancient image, to which alone the Peplos pertained. We therefore reach the same
conclusion at which we have already arrived by a different path.
The ancient image did not, however, go to the Parthenon. The plans of Perikles were crossed, and
that as early as the time of the dedication of the new temple, when the ancient idol was to be moved.
It was not moved, and this was evidently a success achieved by the opposition party, which from
the first had combated the whole project of the magnificent new building, and which had kept
the worship of Erechtheus in its old seat, just as it now prevented the transference of Athena's
image. Now, as we know that the Parthenon was still being worked at in 434, and that accordingly
it had not yet been absolutely completed, this victory of the conservative opposition belongs
to that period preceding the war in which the position of Perikles had already been shaken as a
whole, and in which his adversaries ventured to bring accusations against persons connected with
him. The charge of outraging religious obligations was the most convenient weapon in the hand of
his enemies. It enabled them to effect considerable curtailments in the magnificent project for the
Propylaia ; and it enabled them also to prevent the Parthenon from becoming what it was to have
been, and superseding entirely the cella of Athena in the ' Old Temple.'
But before we proceed further we must study somewhat more in detail the centre of the frieze of
the Parthenon. It has been said that so long as no special significance has been shown to attach
to the scene on the left, which represents a priestess receiving seats, it is not necessary that the
scene on the right either, the delivery of the piece of stuff, should have any such special meaning.
And the contention is just. The attempts hitherto made, however, to explain the scene with
the priestess must be regarded as quite unsatisfactory. First we must be sure of what is actually
happening : a woman, who clearly can be no other than Athena's priestess, is in the act of taking
from a girl the cushioned seat which she carries on her head. A second girl is approaching with a
similar seat on her head, and an unrecognizable object on her arm -f she is likewise advancing towards
the priestess, but is half turning back, as though to look at another who is following her. The artist
clearly meant to indicate by this action that more girls carrying seats in the same way are to be
imagined as following after those whom we see ; since if the business of bringing chairs was to have
been represented as ending with the second girl, she would certainly have been made to look towards
the centre. The seats are of the same shape as those on which the gods are sitting : they are 8l<f>poi
— Zeus only has a Bpovos ; a seat-cushion occurs, too, on one of the seats of the gods (that of the
second god from the left, Dionysos).2 The bringing of seats upon the head in the way here repre
sented was customary when guests arrived.3 When Demos is feasted in the Knights of Aristophanes
(1164 sqq.), the bicppos is the first thing brought, then follows the tpdve^a with the meal. The girls in
the frieze are seat-bearers, Sicppocpopoi. Now we know that maidens of this kind took part in
religious processions at Athens, with the naviqcpopoi4 There is no doubt, then, that the two maidens in
the frieze are to be regarded as representatives of the bicppocpdpoi who formed part of the Panathenaic
procession.6 These girls, like the Kanephoroi, no doubt belonged to the most distinguished families
of Athens ; for that they were of the class of the Metoikoi is a modern fable, and that they carried
the seats after the Kanephoroi for their convenience, an ancient one.6 The frieze informs us that the
1 This is most probably a stool, as Petersen conjectures (Phidias, p. 247). Cf. the viroir6Siov in the inventory
of the Parthenon, C. I. A. ii. 646.
2 Cf. infra, p. 431 ; Back, in Fleckeisen's Jahrb. 1887, 434.
3 Cf. the well-known vase by Exekias, Wiener Vorlegebl. 1888, Taf. 6, I, and a vase of the fine style in
the British Museum. The slave carrying a chair in the latter is engraved in the Catalogue of Sculpture, vol. i.
(1892), p. 157. Cf. also Welcker, Alte Denkm. v. Taf. 19, where two chairs are being carried, though not upon
the head; also Millingen, Div. Coll. 7 (= Wiener Vorlegebl. ii. 8), the Welcoming of Jason. Finally a newly
acquired white lekythos from Athens in Berlin, representing a female slave, who at the bidding of her mistress
carries a stool on her head in the fashion of the Parthenon maidens ; her type of face marks her as a ' barbarian ' ;
style of the Pheidian period.
4 That they walked near the Kanephoroi is clear from Aristoph. Birds, 1 549 sqq. , and the passage of Hermippos
quoted in the Scholia there. The Scholiasts wrongly inferred that they followed behind the Kanephoroi ; we know
this (from Aristophanes) only with regard to the bearers of sunshades. That the Diphrophoroi walked in the
Panathenaic procession is nowhere expressly stated, but is certainly to be assumed, since we know that the
Kanephoroi did so.
5 The old interpretation, that they are Errhephoroi, should not have been taken up again by Petersen, when
the seats had once been recognized as such. The Attic, fifth-century, terra-cotta, /. H. S. xi. p. 144, representing a
girl currying a seat with a cushion, just like the maidens of the frieze, is of course a diippocpipos. Hydrophoroi
also occur not unfrequently among terra-cottas. Waldstein misunderstands the figure, and suggests Kosmo and
Trapezo, who have nothing at all to do in this connexion, but to whom Miss J. E. Harrison (Class. Rev. iii.
378), following older precedent (Michaelis, p. 264), had lately called attention.
6 It arose from an erroneous interpretation of the passage of Aristophanes ; the seat-bearers and the bearers
of sunshades have been confused together, and thus the former made to carry their seats after the Kanephoroi.
Cf. Michaelis, Parlh. pp. 213, 255 ; Thumser, Staatsaltcrlh. p. 424. On the other hand, it is stated with regard
to the bearers of sunshades only that they belonged to the class of Metoikoi. They did really walk behind the

BANQUET OF THE GODS 429
Diphrophoroi walked at the head of the procession, entered the temple, and there gave up their
burden before the priestess. The Kanephoroi, who in the frieze form the head of the procession on the
rio-ht l remain outside ; their load is taken from them there by some of the officials. The other classes
of those that take part in the procession are indicated each by only a few representatives, and so
it is likewise with the Kanephoroi and Diphrophoroi. Regarding the matter in this way, we reach a
fresh confirmation of the fact that the delivery of the Peplos is represented in the frieze : since if, as
we have seen to be the case, the seats have been brought by the procession, this will be true also of
the robe represented in the corresponding scene to the right ; therefore it must be the Peplos, the
only robe that can have come with the procession.
But why are the Diphrophoroi received in the temple and by the priestess in person, and how
can this transaction stand on a level with the delivery of the Peplos ? Michaelis, who has rightly
perceived that the Diphrophoroi are represented,2 is not able to solve this question. He speaks of the
seats merely as ' furniture for the festival,' and can make no statement as to their purpose.3 In the
case of the other iropirela— the sacrificial baskets, the paterae, jugs, hydriai, censers, and the like —
the intention is immediately clear : they are used in the act of sacrificing. But what about the
seats ? I do not know that at any ancient sacrifice there were persons participating who sat down.
That the theory which connected the seats with the Kanephoroi rested merely on a foolish blunder
we have remarked above.
A number of seats and armchairs are, as is well known, entered in the inventories of the
Parthenon among the votive offerings and festival furniture. In other temples also we find the
like as votive gifts ; thus at Olympia there is a throne presented by a Tyrrhenian king, at Delphi one
dedicated by Midas and one by Pindar.4 An armchair is named in an inscription as being in the
temple of Artemis at Delos,5 another as being in the Asklepieion at Athens.0 Marble representations
of thrones dedicated to Nemesis, to Themis, and to Dionysos were discovered at Rhamnus.7 That
such thrones were meant for the gods was in later times sometimes made clear by placing their
attributes upon them.8 Again, as early as the Mykenaian period, small representations of armchairs
in terra-cotta were dedicated to divinities.9 The great throne of Apollo at Amyklai is interesting
as having comprised several KadeSpai : clearly these seats were intended for the divine and
heroic guests of Apollo, whom he was to receive on days of festival. An inscription from Chios
dedicates tr)v o-rpa>rr)v ndi rds KaBeSpas Mrjrpl ;10 thus even several KaBebpai are appropriated to the
mother of the gods. The connexion in which these appear here is significant : o-rparr) is
indeed a word that does not occur elsewhere, but it must evidently mean some kind of couch.11
Such pieces of furniture, under the names Koi'-n;, KKivq, KXio-pos, we find not unfrequently kept in the
temples like the seats ; quite a number are entered in the inventories of the Parthenon, others
are mentioned in the Heraion near Argos, in the Heraion at Olympia and the one near Plataia,
and also in the temple of Athena at Tegea.12 There can be no doubt as to their significance : like
the tables, the rpdnefri, they have reference to a custom widely prevalent at festivals of the gods,
that of the o-rpmo-is tr)s K\i.vt)s Ka) KoVp/o-ts rijs rpanetjis. The deity is feasted, and not he alone,
but guests also are invited to his festival from the circle of gods and heroes. This was a widely
Kanephoroi ; no doubt their sunshades were in reality only ceremonial objects (cf. the Skirophoria), but it may
have been supposed even in early times that they were connected with the distinguished maidens, the Kanephoroi,
behind whom they were carried. Von Wilamowitz (Hermes, xxii. 220) is certainly right in denying that the
connexion existed, but it does not necessarily follow that it was not supposed to exist : he too makes the mistake
of saying that the girls carrying seats and sunshades were of the class of Metoikoi.
1 Petersen (Phidias, pp. 291, 315) conjectures, perhaps rightly, that the girls at the head of the procession to
the left (figs. 16, 17) were also marked as Kanephoroi by painted baskets.
2 Parthenon, p. 256 seq.
3 Recently Michaelis (in the Festschrift fiir Overbeck, 1893) has expressed the view that the seats were
intended for the Peplos. I hardly think this will prove convincing.
4 Paus. v. 12, 5 ; Herod, i. 14; Paus. x. 25, 5. That Pindar sat upon the seat was no doubt a fable of
later times, based upon the dedicatory inscription, which gave Pindar's name.
5 Bull, de Corr. Hell. 1882, 128. e C. I. A. ii. 766, 1. 8.
7 C. I. A. ii. 1570, 1571, 1 191 (Ath. Mitth. iv. 284).
8 Cf. the throne of Apollo, Mon. d. Inst. v. 28 ; Annali, 1851, 103 (Brunn); that of Poseidon in the
relief, Friederichs-Wolters, Gibsabg. 1905.
9 Some were found at Mykenai (noted by me, 1878) and at Tiryns (Schliemann, Tiryns, PI. 23 C.) ; a good
specimen is at Berlin (T.C. Inv. 7812 ; cf. Samm. Sabouroff, on PI. 144, 3).
10 B. C. H. iii. p. 324, No. 11, perhaps of the second century B.C.
11 Haussoullier (loc. cit.) conjectures 'vine etoffe, un tapis.'
12 Paus. ii. 17, 3 : kA.ii/tj rijs "Hpas, in the Pronaos, v. 20, 1. Thuc. iii. 68. Paus. viii. 47, 2: kAiktj
lepa T7js 'ABipids. A representation in stone of a couch dedicated to Asklepios at Epidauros : 'Eip-qp.
apx- 1883, p. 27, 3. On the Akropolis is a marble couch with a pillow, dedicated to Dione, and belonging
to the fourth century (the inscription AeXriov apx*ioX. 1890, p. 145, 3).

43° THE TEMPLES OF ATHENA ON THE AKROPOLIS
spread Greek conception, which found its most pregnant expression in the 8eo£evia.x It was for
this that the seats and couches of the temple were used : the numina of the gods were to take their
places upon them. It is a quite erroneous though common view to regard these dedicated seats,
the property of the gods, as intended for the staff of priests.2
The seats accord with the more ancient custom of not reclining at meals : thus they are found
even in the stratum of Mykenaian civilization. They arise from the conception that is seen in the
funeral custom practised in one part of Etruria. There an armchair is placed in the grave, with the
cinerary urn resting upon it, and a table set in front for food.3 In the one case the spirit of the dead
man, in the other that of the god, is to take his place on the seat made ready for him. At least for
the female deities, who never used to recline, the seats must have been always kept.4
We have express testimony that in the worship of Athena on the Akropolis it was usual to feast
the goddess in this fashion at festivals : the priestess of Polias is praised on account of the Koo-prjms
rrjs -rpa7refnf,5 and her two assistants in this duty are called Koo-pm and rpaive^ii).6 Moreover, there is
literary evidence for the custom of preparing at the Panathenaia a small ' Kline ' composed of
flowers ; 7 this no doubt was done in private houses, while in the temple there was held upon a
grander scale a symbolic feast of the goddess with her divine guests. For this were used the couches
and seats belonging to the temple. It is hardly a mere coincidence that from 01. 86, 3 onward (in
that year they had probably been brought from the ' Old Temple') there was kept in the Parthenon a
set of twelve armchairs.8 They were clearly intended for a reunion of twelve gods ; besides these the
temple possessed also Sicppoi and okXuSku, and no less than eighteen Kklvai.
If we now glance at the frieze, the true meaning of the whole central scene becomes at once clear to
us. As in ordinary life on the arrival of guests, so here seats are being brought in, destined in this
case for the divine guests expected at the feast. They are being received by the priestess of Polias,
to whom belonged the management of the symbolic banquet.
And what faith conceived as the invisible sequel the artist has ventured to set forth as actually
taking place. On either side the gods have already arrived, a company of twelve, and they have
taken their places on the seats made ready for them in the temple. Thence they look on at the
festival. Thus for the first time we really understand the presence of the gods in the frieze ; it arises
from no fancy of the artist, however beautiful, from no general and ideal conception of the nearness
of divine beings, but from a quite positive belief and a ceremonial rite actually practised.0
We must not suppose that the choice of the deities either was left to the pleasure of the artist.
This group of twelve Olympian gods was really invited to be the guests at the festival. The
institution probably dates from the regulation of the greater Panathenaia by Peisistratos. The gods
are the dadexa 8eol worshipped in the state religion of Attica,10 the same to whom was consecrated
the altar erected on the market-place by the younger Peisistratos. This gives us the clue for
recovering the names of the individual gods on the frieze. The relief upon an altar from the inner
Kerameikos at Athens makes known to us with complete certainty at least six members of the Attic
1 Cf. Deneken, De Theoxeniis (Berl. Dissert. 188 1). A usage similar to the Theoxenia may well be supposed
at the Panathenaia, even if the same name was not used in this case.
2 The priests also no doubt had their seats, from which they could look on at the festive proceedings : a
fragment of a seat, found near the Erechtheion, bears the inscription lepeas Bobrov (C. I. A. ii. 1656. Dittenberger
is wrong, as Kdhler has pointed out, in assuming that the fragment is out of place and comes from the theatre).
But these seats belonging to the priests were quite distinct from those of the deities, who are in fact also named as
owners in the inscriptions on their own seats.
3 One set of sepulchral furniture of this kind found in a tomb at Chiusi is at Berlin, others at Florence.
One is published Ann. d. Inst. 1878, PI. Q.
4 An interesting Attic vase of the late black-figured style at Naples (Heydemann, 3358 ; Ann. d. Inst. 1865,
PI. F. ; Schreiber, Kulturhist. Bilderatlas, PI. 20, 3 ; Wiener Vorlegebl. Ser. C. 8, 2) represents two female
deities seated on Vifypoi ; in front of them is a small table for food, and a man praying and pouring a libation ;
there is also a small aedicula, in which we must conceive the images of the goddesses to be ; the table and the
seats, however, have been set out for the goddesses themselves, and they have really come and taken their places.
These deities are certainly Demeter and Kore (as has been already conjectured by Stephani, Compte Rendu, 1868,
p. 160) ; the man's name is Mystes (pvara no doubt is vocative). The interpretations given by Liibbert and
Schreiber are wrong. 5 C. I. A. ii. 374.
6 With regard to them see Tdpffer, Attische Geneal. p. 122, as a recent authority.
7 Hesychios, s. v. TtXaitis.
8 C. I. A. i. 161 sqq. After the war only six are left (C. I. A. ii. 646, 16) ; probably the number was then
made up with the plain seats.
9 An excellent corroboration of my view is furnished by an inscription discovered at Magnesia (cf. the
preliminary report in the Arch. Anz. 1894, p. 79). Here at the great festival the twelve gods are invited, their
idols are set up in the Agora, and Lectistemia arranged before them. What happens here is just what we have
postulated at Athens : the twelve gods are invited and entertained.
10 The evidence is given in Curtius-Milchhdfer, Stadtgesch. p. ix.

THE TWELVE GODS 43 1
group of twelve gods ; ] among them is Demeter. The only figure in the frieze which will in every
point do for Demeter is the one with a torch.2 If this is Demeter, then the god opposite her, the
only one of the company who sits on a cushion, must be Dionysos.3 The goddess between Aphrodite
and Apollo is generally supposed to be Peitho ; but she is excluded, if only because she cannot be
reckoned among the twelve great gods of Olympos. Rather the goddess must be Artemis, who is
as appropriate in this place between her brother and Aphrodite, a goddess closely connected with
her,4 as, on the other side, Dionysos is to Demeter. This company of twelve gods (on the left,
Hermes, Dionysos, Demeter, Ares, Hera, Zeus ; on the right, Athena, Hephaistos, Poseidon, Apollo,
Artemis, Aphrodite) is identical with the one so widely prevalent in later times, except that it
includes Dionysos in place of Hestia ; it is possible that this deviation only occurred in the one
case of the Panathenaic festival, the cause being that Hestia was fast rooted to her hearth, and
could not well be invited as a guest, for that later canonical group of gods was that of Athens even
in early times.5
The Peisistratid altar of the Twelve Gods marked Athens, as Curtius says,6 ' even then as an
abode of guest-friendship between state and state, and of peaceful intercourse of nations.' It was
certainly a genuine Periklean conception, according well with the idea of an Hellenic congress
which was to precede the building of the Parthenon, that here on the temple frieze where the festival
of Athena is represented, by far the most prominent place was given — no doubt owing to an established
usage in the rites of the Panathenaia — to this group of gods, which could claim a significance general
to the whole Hellenic nation ; and that the representation of that temple rite which has reference to
the divine guests is placed on a level with the bringing of the Peplos, the gift that did honour to
Athena. Of course, the place in which the gods are sitting is not an ' ideal ' spot ; 7 the temple, the
Parthenon, is intended. The men who stand immediately in front of the gods are conceived as
assembled in front of the temple while the procession comes up. They form a company of ten (four
to the right, six to the left) ; 8 probably they are the ten officials who managed the property of the
goddess, the rapiai lepmv xpri/jaVcoi/ tfjs deoi, who here are posted before the temple intrusted to
their charge. The man in the long chiton within who is receiving the Peplos is, however, certainly
not a treasurer, as has been supposed ; '-> the priestly robe in itself proves this. Now the priest of
Erechtheus will not do either — it is hardly likely that he was concerned with this particular offering,
the Peplos of Athena ; thus we have to regard the priestly personage as a representative of the
lepon-otoi, who, according to a well-known inscription,10 had, in fact, the whole management of at
any rate the Lesser Panathenaia ;u among other things they offered the various sacrifices to Athena,
1 Ath. Mitth. iv. PI. 20, p. 337 sqq. (von Sybel). The altar, judging by its workmanship, is to be ascribed
perhaps to the second half of the fourth century. Cf. also Preller-Robert, Griech. Mythol. i. 1 10.
2 Flasch interprets it as Artemis, but for the reasons against this cf. especially Back, in Fleckeisen's Jahrb.
1887,436^. Michaelis in Berliner Philol. Wochenschr. 1892, 11 72.
3 As was supposed by Michaelis and Petersen, and has lately been maintained by Back with very good
arguments, in opposition to Flasch (Fleckeisen's Jahrb. 1887, 433 sqq. )
4 Gerhard also regarded her as Artemis. — It is well known that Artemis is found not unfrequently in the
fifth century wearing a cap ; cf. merely as an example the Niobid vase of Polygnotan style, Mon. d. Inst. xi. 40, or
the somewhat older cylix of the severe period, Ath. Mitth. v. PI. 10. Robert regarded the latter as representing
Artemis Brauronia (cf. supra, p. 102, note 3) ; in any case the figure follows as its model an Attic cultus-image
of an early period. The vase of Elite drain, ii. 92 is near in time to the frieze of the Parthenon. — The chiton
slipping clown on the left shoulder is found again in the statue of Artemis Brauronia by Praxiteles (cf. supra,
p. 323). Artemis Brauronia as a goddess nearly allied to Aphrodite, and presiding over the sexual functions
of women, has clearly influenced the representation of Artemis in the frieze.
6 I recently had the opportunity in Copenhagen of accurately examining the terra-cotta fragment with the
Athena of the Parthenon frieze, Herr Sophus Miiller having kindly taken the terra-cotta down from the wall into
which it had been fixed. I consider the fragment to be genuinely antique : the thickness, the clay, the whole
appearance are exactly those of the ' Campana ' reliefs. In addition to this the (very slight) traces of blue colour
on the ground, and, above all, the genuine incrustation on the surface, are proofs of their authenticity. The other
fragments noted by Waldstein (Essays on the Art of Pheidias, PI. ix. xiii.) are therefore presumably also genuine.
The agreement with the small casts in Rome, derived from the old mould made for Choiseul-Gouffier, which
puzzled Waldstein, is to be explained by reference to the same original. The Roman terra-cotta copies, which
must have been made in Rome probably in the Augustan age, are evidently derived from casts. [A further frag
ment belonging to the same series, and reproducing figs. 133 and 134 (Mich.) of the north frieze, was recently
acquired in Rome by Mr. A. IT. Smith. Rdm. Mitth. 1894, p. 94 ; cf. Sal. Reinach in Chron. d Orient, 1894,
p. 17.— E. S.] 6 Sitzungsber. d. Berl. Akad. 1890, p. 1 154.
7 As is supposed by Michaelis, Parthenon, p. 221.
8 Michaelis thought that they were rather the nine Archons ; he separated figure 18 from the rest. But this
will not do ; 18 and 19 belong together as much as 20—21, or 22—23 ; theY are three groups of men quite
similar to each other; 18 may also have held a staff. Cf. Petersen, Pheidias, p. 290, note 3.
9 So Bbtticher, and Michaelis agrees with him, pp. 221, 257.
10 C. I. A. ii. 163 ; Michaelis, Parth. p. 332 ; Dittenberger, Syll. 380.
11 According to Aristot. 'AB. iroX. 54, the fifth Panathenaia of every fourth year were not managed by the
annual Hieropoioi ; unfortunately the further details with regard to the Panathenaia are much damaged.

432 THE TEMPLES OF ATHENA ON THE AKROPOLIS
and therefore probably appeared in the long chiton, the sacrificial robe of priests. Analogous to this
function of receiving the Peplos would be the Eleusinian ceremony, at which the leponoioi received
the grain offered by the various states.1 IV. The Erechtheion.
Thus we have in the Parthenon, just as in the Propylaia, a lasting memorial of what Perikles
desired but did not wholly accomplish. The old sacred image of Athena remained in the ' Old
Temple,' and the Parthenon was not, as had been intended, the centre of the worship of Athena on
the Citadel, but only an appendage, however magnificent, of the ' Old Temple.'
Now it was the task incumbent on the opposition so to restore or rebuild the old temple by
the sacred ' tokens ' that it might be a fitting place for the ancient image. But they could not set to
work at once. They had succeeded in cutting down the projected Propylaia of Perikles, but not in
putting an end to the erection altogether, and till this was finished there could be no thought of anew
temple. Then, before the Propylaia were yet completely finished, came the war, and with it the
cessation of all building on the part of the state.
Not till after the peace of Nikias could there be any thought of building new temples,
and it is just this time that Michaelis has recently fixed upon, on strong grounds, for the beginning
of the Erechtheion.2 He at the same time hazards the suggestion that Alkibiades had something
to do with the project. This, I think, is practically impossible. The undertaking does not
in the least accord with the character of Alkibiades, or his attitude in the strife of parties.
But there is another who probably has a right to be named in this connexion — Nikias, the foremost
man, the most highly respected in the Athens of his day, the very person who had been
successful in negotiating the peace. He was the head of the conservative party, and personally a
man of strictly orthodox belief and timid piety : t)v yap n, ndi ciyav Qeiao-pat re koX ra toiovtcp
npoo-Kfipevos (Thuc. vii. 50), in strong contrast to Perikles, who was Seicridaipovias n.a6vweprepos (Plut.
Per. 6). Political and religious considerations alike must have determined Nikias to continue the
traditions of those who opposed the building schemes of Perikles. We saw that the task had fallen
upon them of providing a worthy place for the old image more especially, and also for the worship of
Erechtheus. This was done by the building of the Erechtheion. For it follows necessarily from our
previous arguments that the Erechtheion must be regarded as the rebuilding of the ' Old
Temple.' The latter was now demolished, with the exception of the stylobate,3 which was left as a
level platform. A part of the peristyle had already been taken down and used in the building of the
north wall, perhaps as early as the time when the Parthenon of Themistokles was in progress.4 The
cellas could not be removed till the Erechtheion was erected. The new building was placed
close beside the old, only projecting a little over its stylobate. Thus worship could go on
undisturbed in the old cellas until the new were ready for the transfer. The date of this is not
quite certain. When the work was taken up again in 409, the whole roof of the cella of Athena was
still missing ; it was not put on till 409—408. Possibly the ' Old Temple' was removed at the com
pletion of the Erechtheion, though it seems to me more probable that this had taken place
previously. The expression 6 veios 6 ipwoXei iv a r6 dpxa'iov dydKpa, by which the building in
scription designates the new structure, would be meaningless if— not the new temple, but — the old,
closely adjacent to it, was the temple containing the idol. Further, the inscription proves, as is well
known, that the porch of the Caryatids was the portion of the building most advanced when the works
were interrupted in 413 ; 5 it alone was quite finished and polished except in one or two trifling details,
evidently because it was the chief ornament of the building, and would be the first to strike visitors
to the Akropolis. If, however, the high cella wall of the old temple rose straight in front of the
1 Dittenberger, Syll. 13. 2 Ath. Mitth. xiv. p. 363.
3 Dorpfeld (Ath. Mitth. xv. 439) points out the fact that one stone of the stylobate is still in situ, and several
lie near on the foundations. This is, of course, no proof for Dorpfeld's theory that the temple remained standing
throughout antiquity, but only shows that the stylobate was not demolished with the rest of the edifice : this is
very intelligible, since it afforded a fine level platform. We may further quote the analogy of the old Heraion
near Argos, which was not completely pulled down, the BepiXia being left standing (Paus. ii. 17, 7).
4 Dorpfeld has been so kind as to inform me by letter that the part of the north wall which contains the
entablature of poros stone from the old temple is calculated for a lower level of the surface of the Citadel within
than that which existed in the time of Perikles ; and that when the north wall was built the numerous archaic
statues came to be buried in the earth ; this seems to point to the time of Themistokles as the most probable.
We must suppose that the work of removing and demolishing the peristyle was then begun, and given up at the
same time that the building of the old Parthenon was broken off. 6 Cf. Michaelis, loc. cit.

THE ERECHTHEION 433
Caryatids and completely concealed them, the early termination of this portion becomes unintelligible.
Previous to the roofing of her new shrine, the old image could be provisionally guarded in her own
closed aedicula. Even in the new shrine it seems to have been contained in a separate aedicula, the
painting of whose ceiling is recorded in one of the inscriptions (Michaelis, Ath. Mitth. xiv. 361). It
is, however, quite possible that during the building the image was kept in some other place of safety
within the sanctuary. Cultus-image and temple are by no means inseparable in ancient usage :
there are countless instances of cultus-images that never had a temple. The ancient holy idols
were habitually placed in a small portable shrine or aedicula. It is also several times stated that
cultus-images were kept in the house of the priest for the time (cf. the oltda Upa ov r) Upeia oiVei
at Eleusis, 'Eqbrjp. dpx- 1883, p. 109, n. 74 ; a. 50). J Therefore on the Athenian Citadel also
there would be ample means for guarding the ancient image in a manner conformable to ancient
custom. At any rate, there is nothing that can be urged on this score against our previous
assumption. After the new building had received the dpxdiav aydkpa, it was officially called after this
as being the most important thing it contained ; 2 but in common parlance it kept the name of the
predecessor which it had supplanted, and was called the ' Old Temple.' In the contest over the
building of the Parthenon, long waged between the parties of Athens, the ' Old Temple ' was
certainly always opposed to the new one. The Erechtheion, as the work of that party which had
desired the maintenance of the old temple, as the temple of the old sacred image, near by the old
sacred ' tokens,' remained the ' Old Temple,' even though the building was new ; it represented old
Athens, while on the other hand the Parthenon always represented new Athens.
It is characteristic of Nikias that he — as a private person — dedicated on the Akropolis a gilt
' Palladion ' (Plut. Nik. 3), i.e. clearly an archaistic, idol-like image of Athena wielding the spear.
In this act he was following Kimon, the great hero of his party, who, after the battle of the
Eurymedon, likewise caused a gilt Palladion to be dedicated. Kimon's image was sent to Delphi.3
Thus in their new dedications this party adhered to the old type of the goddess. How much more
must their programme have required that they should do their utmost to display the ancient sacred
image in surroundings worthy of her !
The Erechtheion was smaller in plan than the old temple ; as a grand and spacious temple
they had the Parthenon, and small chambers sufficed for the old image and for the worship of it and
of Erechtheus. They dispensed even with a peristyle, and only set porches in front of the cellas.
The Ionic style of architecture was selected, perhaps in accordance with the precedent of the old
temple, the architectural forms of which are unfortunately not known to us. The choice of site was
determined by the condition of keeping near the sacred ' tokens ' ; in fact, the temple was to be
even more closely attached to these than its predecessor had been ; the cleft in the rock was
included within the building — the Olive of course was to remain in the open. This arrangement, how
ever, involved the necessity of building on a site which included considerable differences of level :
the western part, where was the cleft in the rock, lay lower down than the eastern. It is well known
what skill the architect displayed in solving the difficulty arising from these conditions.
With regard to the interior arrangement of the Erechtheion, there is, as is well known, consider
able uncertainty. This much only is sure, that there were two cross walls dividing it into three
chambers. The old theory of a division into two storeys may be regarded as completely refuted.4 Now
these three chambers clearly answer to the three parts of the ' Old Temple.' In the new building,
as in the old, there is to the east a larger cella, that of Athena. Here too, to westward of this, lie two
other rooms, one beyond the other. We are, accordingly, fully justified in supplying what is
destroyed in the Erechtheion after the pattern of the old temple : that is to say, the central section
may be regarded as having been in the Erechtheion also divided into two chambers,6 accessible
probably only from the third, the western, section.6
Thus we obtain a fixed point of startling importance for the restoration of the Erechtheion
and the comprehension of its arrangements for worship.
The temple as a whole is a unity, and as such is called in the inscription relative to its building
'the temple with the old image,'7 and again, 'the old temple of Polias' ; Pausanias even calls it
1 Cf. Paus. iv. 33, 2 ; vii. 24, 4 ; ix. 40, 12. 2 Cf. supra, p. 432.
3 Plut. Nikias, 13, 3 ; cf. Paus. x. 25, 4. The inscription gave only the name of the Athenians, since it was
a votive offering of the state, but the originator of the idea was of course Kimon.
4 Julius, Das Erechtheion, 1878 (in Baumeister's Denkmdler, i. 488) ; Borrmann, Ath. Mitth. 1881, 372 sqq.
' They must, however, have been divided only by a slight partition wall that did not affect the foundations ;
there is in the foundations no trace of such a wall.
6 That the Erechtheion had no door of communication between the east cella and the west sections is a view
that has been recently supported by Petersen (Ath. Mitth. x. 6 seq.) with good reason.
7 C. I. A. ii. 464 ; Strabo, 9, 396. Cf. supra, p. 432.
3 K

434 THE TEMPLES OF ATHENA ON THE AKROPOLIS
simply ' temple of Athena.' 4 But this unity consisted of two distinctly separated parts. The
eastern part on the higher ground was the cella of Athena, the special temple of Polias, with its
porch turned towards the great altar of Athena. To the larger and lower-lying western part
belonged the north and south porches. The great north porch stands partly in front of the central
section, which shows that this and the western part were closely connected, forming a single whole —
namely, the o'U-qpa 'Epe'x8eiov KaKovpevov, as Pausanias has it. Now this consisted of three apartments —
the west section, and the central section, divided, as we saw, by a wall into two chambers. The west
wall, towards the Pandroseion, had windows above, with half-columns, clearly in order to admit light
to the central section.2 The wall between the western and central sections had two doors opening
into the chambers, probably of the same height as the great north door ; up to this same level it was
perhaps divided into panels by a row of pillars.3 No doubt the two chambers received the more
precious and more sacred objects, and the western section was the antechamber to them. This
clearly was so in the old temple, although there the front apartment is larger. The two chambers
were almost exactly of the same size in the new building as in the old.
Among the sights of the Erechtheion mentioned by Pausanias, the position of one at least is
unquestionable : it is the trident-mark in the rock.4 It lies under the floor of the north porch, and
from this place a covered passage leads straight to the central section, opening into its north-west
angle — that is, into its northern chamber. According to Pausanias, we must conceive the trident-mark
and the Salt Spring as lying quite close together. Again, the Spring is expressly mentioned as evhov,
within the o'Urjpa. From this we must infer that the above-mentioned covered passage joined the two
tokens ' of Poseidon's power, that, therefore, the Salt Spring was in the north chamber of the central
section ; from this, which was, of course, also down in the rock-foundation, the way led through,
below the north wall of the building, to the mark in the rock under the north porch. The placing of
the porch at this point, and the great depth at which it stands, are naturally explained by the desire
to include the 'tokens' in the rock within the limits of the building.
The inscription relative to the building, where it speaks of that part of the interior walls which
had not received its polish, mentions a place which it calls to irpoo-topidiov. This word is evidently
formed from npoo-ropwv, the enclosure of the mouth of a well or spring,6 and therefore denotes the
place, the apartment (o'Urjpa is to be supplied) in which is the irpocrropiov, the puteal ; and this can
scarcely be other than the spring of salt water. The Prostomiaion, then, denotes the north chamber
of the central section.
The inscription unfortunately gives us a definite name for only one other part of the building :
this is rb KeKpbmov. It is true that it has hitherto been customary to place the Kekropion outside, to
the south of the Erechtheion. But since 1886 we have known that the stylobate of the old temple
lay there, and this certainly cannot represent the Kekropion ; and apart from this an accurate study
of the testimony given by the inscription shows that it was a part of the Erechtheion itself. The
context in the inscription (C. I. A. i. 324 a, col. 2, 24) in itself makes this highly probable ; only the
name of part of the actual building could be expected in this passage. But the clearest proof of all
is given by the fact that the ' Caryatid Porch' is consistently called r) irpoo-rao-is 17 npbs ra KexpoirLa,
not rrpbs rov KeKponlov. It is the porch added to the Kekropion, not the one that looks towards the
Kekropion. As the fwa irpbs ra \i9a> ta 'Ekevo-iviw in the same inscription are the figures on the stone,
fastened to the stone, so the 7rpdo-rao-ts irpbs ra KeKpoiricois the porch attached to the Kekropion. The
Kekropion, then, must be behind the porch ; so the name must denote the western section of the
temple. For the ' Caryatid Porch ' does, in fact, precisely stand in front of this section, not over
lapping the central apartment as the north porch does, but ending exactly in the line of the wall
between the sections : thus it is, in a sense quite proper to itself, the porch added to the western
section. The distinction between irpbs with the dative and irpbs with the genitive is very accurately
observed in this inscription. The rolxos irpbs rod Jlavdpoo-eiov is the wall towards the Pandroseion, the
western wall, which faced the Pandroseion situated outside ; in the same way the western pediment
is designated 6 irpbs rov TJavSpoo-elov aleros. If the ' Caryatid Porch' was to have been designated,
1 r$ v*'p rijs 'ABrivils (Paus. i. 27, 2) no doubt refers to the whole Erechtheion, whereas coils rrjs UoXidSos
(27, 1) denotes the special cella of Athena, as is shown by the context.
2 Cf. Julius in Baumeister's Denkm. i. 488.
3 c.f. Julius, loc- ciL Borrmann (Ath. Mitth. 1881, p. 388 seq.) reports that the wall above the level of the doors
s undivided, but below that is carried down with only half its strength ; probably there were projecting pillars
4 Ath. Mitth. 1881, 380 seq. Cf. Julius, Das Erechth. p. 25.
""So irpoirvXatov from irpoirvXov. Michaelis and Petersen can hardly be right in deriving the word from
trr6piov, and explaining it as the space in front of the spring (Petersen), or in front of the ariaia, meaning the
doors (Michaelis) ; Ath. Mitth. >.. 4.

THE ERECHTHEION 435
according to the common view, as looking towards the Kekropion, then, being outside the Erechtheion,
it must have been called irpbs tov KeKpoiriov ; but it is always called irpbs to KeKpoiricp. On the other
hand the corner in which most of the unfinished blocks lay is indicated as r) yavia r) irpbs rov KeKpoiriov ;
it is the corner looking towards the Kekropion. We may further remark that these phrases defining
the 'Caryatid Porch' and the corner by their relation to the Kekropion furnish a valuable con
firmation of the fact that the old temple had at this time (409—408) been already removed, since other
wise they would surely have been determined with reference to that.
Thus we gain the name Kekropion for the western section of the Erechtheion, or at any rate for
its southern part. This, then, is also to tov Kenpoiros iepov, in which, according to an inscription of the
year 334 — 333 found on the Citadel, the Epheboi of the tribe Kekropis put up their honorary decree.1
And here was the domain of the priest of Kekrops, taken from the clan of the Amynandridai.2
Again, the conclusions reached for the Erechtheion may be applied to the old temple ; there,
also, the shrine of Kekrops was in the west cella, the section in front of the two chambers. As
a hero who was half in the form of a serpent, he was naturally worshipped according to the rites that
pertain to those beneath the earth. This might cause it to be widely believed that the actual grave
of the hero existed at this seat of his worship.3
Now let us see what Pausanias tells us. The one eu-oSor to the Erechtheion which he mentions
must, as Petersen has shown,4 be the great north porch with its richly decorated doorway. In front
of this entrance, says Pausanias, stands the altar of Zeus Hypatos. This has been identified by
Petersen, and no doubt rightly, with the altar toO Bvqxov in the north porch, which is mentioned in
the inscription relative to the building.5 On this altar of Zeus were offered only cakes, according to
the most ancient form of ritual, and it was regarded as having been set up by Kekrops (Paus. viii. 2, 3),
which agrees very well with its being near the Kekropion. It no doubt stood at this point, near the
divine ' tokens ' in the rock, even before the erection of the Erechtheion. When that took place, the
altar was, as the inscription shows, completely rebuilt. Its position exactly in front of the entrance
to the Kekropion was no doubt due to the legend of its foundation.0
Pausanias does not mention the Kekropion at all, but this should be no ground for wonder,
considering how much he passes over. He hurries, as is always his way when he enters a country,
a town, or a temple, immediately to the most important sight which it contains ; then, starting from
that and retracing his steps, he notices whatever else there is.7 In the Erechtheion, when speaking of
the altar before the entrance, he immediately tacks on a mention of the three principal altars within —
namely, those of Poseidon- Erechtheus, Boutes, and Hephaistos : thus he gets his favourite antithesis,
so marked a feature of his style. From the mention of Boutes he passes on to speak of the paintings
of the Boutadai. These things cannot have been in the Kekropion ; but, apart from their
want of appropriateness to the Kekropion, the fact that these altars were the principal
centres of worship in the Erechtheion would in itself require us to set them rather in the
place of honour, the central section.8 This was, as we saw, divided into two chambers ; in the
northern one was the Salt Spring. No doubt this chamber also contained the altar of Poseidon-
Erechtheus, the god to whom the 8d\ao-o-a 'Epex6r]is owed its origin. We have the south cella, then,
free for the worship of Boutes and Hephaistos. The latter had his own proper temple in the lower
city ; here he is only a secondary personage associated with Boutes as the original ancestor of the
Boutadai, whose votive pictures0 we must conceive as hung on the walls of this cella. Boutes had
1 AeXriov dpx. 1889, p. 1 1 ; Bull, de Corr. Hell. 1889, 257.
2 C. I. A. iii. 1276. Cf. Tdpffer, Att. Geneal. 160 seq.
3 So Antiochos of Syracuse (ap. Clemens, Protr. 3 ad fin. ; Antioch. Frg. 15, Miiller). His information,
however, concerning points of Athenian religion can hardly have been very exact. If it had been more widely
recognized that the grave of Kekrops was on the Citadel, it is not likely that Clemens would have given the full
quotation from Antiochos. 4 Ath. Mitth. x. p. 7 seq.
5 Loc. cit. The altar found in the excavations to the east of the north porch (Lolling, Hellen. Landeskunde
u. Topogr. p. 351) must be another. The altar of Zeus is said by Pausanias to be immediately before the entrance,
therefore in the porch, where the inscription tells us that the altar of the Thyechoos stood ; and this is in fact called
after the rite which we know to have been followed in the worship of Zeus at this spot.
6 That when restorations took place altars were by no means always erected again upon the same sites has
been shown also by the excavations at Olympia ; the old strata of remains indicating an altar do not always
coincide with the later altars (see, e.g., to the west of the Metroon, Olympia, vol. iv. Die Bronzen, p. 4). The
legends of the foundation of altars refer to the institution of the worship, not to the actual altar-block.
7 Cf. Gurlitt, Pausanias, p. 21 sqq., 75 sqq.
8 This in itself is a sufficient reason for rejecting the ordinary theory, which places those three principal altars
in the antechamber. The worship here was in each case without an image, therefore concentrated itself entirely
upon the altars, which thus were the most important objects in the oturipa.
0 The ypaipai mentioned by Pausanias consisted of the great Pinax containing the genealogy of the Eteoboutadai
which had been dedicated by Habron, son of Lykourgos (Ps. Plut. Lives of the X. Orators, 843), with perhaps
others of the same kind subsequently added,

436 THE TEMPLES OF ATHENA ON THE AKROPOLIS
his own special priests.1 Thus we reach a highly satisfactory result: each priest who had here the principal
seat of his office— namely, the priests of Kekrops, Poseidon-Erechtheus, and Boutes (to the priest of
Hephaistos his functions upon the Akropolis were only secondary)— possessed his own distinct
chamber, and the threefold division of the west cella just answers to these three priesthoods. Of
course this arrangement in the Erechtheion is simply taken over from the old temple ; the chambers
were, as we noted above, of nearly equal size in the two temples. Only the one which pertained to
Poseidon-Erechtheus had the advantage in the new building of possessing in addition the Salt Spring,
the sacred 'token ' of the god, united to his altar. It was in fact simply in order to attain this object
that the position of the temple had been changed at all.
After Pausanias has mentioned the altars in the Erechtheion, and in connexion with them the
paintings of the Boutadai, he continues: ko'i — cWXoui/ ydpiurirb owr/pa — Kai 2 vftmp io-riv evhov 6a~Kdo-aiov ev
cppean. We have now attained an excellent interpretation of this designation of the apartment as twofold
which has given rise to so much controversy. Pausanias refers to the division of the central section
into two chambers : this makes the expression far more intelligible than it is if we follow the hitherto
prevalent view that the antechamber and the principal cella are the two components of the double
whole, though dissimilar to each other,— quite apart from the further objection to the old view
that it severs the connexion between the altar of Poseidon-Erechtheus and the spring, and places
all the principal altars in the antechamber, which we have found to be the Kekropion. But, it may
be asked, why does Pausanias wait to speak of the division of the chambers till after he has named
the altars in both? Clearly this is caused by considerations of style merely, and in fact such
considerations always with this author prevail over those that concern the subject-matter. The
enumeration of the altars within, introduced by the mention of the altar outside, was not to be
interrupted. Thus the notice of the division of the chambers is deferred, and is brought in parenthe
tically when the writer is to mention, as being in one of the chambers, the principal wonder of the
building, which he has already described as a whole with reference to its altars.
We have shown, then, that there were three chambers assigned to the worship of three deities—
Poseidon-Erechtheus, Boutes, and Kekrops . these were all of a Chthonian nature, and on this
account their cellas faced west. This is obvious with regard to Boutes and Kekrops, as heroes ;
and Poseidon was worshipped here only as Erechtheus — that is to say, as a Chthonian balpmv
dwelling in the depths. Therefore mythology, which divided god and hero, told the story that
Erechtheus had been banished to the depths of the earth by means of blows dealt by the trident.3
Again, it must have been at some spot in the Erechtheion, and previously in the old temple,
that the honey-cakes were set each month for the great snake, the guardian of the Citadel.4 And it
is highly probable that this spot was in the cella of Athena itself. The snake always appears in the
most intimate connexion with Athena. Pheidias represented it coiled by the shield of his Polias.
There seems to have been no fixed tradition as to who the snake really was ; Herodotos and Aristo
phanes call it simply the guardian of the Citadel. It was a sort of genius loci, not specially
individualized ; only in later days was it conjectured that the snake was Erichthonios.5 They con
ceived of him as buried in the temple of Polias,0 and surviving in a sense in the form of the snake.
Again, Erichthonios is only a second individuality divided off from that of Erechtheus. There is no
worship of Erichthonios ; he figures in myth, not in ritual. There was certainly no real grave of his
in the temple. But the snake also had no special priest ; it was tended by the priestess of Athena.7
Thus neither the snake nor Erichthonios had any influence on the partition of the Erechtheion.
We have explained the sections of the Erechtheion, and thereby those of the old temple which
it succeeded. With regard to the fortunes of the building, we know from Xenophon that as early as
1 C. I. A. ii. 1656 = iii. 302.
2 The double Kai given by the manuscript tradition has been rightly defended by Petersen (Ath. Mitth. x.
3) ; the passage gives no ground for objection, and it is pure caprice to emend, to suppose lacunae, or the like.
3 Eurip. Ion. 281 seq. irXnya'i rpiaivijs thrust him into a x*»P* X^°vos ; clearly the myth conceived of Erechtheus
as living in the cleft of the rock under the north porch which was ascribed to the action of Poseidon's trident. — In
Hyginus, Fab. 46, Erechtheus is struck by lightning, and this is surely no more than a later transformation of the
blow from the trident, in a form divorced from the local circumstances.
4 Herod, viii. 41 ; Aristoph. Lysistr. 758. The sacred place on the Citadel is named as the scene
without further specification. Hesychios and Eustathios are the first to speak of the temple of Erechtheus or
of Polias. 5 Paus. i. 24, 7 ; Hygin. Astr. 2, 13. — Von Wilamowitz (Kydathen, p. 141) has conjectured that, at the end
of the play by Euripides, Erechtheus was transformed into the oUovpbs itpis. But the line from an unknown
tragedy (Eurip. Frg. 922) to which he refers speaks only of a metamorphosis of half the body into a snake. More
over, we should expect to find a mention somewhere in literature of the identity between the snake of the Akropolis
and Erechtheus, if it had been expressed at the conclusion of the celebrated piece by Euripides.
6 Apollodor. Bibl. 3, 14, 7, 1 ; Clemens Al. Protr. 3 ad fin. But he is not by any means, as Rohde asserts
(Psyche, i. 128), identified with Erechtheus. 7 Cf. Herodot. loc. cit.

KALLIMACHOS 437
406 — 405 it was injured by fire. This happened just at the time that Athens was at the lowest depth of
her difficulties, and it seems accordingly that for the moment nothing was done beyond temporary
repairs. Not till 395 — 394 do we find an inscription (C. I. A. ii. 829) recording a real restoration of the
part destroyed ; and to this time also seems to belong an inscription commending the people of
Karpathos for having presented a cypress to be used for the roof of Polias's temple.1 Other
inscriptions (C. I. A. ii. 733 sqq.) show that in the fourth century votive gifts in the shape of
miscellaneous small valuables had accumulated in the Erechtheion also.
But by far the most valuable object kept in the Erechtheion — namely, the golden lamp of
Kallimachos surmounted by the palm-tree of brass — had been in all probability dedicated at the very
beginning, when the temple was first erected. The palm clearly passed through the roof, serving
as a chimney, and was thus a part of the actual building. Thus it is permissible to conjecture
that Kallimachos, whom we know to have done architectural work, took part also in the building
of the Erechtheion in general, apart from the lamp. The decorative marble work of this temple,
•unsurpassed in carefulness of execution, would accord excellently with the elegantia et subtilitas
artis marmorariae1 for which this artist was renowned, and which earned for him the surname
Katatexitechnos. And it is certainly no mere accident that the Erechtheion is the first monument
at Athens in the ornamentation of which is introduced the so-called akanthos,3 the same motive
which helped Kallimachos to create the Corinthian capital.4 And we may yet further conjecture
that Kallimachos stood in specially close relation to Nikias and his party, with whom originated the
building of the Erechtheion. There is something particularly appropriate in the commission for
the Erechtheion : here no new statue was to be erected ; the builders desired nothing beyond the
old sacred image. These pious souls use their money for making magnificent cultus-appliances. And
the strength of Kallimachos must have lain in decorative art. On that account he ranked, as Pausanias
says, 'after the greatest masters,' while only one statue of a god made by him is recorded.5 There
is a special interest in the form used for conducting the smoke of the lamp. Benndorf already
rightly perceived that it must have some sort of connexion with the palm of the votive offering
dedicated after the victory on the Eurymedon.6 Kimon had then caused to be erected a gilt Palladion
on a palm of bronze, symbolizing the East vanquished by Athena's proteges. Nikias was the heir to
the political ideas of Kimon's party ; the palm-tree in the Erechtheion we consider as being, in
its conception, the work of Nikias, and it took up once more that favourite Kimonian allusion to
the national victory over the East.
But in yet another particular was the tendency of Nikias mirrored in the artist Kallimachos.
The literary notices of him contain what has hitherto seemed an irreconcilable contradic
tion. While the lamp in the Erechtheion and the creation of the Corinthian capital definitely
point to the last third of the fifth century as his date, he is on the other hand named in con
nexion with Kalamis, and a relief which bears the artist's name, and clearly professes to be a copy
1 Bull, de Corr. Hell. xii. 154 sqq. (Foucart). I have already remarked (supra, p. 415, n. 4) how untenable is
the hypothesis of Szanto, who supposes Dorpfeld's temple to be meant here. — It has been proved by Schultz and
Gardner (/. H. S. xii. p. 1 sqq. ) that the north door is a later restoration, which is perhaps connected with
this fire.
2 Vitr. 4, 1, 10. Michaelis emends the marmoreae of the MSS. to marmorariae (Paus. Descr. Arc.
p. 26). 3 Cf. the remarks on this subject in Samm. Sabouroff, vol. i. Skulpt. Introd. p. 8. The Akroterion of the
Parthenon decorated with akanthos must have been added after the completion of the building. Cf. loc. cit. n. 5.
4 I have shown (loc. cit. p. 9) that the Corinthian capital cannot in any case have been invented earlier than
the time of the Peloponnesian War. The recent attempts to make it out earlier are all refuted by the certain
facts with regard to the development of the. finials of stelai. I mistakenly (loc. cit.) followed Benndorf in regarding
Kallimachos as an earlier artist, and therefore necessarily rejected Vitruvius's statement that he invented the
Corinthian capital. This was an error. It is true that the statement is dressed out as an anecdote ; the story of
the basket with the akanthos growing round it as the model of the capital is of course only due to a later explana
tion of its form. But the origin of the Corinthian capital belongs to a time of no great obscurity ; the kernel
of the tradition, the part assigned to Kallimachos as creator of the form, cannot but be accepted as true. Artists
themselves were by that time beginning to write books. The date of Kallimachos, known from the evidence of
the lamp, and the date at which the Corinthian capital must have arisen agree exactly. Thus there is no ground
for doubt. — The temple of Phigalia, too, whether built by Iktinos or no, cannot belong to a time earlier than the
Peloponnesian War. The style of the frieze is closely akin to the Nike balustrade, and essentially diverse from
that of the Parthenon. The antefixes, likewise, decorated with akanthos — there are still fragments remaining on
the spot — point to the time of the Erechtheion at the earliest. Probably the temple belongs to the time of the
peace of Nikias ; its erection at that time would be historically also very intelligible.
6 A seated Hera at Plataia, in the same temple for which the Elder Praxiteles made the principal statue
(see supra, p. 102). The date (circa 425) is again in excellent accord with the other dates known for
Kallimachos. 6 Benndorf, Tiber das Kultusbild der Athena Nike, p. 40. He only drew the mistaken inference that the
lamp also was set up soon after the victory on the Eurymedon,

43§

THE TEMPLES OF ATHENA ON THE AKROPOLIS

of a work of his, shows an archaistic style.1 This becomes intelligible when we have recognized
Kallimachos as the artist of the pious Nikias, who erected a ' Palladion.' 2 There must really have
been archaistic works by Kallimachos. The notices we have of him lead us to conjecture that he had
the skill to combine the grace and delicacy of archaic works with the highest degree of elegance
belonging to the free style of his period ; and, moreover, that he especially distinguished himself in
delicate and exact metal-work : he made the grand golden lamp, no doubt adorned in the richest
fashion, and he received the surname Katatexitechnos.
Now there is a whole set of other works which answer exactly to the conception which
we are thus led to form of Kallimachos, so much so that we surely are justified in connecting
them with him. In connexion with the saltantes Lacae?iae mentioned by Pliny as a renowned
work of Kallimachos, we may recall the dancing-women wearing the kalathiskos,3 which have
come down to us in diverse variants, in reliefs and on gems of later date. They are charac-

Fig. 179. — Kalathiskos dancers on either side of Palladium. Terra-cotta plaque (Berlin).

terized by a certain graceful charm, due to a slightly archaic manner, and combined with
great elegance and a free treatment of the draperies that is clearly influenced by metal-work.
These dancers more than once occur beside a Palladion,4 and this was probably the case in the
1 Cf. Benndorf, 10c. cit.; Ldwy, /. G. B. p. 331 ; Friederichs-Wolters, 435 ; Hauser, Neuatt. Rel. p. 58.
2 Cf. supra, p. 433. May the relief given Gerhard, Gesamm. Abh. PI. 23, 3 (cf. Benndorf, loc. cit. p. 39 ;
Kekule, Balustrade, 1st ed. p. 9), be traceable to a votive offering of Nikias ? It is a Palladion with the snake of the
Citadel, Nike, and a victorious general. The archaic style would accord excellently with such a view, as well
as the subject. It seems to me that the head (Coll. Barracco, PI. 25, 25 a) gives a good notion of what a head
of a ' Palladion ' by Kallimachos would be like ; the features (especially lower part of face) showing conscious
archaism adapted to a type derived from the Pheidian ideal of Athena.
3 Cf. the instances collected by Stephani, Compte Rendu, 1865, 27 sqq., 60 sqq.; Hauser, Neuatt. Rel. p. <)6seq.,
100. In spite of diverse variants one can see that the later monuments must be based upon a celebrated
original. We must distinguish from them some works of an earlier date, such as the gold plaques from the
Crimea, some terra-cottas from Greece (eg. Dumont-Chaplain, Ciram. Gr. ii. 10, 1), and the reliefs from
Gjdlbaschi : these are independent. On the other hand, two slabs of Pentelic marble, each with a relief of a
dancing-girl, recently brought from Italy to the Berlin Museum (Arch. Anz. 1893, p. 76 .My.) are also to be referred
to that celebrated original. They are good specimens of the so-called later Attic school, by no means genuine
archaic works (as they are called by Kekule, loc. cit.) ; the drapery is carelessly treated, the heads were veiy
beautiful. A paste in Berlin (Tdlken, iii. 1480), perhaps of the time of Augustus, also belongs to the very
best class of replicas. Among these rank also the excellent reliefs of a cup from Arezzo, on which the
Palladion again stands in the centre upon a pillar (Notizie degli Scavi, 1884, PI. 7). It would certainly be
quite intelligible if dancing-girls of this type, with the short tunics, were designated in the later jargon of art
as ' Lakonian women.'
4 Especially excellent is the Roman terra-cotta relief in Berlin (Fig. 179).

KALLIMACHOS 439
original. A Palladion surrounded by these dainty worshippers would accord excellently with the
accounts given of Kallimachos. The type of head seen in the Palladion of the relief in Berlin
(Fig. 179) is exactly what one would expect in the period immediately after Pheidias. In general the
Palladion may be conceived as approximately resembling the well-known ' Dresden Pallas ' ; x the
combats of giants on her robe may well be intended to remind us of the ornamentation on the
Peplos of the goddess who dwelt in the Erechtheion, and the motives of the groups and type of the
giants point precisely to that period to which Kallimachos belongs. But above all we must speak
here of the statue of Artemis at Munich,2 that remarkable archaistic work whose exquisite draperies,
at once floating and transparent, declare it to be a product of that period to which belong the Nike
of Paionios and the reliefs of the Nike Balustrade (Fig. 180). Moreover, we can now understand
those magnificent, dancing Bacchantes, so often reproduced in later times,3 who combine certain
features of an archaistic style with the highest degree of freedom and movement in the rendering of
folds ; all this corresponds completely with the above-mentioned Artemis. The drapery clings so
closely to the body that this shows almost as plainly as if it were naked, and at the sides the garments
spread in those peculiar, rushing lines which seem to betray the influence of work in metal. The
Nike balustrade enables us to fix historically this style of drapery. Now, it is quite natural that we
should find in these Bacchantes a strong resemblance to the Aphrodite of Alkamenes ;* this would
be a work of the same period, and probably only slightly earlier than that of Kallimachos.
In the so-called New Attic period it is clear that Kallimachos was imitated to a great extent ;
many of the charming motives of the later Attic reliefs, with their elegance and their slightly
archaistic look, may be ultimately due to his decorative genius ; this may especially be the case
with those splendid decorations on candelabra and pedestals, in which the akanthos is combined
with figures displaying archaistic motives. The fact remarked by Hauser,5 that these decora
tions show many signs of a tradition derived from the older style of metal-work, would again
agree excellently with Kallimachos the Toreutes. On account of the relief which bears the name
of Kallimachos we may also ascribe to this artist works of the archaistic style properly so called,
i.e. such as are wholly based on the archaic style and seek to bring it to the highest pitch of
\eirr6n\s and x°Pls- 6 I am inclined to believe that there is still preserved to us in the so-called
Altar of the Four Gods on the Akropolis 7 an original of the kind by the artist in question : for fresh
ness and delicacy its reliefs are a whole heaven above the ordinary products of the archaistic style.
Moreover, the Lesbian Kyma which runs round below the figures is executed quite in the taste and
manner of the Erechtheion. The work was the pedestal of a statue, perhaps of a ' Palladion,' s
certainly of a figure in archaistic style. The four gods on the pedestal clearly have reference to the
ancient forms of worship which had found a new and beautified home in the Erechtheion : Hephaistos
and Athena, the parents of Erichthonios, stand facing each other, and behind Athena comes a god to
whom we may give the name Poseidon-Erechtheus ; 9 lastly there comes Hermes, who was also
1 Which seems to me a copy of an original in bronze.
2 Friederichs-Wolters, Gipsabg. 450. Cf. supra, p. 52, n. 4. The terra-cotta replica of this figure mentioned
by Rossbach in Ath. Mitth. 1890, 69, cannot but be modern, as it seems to me after an actual examination of
the work. The figure is heavy, and covered with a poor varnish of black tint that is not ancient. It is an exact
copy of the statue at Munich : the difference in the crown of which Rossbach speaks has no real existence. The
figure was sold in the Castellani auction in 1884, and taken to Lyons.
3 Winter has discussed them in 50th Berliner Winckelmannsprogramm, p. 97 sqq.
* The affinity to the type of Venus Genetrix, as it is called, has been perceived by Winter, loc. cit. I 1 7.
But his mistaken view as to the date of the relief obliged him to draw an equally mistaken inference as to the
period to which the Aphrodite belongs. Cf. supra, p. 19, 82.
3 Hauser, Neuatt. Rel. p. 121 sqq.
6 Hauser (loc. cit. 159 sqq.) thinks that the introduction of the archaic style properly so called can be
exactly dated by means of the figure of Athena on the Panathenaic amphoras. Perhaps about 350 the pretty
archaic treatment was applied even to the established type of the prize amphora, almost equivalent to a coat of arms
of the state. But this is a reason for inferring that the style had been established for some time before circa 350,
rather than that it was only then introduced. We may note here that Hauser is certainly right in dating the two
Amphoras from Taucheira (Mon. d. Inst. x. 48, c, d, and h, 13, 14) before 367. He might have gone further :
those two vases belong, as is proved especially by the style of the reverse, to the second half of the fifth
century, and serve to bridge over the gap in the series of Panathenaic amphoras according to the dates hitherto
accepted. 7 Friederichs-Wolters, Gipsabg. 421 ; Hauser, loc. cit. 34.
8 According to Michaelis (Ath. Mitth. i. 298) the side on which Hephaistos is was the one that faced the
way the statue looked : this does not at all exclude the conjecture given above.
9 Hauser (loc. cit. ) regards him as Zeus, who certainly was also represented under this type. But the type is
suitable for Poseidon too, and the little that remains of the end of the staff, so far as I can see m the cast, agrees
at any rate better with a trident than with a sceptre. The shape of the trident was like that on the Capitoline
puteal (see injra, p. 441). Cf. Michaelis, Ath. Mitth. i. 298. Hauser's suggestion of the birth of Athena is not
happy. Poseidon moreover has been already rightly recognized by Botticher, Erkl. verzeichus der Gipsabg. 1278.

Fig. 180.— Archaistic Artemis from Gabii (Munich).

KALLIMACHOS AND ARCHAICISM 441
worshipped in Athena's cella. The artist has brought together the four great divinities, whose rites
were performed in the Erechtheion, excluding the heroes. And the person who erected this altar
certainly belonged to that piously conservative party to whom we owe the building of the Erechtheion
itself. The marked agreement in style and the peculiar delicate workmanship indicates that the well-
known Corinthian puteal J is also to be ascribed to the same artist, or at least to some person closely
connected with him. The place at which the relief was found creates no difficulty : that Kallimachos
did work at Corinth is attested by the very name of the Corinthian capital, which could only have
been given to it if, as the well-known anecdote relates, Kallimachos used his creation first in Corinth.
The two principal deities on the puteal in question are Apollo and Athena ; here again no action at
all is represented,2 but only an assemblage of deities that was worshipped in a particular shrine.
There is also a group of three Charites or Horai,3 and in the figure of one of these the artist has been
led away by the charm of a graceful motive to depart somewhat from the part he is playing.
Another well-head, in the Capitoline Museum,4 is only a poor and tasteless piece of copyist's work,
admirably calculated to show us the difference that exists between original and copy even in things
archaistic ; but we can see that it is traceable to an original of the same kind, and very likely also
by the hand of the sanie artist as the Pedestal of the Four Gods and the Corinthian puteal. This is
shown by the marked similarity in style and in individual motives.5 The principal figures are of
course here also the two that face each other, and those who immediately follow them. We have here
on the left Zeus, Hera, and Athena, on the right opposite them Hephaistos, Poseidon, and Hermes.
Many mistaken guesses have been made as to the reason of this assemblage. Our previous discussion
brings us easily to the right explanation : these also are the deities of the Erechtheion. Athena, as
is right here in the festal procession, gives precedence to Zeus Hypatos with his spouse ; opposite them,
just as on the Pedestal of the Four Gods, come again Hephaistos, Poseidon-Erechtheus, and Hermes.
Behind these leading gods there follow yet other Olympian deities, selected with greater freedom.0
The spring (<i>peap) in the Erechtheion no doubt received when the temple was built a new ornamental
enclosure which was probably designed by Kallimachos. Should we not regard the puteal in the
Capitol as a copy of it ? The sacredness and renown of this spring would make it very intelligible
that its decorations should have been copied in Roman times.
On account of the kindred style I should be inclined further to conjecture that the so-called
'Tripod-Basis' at Dresden7 is a copy after Kallimachos, and probably after an original in bronze.
Here the archaistic reliefs are again combined with an ornamentation which, in its treatment of the
so-called akanthos and the flowers, stands in close relation to the Erechtheion, while the winged
Silenoi introduce an element that belongs to older times. The purpose of the work— it is a stand
for some monumental candelabrum or lamp-carrier, and was no doubt destined originally for a
shrine, probably of Apollo— is also suggestive of Kallimachos, who made the lamp for the Erechtheion.
Finally, I incline to regard the armchair found in front of the Pronaos of the Parthenon 8 as an
original, hardly indeed from the hand, but at least from the studio of Kallimachos. It shows once
more that remarkable combination of dainty archaism and ' swallow-tail ' folds with an ornamentation
resembling that of the Erechtheion.0 How highly this work was esteemed is to be inferred from the
1 Cf. recently Hauser, loc. cit. 162 seq. He indicates the archaic style, and is inclined to date the monument
shortly after 350. Wings on the foot occur in archaic monuments not only on shoes but also attached to the
barefoot (cf. e.g. the old Ionic monuments, Mon. d. Inst. vi. 46, 3 ; Ohnefalsch-Richter, Kypros, PI. 31, 6), so
that this detail in the Hermes is not to be regarded as an argument of late date. Hauser (p. 163, n. 3)
rightly calls attention to the carved Astragal ; this again answers completely, as it seems, to the style of
the Erechtheion.
' 2 The usual interpretation, which regards it as the introduction of Herakles to Olympos, is untenable (I also
followed it formerly, in Roscher's Lex. i. col. 2239) : Zeus could not possibly be absent.
3 The triple group is indicated by holding hands, in accordance with the established type. The interpreta
tion of one of the figures as Hebe was a mistake.
4 Friederichs-Wolters, Gipsabg. 424 ; Hauser, loc. cit. p. 60 seq.
5 Athena, Hephaistos with the hammer, Apollo, Herakles, should especially be compared.
6 Here, as on the Corinthian puteal, Herakles follows after his protectress Athena, which confirms the Attic
origin. It was a matter of boast that Herakles was worshipped as an Olympian god first in Athens, and archaic
Attic vases show Herakles already present on Olympos at the birth of Athena. That the introduction of
Herakles to Olympos cannot be meant has been rightly remarked by Hauser.
7 Friederichs-Wolters, Gipsabg. 423 ; Hauser, loc. cit. 52, 117 seq. ; Reisch, IVeihgeschenke, p. 99. The view
which I formerly advanced (Roscher's Lex. i. col. 2232), that the work was an original of the fourth century,
I cannot maintain after a second examination of the work itself. The workmanship is later, but it is certainly
a mere copy from an earlier original. That it is a pedestal not for a choregic tripod but for a candelabrum has been
shown by Hauser and Reisch. 8 Friederichs-Wolters, Gipsabg. 1332.
9 The stylistic treatment of the akanthos is characteristic, as is the retention of the conical arrangement of
the palmette-leaves, and the form of the spiral, which is not yet, as it becomes later, purely an adaptation of
3 L

442 THE TEMPLES OF ATHENA ON THE AKROPOLIS
fact that there was a copy of it at Rome.1 On the upper edge of the original is an inscription, which
if it were complete would furnish us with an exact date, since it names the Archon Eponymos. I
conjecture that it should be restored thus : eirl Arjporrr]pdro apxovro[s, and that the work therefore
belongs to 393 -392.2 There is of course not the slightest difficulty in supposing Kallimachos to
have been still at work at the beginning of the fourth century.
The archaistic fashion of the time of N ikias no doubt found other adherents also among the artists.
Even Pheidias's pupil Alkamenes must have followed it in one case : his triple Hekate on the Nike
bastion was, as is shown by the numerous imitations, an archaistic work.3 There was an old and
simple type of idol, popular at Athens, in which Hekate was represented as a pillar with three heads.4
Developing this, Alkamenes 5 created the three figures with their backs resting against the pillar. He
gave them the Doric peplos girt over the diplois, as Pheidias was fond of representing it, with an
Ionic under-garment. The attitude was the stiff pose of an old idol, the folds were archaistic, without
however degenerating into, the exaggerated prettiness of Kallimachos. The heads seem to have
displayed the free and grand type of the epoch. The work was certainly contemporary with the
building of the new temple on the Nike bastion, when it would seem desirable to replace the old
idol which had probably existed down to that time.

V. The Temple of Athena Nike.
Thus we have passed to a new subject, the temple of Athena Nike, the fifth extant temple to
Athena on the Citadel, and the last to be studied here. It fits in excellently with the results we
have so far reached.
Recent investigations into the date of the temple of Nike have at least established with certainty
the terminus post queni. Julius and Bohn did indeed consider that they had fixed also the
terminus ante quern ; they tried to prove that the temple of Nike was planned and erected
during the time of the building of the Propylaia. Dorpfeld however has removed the basis of their
assertion by showing 6 that the original plan of the south-west wing of the Propylaia was curtailed
before the building of it had begun at all, that therefore this curtailment could not have been due to
the temple of Nike, even if, as is generally supposed, that was planned while the building was in
progress. Some sacred place existing on the bastion, before the Propylaia were begun at all, must
have been the cause. Next Wolters proved 7 that this sacred place was certainly not the now extant
temple of Nike, and that even the curtailed design of Mnesikles, and indeed the whole Propylaia as
actually built, including even its very latest parts, showed no trace of the existence of the present
temple of Nike. The bastion was higher when the Propylaia were built. Mnesikles did not carry
the steps completely round the anta of the south-west wing, because at that time the third step of the
Propylaia lay still below the top of the bastion ; to divide off that space he erected here in front of the
anta the marble barrier which Wolters proves to have existed. This remained standing even later,
vegetable forms. The monument of those that fell in the Corinthian War also belongs to this earlier stage. Very
soon after this, however, must have occurred the change of type to that of the usual fourth-century stele. Cf.
Samm. Sabouroff, i. Skulpt. Introd. p. 8.
1 Now in Berlin, Skulpt. No. 105 1 ; a good copy of the so-called later Attic school. That it is a question of
a copy only is very clear from closer comparison, especially of the akanthos.
2 The inscription is in C. I. A. ii. 1524. Kdhler proposes Avaiarparos (369 — 368), or KaA.Xfo-TpaToj(355— 354).
But the ornamentation in itself shows the chair to be of an earlier date. Then we come to the two archons named
Demostratos, 01. 96, 4 and 97, 3. To distinguish the second from the first, the name of his deme used to be added
(Kdhler, on C. I. A. ii. 660) ; thus we must suppose the earlier one to be meant here.
3 Cf. my statement, Ath. Mitth. iii. p. 194. But I did not at that time perceive the archaistic character of
the work, and thought it was necessarily to be inferred that ' an Alkamenes ' worked in an old-fashioned style (I
should have said an archaistic). And in fact there was only one Alkamenes (cf. supra, p. 90). That the numerous
Attic repetitions of one and the same type reproduce no other than the work of Alkamenes, the Hekate of the
ir6xis, there can be no doubt. The quite solitary and unimportant Hekataion from Rome which Petersen has
lately put forward as that of Alkamenes (Rdm. Mitth. 1889, p. 73 seq.) cannot stand against the weight of evidence
afforded by the above-mentioned Attic works (some of which are completely established as the Hekate of the
Akropolis by the addition of the Charites). The high girdles and broad hips to which Petersen takes exception
(Oesterreich. Mitth. iv. 140.^.) are details due to the rather late date at which all those reproductions were made.
These are by no means exact copies, as is in fact shown by small variations in many particulars, but only repeti
tions of the type of the statue on the bastion. Similarly even the Parthenos of Pheidias was carved in free Attic
reproductions with a higher girdle. Cf. also the Artemis in ihe Jahrb. d. Inst. vii. (Arch. Anz.) p. 107.
4 This older type, too, was frequently reproduced in later times (cf. Petersen, loc. cit.)
6 Pausanias is clearly correct in his statement that Alkamenes was the first to create the triple type ; this is
corroborated by the monuments which show us no earlier form of the type than that which is to be assigned to
Alkamenes. 6 Ath. Mitth. A. 1885, p. 47. t B-onner Studien, p. 92 sqq.

TEMPLE OF -ATHENA NIKE 443
though it might have been discarded after the lowering of the bastion at the time when the temple of
Nike was built. It is quite clear, then, that it was not till after the completion of the Propylaia that
the temple of Nike was planned and built : the Propylaia in their peculiar curtailed condition were
already quite finished, and it left them absolutely untouched, merely offending against the earlier building
in one point — namely, that by the lowering of the bastion a foundation step of the south-west wing was
laid bare. Now the holy place which stood on the bastion at the time that the Propylaia were built,
and which caused the curtailment of the south-west wing, was that of the great altar of Athena Nike,
which kept its position in later times also ; whether there was already a temple behind it we do not
know, but it is very improbable.1
The temple of Nike, then, is later than the Propylaia. We have now to ask how much later.
The erection of the south-west wing took place, as we have noted, before the temple of Nike was begun,
and this part of the Propylaia was probably among the latest to be built ; for it was here that the curtail
ment of the original design was most extensive, resulting in a finish towards the west that was positively
ugly. This probably happened in the time when the position of Perikles was already shaken, shortly
before the war, rather than at the beginning of the time of building. Thus there remains no interval
at all before the outbreak of the war within which to date the temple of Nike. This is confirmed by
the consideration that the Propylaia themselves were not completed : the blocks were left rough with
the bosses upon them, and the intended adornments were not all executed, clearly on account of the
intervention of the war. This was the case with the Akroteria, and perhaps sculptures in the pedi
ments were also intended.2 How then could the temple of Nike have been built at that time
perfected as it is to the most delicate finish, and richly decorated with sculptures ? And what pretext
could there have been just at that time of anxious suspense for setting up a temple to the goddess of
victory, with reliefs commemorating the Persian War ? Then, finally, we find that the temple of Nike
is nowhere cited among the buildings of the time when Perikles was at the head of the state,3 and,
above all, that it forms a contrast which is out of harmony with the Propylaia, the creation of Perikles.
Already Kekule' was conscious 4 how discordant the small and obliquely placed toy temple is with the
imposing erection of Mnesikles ; and Wolters 5 also holds that the remodelling of the shrine of Nike
can hardly have had the approval of Mnesikles. In fact, there can be no doubt that the erection of the
little temple proceeded from the same party, antagonists of Perikles, which had previously ruined the
magnificent but reckless project of his architect for the building of the Propylaia. But this party
must first have been able to point to external successes of the state, to victories, apart from which the
temple of the goddess of victory would be unintelligible.
The time previous to the outbreak of the war is excluded, then, for these reasons, as it already was
by the considerations previously adduced touching the relation of the temple to the Propylaia, which
were themselves not yet completely finished before the war. Equally excluded are the first years
of the war, when Perikles still stood at the head of affairs, and the time during which the plague was
raging is wholly impossible. Even the period immediately after this affords no event which could
account for the erection. The punishment of the revolted Mytilenaians is of course out of the
question ; moreover, Kleon's influence was at that time quite predominant. The earliest occasion
for the erection is afforded by the prosperous campaigns of Nikias and Demosthenes, in the year
426—425, and especially by the great success of Demosthenes in the Amphilochian War. Here for the
1 [Subsequent observations by Dorpfeld and Wolters (unpublished) somewhat modify the above statements : the
result, however, remains the same, as I have lately convinced.'myself at Athens. — A. F. ] With regard to the altar, see
Bohn, apud Kekule, Balustrade, p. 30, PI. 8. The worship of Athena Nike is of course ancient, and therewith also
her property, her ' treasure.' The existence of a treasure does not, however, in the least imply that of a temple, since
the property of the goddess could be kept in any safe place within the Citadel. The building of a temple was sure
to make considerable inroads upon the treasure, but could not call it into existence. In C. I. A. i. 32 (OI. 86, 2)
the general expression xptpata rr/s 'ABifvaias must be regarded as including the two subdivisions — (a) the money
of Polias, (b) that of Nike ; these two together make up the amount which it was the duty of the rapiai rijs
Beov to administer. In this record there was no reason at all for indicating the subdivision in express terms. It
is different of course in the lists of sums given out and calculations of interest belonging to the time of the war ¦
here the amounts belonging to the two parts are entered separately. From C. I. A. i. 273, iii. c. and iv. a, it is
clear that as early as the second year after that inscription of OI. 86, 2 money was lent from the treasure of Nike
also. This had been kept 'as a part of the Xvhpara rijs 'AB-qvaias in the Opisthodomos of the Parthenon.— I cannot
therefore but consider as unsuccessful the attempt of Loschcke (apud Kekule, Balustrade, p. 28 ; cf. Petersen, in
Zeitschr. f. d. Oest. Gymn. 1881, p. 280) to infer from the absence of any mention of the money of Nike as a
subdivision in C. I. A. i. 32 that the worship of Nike was not instituted till after 434.
2 There are on the pediments the bases for the Akroteria, which never were executed ; cf. Bohn, Die Propylaen,
p. 20. That the pediments were not intended to remain empty is in my eyes highly probable. In the Parthenon
also the groups of the pediment were the last part of the work ; in the Propylaia this point was never reached.
3 This has been already brought out by Ross, Der Niketempel, p. 10.
4 Balustrade, p. 25. Petersen was wrong in combating this, Zeitschr. fiir Oesterreich. Gymn. 1881, 280.
5 loc. cit.

444 THE TEMPLES OF ATHENA ON THE AKROPOLIS
first time a brilliant victory over the enemy had been achieved. And the men of these successes
belonged to the conservative party,1 which, as we saw, must have brought about the building of the
temple to Nike. Demosthenes came home in triumph with three hundred panoplies, which he dedi
cated in the temples of Athens.2 His friend Nikias, who had met with success in Boeotia, devoted
himself with his party in Athens above all to religious matters. There ensued the superstitious ' puri
fication' of Delos, and at the same time the magnificent renewal of the festival on the island, and also
apparently the restoration of a temple there.3 We see that the party was strong and could bring
plans to execution : must it not have tried to set up a monument in Athens also ?
This earliest possible date — the time after the Amphilochian victory, 425 — is actually pointed
to by the architectural forms and the decoration, which still completely accord with what we find in
the Propylaia, and are somewhat older than those of the Erechtheion. The ornaments, which
by the time of the Erechtheion are plastic, are on the Nike temple only painted, according to
the older method. The akanthos and flower-sprays of the Erechtheion are still unknown. The
ornamentation comprises only the old motives of the palmette and lotos-flower. The capital is on
the model of the Ionic capital of the Propylaia.4 Yet one detail clearly shows that it is the later, and
in itself defeats any attempt to make the temple older than the Propylaia. The palmettes which
terminate the volutes already project somewhat over the egg-and-tongue moulding of the kymation,
and the leaves of these palmettes already depart essentially from the strictly conventional representa
tion on the Propylaia, displaying a more naturalistic treatment approaching that of the palmettes on
the Erechtheion.5 Thus architecturally the temple of Nike comes between those two buildings. B.C. 425
is accordingly a suitable date ; then, after the Peace of Nikias (421), followed the Erechtheion.
Now that we have reached this point in our investigations we can derive considerable aid from
an inscribed fragment of a stele from the Akropolis ° belonging to the time about 350 — 320 : it
informs us of a commission named for the purpose of managing the restoration of the statue
of Athena Nike, which was set up by the Athenians from the victory over the Ambrakiots, the
array at Olpai, and the Kerkyraian oligarchs, i.e. from the victory of Demosthenes in the
Amphilochian War. This exact designation of the statue is clearly taken from the dedicatory
inscription itself.7 On account of the restoration of the image a sin-offering to the goddess is decreed,
to be performed by the priestess of Athena herself. U. Kohler, to whom we owe the explanation of
this important record, supposes the statue in question to have been some bronze figure on the Citadel.
But it is difficult to see how such an image, after standing for a comparatively short time, could have
required a comprehensive restoration, since bronzes do not suffer from standing in the open air, and
no event destructive of monuments had taken place in that period on the Akropolis. But the matter
would become completely intelligible if the image was of wood, perhaps partially covered with gold
and ivory. Moreover, the whole elaborate preparation — the specially nominated commission, the
solemn sin-offering performed by Athena's priestess — points to the conclusion that we are here
concerned with a temple image. Now the date and occasion we arrived at for the temple of Nike
are the same which the inscription mentions for a statue of Nike. This gives excellent support and
corroboration. The statue is no other than the temple image for which the little building was set up.
The periegete Heliodoros 8 spoke of it as a ^oavov ; therefore it was a wooden image, but consisting
undoubtedly, as was customary at this period, of more precious materials also, such as gold or ivory, or
even marble.9 The inscription calls it simply aydkpa, which is in fact the ordinary designation for the
Parthenos also in the official records. The image was doubtless the principal object in the little temple
Demosthenes, it is true, was wholly a soldier, and no politician ; but still the party with which he stood was
clearly that of Nikias. In the Knights of Aristophanes (424 B.C.) the two of them, Demosthenes and Nikias, are
slaves with the ' Paphlagonian ' Kleon.
2 Thuc. iii. 114.
Cf. Arch. Ztg. 1882, 363. The Akroteria, which I have reconstructed (ibid. p. 335 sqq.), seem to belong to
the same period. The small temple on the Ilissos is also shown by the agreement of its architectural forms with
those of the temple of Nike (cf. Ross-Hansen-Schaubert, Niketempel, p. 11 ; Puchstein, D. Ion. Kapital, p. 14)
to belong to the same period. It was no doubt intended for the worship of Demeter in Agrai. The building of
the temple was probably accompanied by a renewal of the old festival of the Lesser Mysteries. This fits
excellently into the picture of the time when Nikias was in the ascendant.
4 Cf. Puchstein, loc. cit. p. 14 sqq.
'; In the original or a cast this distinction is still clearer than in the reproductions.
Published, restored, and explained by U. Kohler in Hermes, xxvi. (1891), p. 43 sqq.
^ Cf. Kdhler, loc. cit. 48. s Apud Harpokration, Nlmj 'ABt\v*.
We know of two works by Damophon expressly designated as \6ava, which had face and hands of marble ;
Paus. vii. 23, 5, viii. 31, 6. The Athena Areia at Plataia ascribed to Pheidias was a \iavov, and moreover
iirixpva-ov, but had face, hands, and feet of marble (Paus. ix. 4, 1). An inscription of Delos belonging to the
second or first century B.C. testifies to a tfavov of yet later date : Lebegue, Recherches a Delos, p. 160 : Xappipm
...dveBijKe rh £6avov.

IMAGE OF ATHENA NIKE 445
of Nike ; this was simply the aedicula for the image, which apparently stood in the centre,1
with only a narrow passage left vacant round it. The dedicatory inscription for image and aedicula
was graven on the pedestal of the image, and we have an abridgment of it in the above-mentioned
record. Thus the only two statues of Athena Nike of which we have any knowledge2 — that of the
inscription and that for which we have literary evidence as existing in the temple of Nike— are
identical. This conclusion appears to me quite inevitable, and it is no objection to it that the image
in the building is a ' temple image,' and that of the inscription a ' votive offering.' This distinction
does not belong to ancient times, if it implies that a temple image was not also a votive offering that
could be designated as such. There is a sufficient number of passages in ancient writers expressly
speaking of a temple with its image as a votive offering after a definite event ; and we suppose that
the temple of Nike and its image belong to the class which we find attested in these cases.3 Every
temple image is a votive offering to the deity ; and the ritual at the solemn 'ibpvo-is of a temple
image seems, from all the indications we possess, to have differed only in degree from that which
accompanied the dvdBeais of an ordinary ex-voto.4
The dedicatory inscription of Athena Nike, as preserved in the record mentioned above,
possesses yet another point of special interest. Among the enemies defeated, the Ambrakiots
and Kerkyraians are designated by name, but the Peloponnesians are called simply r) ev"OXirais
arparid. The only possible reason for this, as it seems to us, is an intention to spare their
feelings, i.e., in other words, the dedicatory inscription was drawn up according to the
proposal of the party of Nikias, which always aimed at peace with the Lakedaimonians. The
wording of the inscription, then, excellently corroborates the conclusion already reached, that
the temple of Nike owes its origin to Nikias and his party. Again, the tendencies of these
people are expressed in all the works of art belonging to the temple. The statue of Athena
itself shows her as the goddess of peace, not only by the motive already known to us from
the Athena Lemnia, the bearing of her helmet in her hand, but above all by the absence
of the spear, and the attribute of the pomegranate in the other hand. The pomegranate
strikes us as strange at this period ; it is a distinctly archaistic motive. In early art, it is the
favourite attribute for all goddesses connected with peace, with the promotion of fruitfulness
and growth ; in the case of Athena, we can point to only one or two instances on vases
of earlier times.5 After what we know, however, as to the art affected in the circle of Nikias,
we must not any longer be at all astonished at an archaism in an image of a god. We may
even conjecture that the whole work was strongly archaistic in tone. And we now remember
that, in the archaistic reliefs which we traced back to the tendency represented by Kallimachos,
Athena appeared quite regularly as a goddess of peace, with her helmet in her hand. There
is much, then, to suggest that the £6avov of the temple of Nike was also a work by
Kallimachos. But not less clear is the evidence given by the frieze of the temple. The combats which
we there see represented are not taken from the present, from the war against the Pelopon
nesians, but from the time of the great national struggle of Hellas against the Persians, the
time of which Miltiades and Aristeides" were the heroes— from that past to which those
1 Landron, apud Le Bas ( Voyage Arch., Archit. Athenes, PI. 2) makes the image stand in the centre, probably
from traces on the floor.
2 The Nike in bronze dedicated on the Citadel after the victory of Sphakteria, in the summer of 425 (Paus. iv.
36, 6), was equally a Nike (probably similar to that of Paionios), but not an Athena Nike.
3 e.g. Paus. i. 18, 6, Hadrian aveBi\ne the Olympieion with its image ; v. 10, 2, the temple of Zeus and the
image, were set up at Olympia out of the spoil of Pisa ; i. 14, 5, the vabs EvitXeias at Athens is an avdB-qpa airb
Mi)Suv ; the Heraion at Olympia was an avaBifpa %KiXXovvriav according to Agaklytos apud Suidas, KvtyeX. dvdB. ;
Anthol. Palat. 6, 53, EtiSripos tbv vr)bv...r6v5' dveB-qnev tip . . . Zeipvpip. Cf. also the story concerning Perikles in
Plutarch, Per. 14, where he proposes to inscribe his name on the dvaB-hpata, i.e. particularly the Parthenon and
its image : there is a similar story concerning Alexander and the temple at Ephesos.
4 The principal symbols of dedication and consecration were always woollen fillets, branches, garlands, also
oil and anointing, and sprinkling with water. At an Upvtris offerings of food were also used with the intention
of attracting the numen of the deity to the place (this is clear from Aristoph. Plut. 1197 with the scholia). This
important circumstance is completely passed over even in the newest handbook of ancient sacred ceremonies
(Stengel). 5 I have mentioned in Roscher's Lexikon (i. col. 689, 1. 39) an Attic Oinochoe of the later black-figured period
at Altenburg, which shows Athena seated with the pomegranate in her left hand. Cf. with it the Lekythos of
circa 500—480, given by Kekule, Balustr. p. 25, and Arclu Ztg. 1885, PI. 12, 2. Benndorf's attempt to derive the
pomegranate from Side has been already refuted by Kekule.
6 That these two names just summed up in themselves the ideals of the conservative peace party is shown
by the beautiful passage of the Knights of Aristophanes, v. 1316 sqq., 1325 ; this play is contemporary with the
building of the temple of Nike.

446

THE TEMPLES OF ATHENA ON THE AKROPOLIS

who followed Nikias's way of thinking, and shrank from the war with Sparta, looked as their
ideal. In the detailed interpretation of the frieze it is true that unanimity of opinion has not been
attained. All that is undoubted is that the long sides represented combats of Greeks and
Persians. The west side shows Greeks fighting against Greeks, the east side an assemblage of
gods. But W. Vischer and Overbeck have already rightly shown1 that the four sides must of
necessity be interpreted as a single whole.2 The reliefs run without a break round the whole
temple, just as in the frieze of the Parthenon. They must, like it, represent a single action. And
this, as has been already recognized by these writers, can be no other than the battle of Plataia.
The view has also been advocated that only a ' general representation of all the combats of the
past' is intended,3 but this seems to me foreign to ancient notions. It is a mistaken conception
of the ideal character of Greek art that ascribes to it such obscure and confused generalizations.
No evidence for the battle of Plataia is to be found, as Overbeck imagined, in the supposed
Boeotian form of helmet on the west side;4 but there is evidence in the intentional arrangement
which places the comb it of Greeks at the west end as a distinct integral part of the battle, between
the two principal combats against the Persians.
This suits no other battle than Plataia, and suits
it excellently. That battle, as is well known, was broken
up into a number of encounters distinct in time
and place. Of course we must expect to find

B'ig. 181. — Left-hand corner of the south frieze of the Temple
of Athena Nike. (From Lebas, Voyage Archiol.)

special emphasis laid on the part played by the
Athenians. They could boast of two successful
actions : one, of which they were especially proud,
took place quite at the beginning,5 when three
hundred picked men of the Athenians, led by
Olympiodoros, son of Lampon, voluntarily under
took to oppose the Persian cavalry under Masistios,
the other Greeks having refused to do so. The
most momentous episode of this fight, which made
a powerful impression on both sides, was the death of the brilliant Persian leader Masistios.
His Nisaian horse was wounded in the side, reared high on his hind legs, and threw his rider,
whom the Athenians then killed. An exact illustration of this incident related by Herodotos
is to be found in the slab in the left corner of the south frieze (Fig. 181). Here a Persian more
richly dressed than the rest — the story speaks of the rich robe of Masistios — is falling from a
horse who rears up and has evidently been wounded in the side. This group is not, like many
others in the frieze, of a typical nature, and so far as I know it nowhere recurs in this form ;
it has a thoroughly individual character, which is excellently explained by the episode of
Masistios. On the next slabs of the south side was carved the continuation of the fight
of the Athenians against the Persian cavalry. If we accept the arrangement of the slabs
advocated by Kekule" ° — and there is much that favours it7 — the battle here is much more exclu
sively a cavalry engagement than is the case according to the arrangement proposed by
Ross. The west side, again, undoubtedly represents the second combat of the Athenians —
namely, the share which they took in the main action by fighting with the Hellenes on the
Persian side, especially the Thebans. This was a very fierce conflict. The inscription on the
golden shield dedicated by the Athenians at Delphi after the battle even ran : dirb Mij&W Kal

1 W. Vischer, Erinnerungen und Eindriicke aus Griechenland, p. 131. Overbeck in Bergk's and Casar's
Zeitschr. fur die A Iterthumswissensch. 1857, p. 289 sqq. ; Gesch. d. Plastik, 3rd ed. i. 363 sqq.
2 This overthrows also the most recent interpretation proposed by Yorke (J. H. S. xiii. 279) ; he holds that
Marathon, Salamis, and Plataia are represented. The battle by the ships at Marathon he supposes not to have
been indicated from want of space (!)
3 Friederichs, Bausteine, p. 189 ( = Fr. -Wolters, p. 283).
4 The ' pilos '-shaped helmet is not specially Boeotian, as has been already noted by Friederichs, loc. cit.
The Kpdvos lioianovpyes recommended by Xenophon (De Re Eques. 12, 3) to horsemen may indeed be judged
from the description to have had the ' pilos ' shape, but this does not imply that it was specially Boeotian,
since it was much used by other Greeks also.
5 Herodotos, ix. 20 — 24. Cf. Busolt, Griech. Gesch. ii. 197 sqq.
6 Kekule, Balustr. (1869), p. 17.
7 The correspondence of the slabs G and O (Ross) is such that one would be decidedly inclined to bring
them to one side. Kekule's proposed arrangement for the south side produces a fine composition complete in
itself.

FRIEZE OF THE TEMPLE OF NIKE

447

erjfiaicov.1 The third, northern, side probably referred to the main action of the Lakedai-
monians and Tegeans against the Persians, and to the conclusion of the whole. Unfortunately
there is but very little extant on this side, if we follow, as seems best, Kekule"s arrangement
of the slabs. To it belong the fragments of Slab f. (Ross), which shows the Greeks
fighting against infantry. Their opponents are draped, and are probably Persians ; but,
unfortunately, decisive indications are wholly wanting. To this side belongs moreover Slab m.
(Fig. 182), where on the left appear indistinct remains of fallen figures. We see a draped
man, who is therefore probably a Persian ; then follows a Greek, who seems to be striking
down at him from above with a spear ; with his right foot he treads on a head,2 which is
severed from the naked corpse lying stretched out beside it ; 3 this corpse appears to have
been stripped and mutilated. Next follow two Greeks hurrying at a run towards the right ; in
front of them a Greek in a chiton4 has his left knee resting on a mound of earth, below which
lies a draped and bearded corpse ; to judge by the outline of the head the latter is a Persian ; 5
the Greek was clearly purposing to spoil the corpse ; he is hindered in this by a Greek who
opposes him, laying hold of him so vigorously with his left hand that the helmet falls from his
head. Behind this attacking Greek two riderless horses are galloping away. Thus we get the
following points : Battle of the Greeks with Persian infantry. The Greeks are victorious. They
are hastening at a run, probably in pursuit of the fugitives, over the field of battle. The field
is full of corpses, which are being plundered. But the retreat of the enemy is not wholly
uncovered : it is being protected by Greek allies, apparently cavalry ; they, however, are also

Fig. 182. — Portion of the north frieze of the Temple of Nike. (From Lebas, Voyage Archeol.)

being overthrown, so that the riderless horses leap away. All this, too, answers to the battle of
Plataia ; the principal action was that of Lakedaimonians and Tegeans against the Persian foot.
The Persians fled to the camp, the Greeks pursued, no resistance being offered except by the
cavalry, especially that of the Boeotians0 (this accounts for the Greek beside the galloping
horses). They made their way in spite of this opposition to the camp, which was then taken,
the Athenians especially helping in the capture. It is possible that a slab is lost which came
next and indicated the storming of the camp. Of the last slab, which came next beyond, we
have still some slight remains. Here there is no longer any battle. As in the east frieze of the
Theseion at the right-hand end there appear tranquil figures, evidently indicating the results of
the battle, so here we see first of all a low podium, and upon it apparently the foot of some
piece of furniture, shaped like the claws of a wild beast, 7 then the lower part of a large vessel ;
towards these a figure is running — there remain parts of the bare lower portion of his legs ; 8

1 Aeschin. in Ctes. 116.
2 This cannot be seen in the reproductions ; the foot treads on the face.
3 '1 he head lies in quite a different line from the body, and also too far away to be considered otherwise than
as severed from the corpse. On the latter there is only a small cloth between the legs. The right arm also seems
to have been hacked off.
4 The Corinthian helmet that is falling from his head proves him irrefutably to bea Greek.
6 Only the outlines can now be seen, since the figure is much injured. The way in which the head is placed
on the shoulders and the outline of the beard are quite similar to what we see in the Persians of the frieze.
Trousers cannot now be made out.
6 Herod, ix. 68.
7 Clearer in the drawing by Landron, apud Le Bas, than in Ross. I have been obliged to depend on the
publications for this slab, as there is no cast.
8 Landron gives a small fragment remaining of the left foot, which is placed far back.

448 THE TEMPLES OF ATHENA ON THE AKROPOLIS
then follow the legs of a boy who is standing still. After the battle there was erected, as is
well known, an altar to Zeus Eleutherios, and a festival instituted, to be repeated every fifth
year, at which there was a contest of runners who ran armed before the altar.1 This, I am
inclined to infer from the remains, was indicated on the slab : on the left a step of the altar,
or the basis of the holy table of Zeus, in front of it a great krater, pertaining to the festival,2
and then perhaps a representative of the armed youths running in honour of the god.
The assembled gods on the east side are to be understoodjust in the same way as the deities that
in the frieze of the Theseion appear among the combatants : they are taking an interest in the battle,
and it is they who bestow the victory. There on the Theseion, where there was only a single side
at the artist's disposal, they are introduced at two suitable places in the battle itself; here, where the
representation extends round the four sides of the temple, a whole side is reserved for them. We
must of course suppose complete unity of time, and the gods can only be conceived as present at the
battle.3 From this alone it is clear that the latest interpretation put upon the east frieze by Bruno
Sauer4 cannot be correct : he considers that we have here a trial scene, taking place previous to the
expedition of Dareios. Hellas sits on the defendant's bench, Asia is the plaintiff, the Greek gods are
the judges, who are to declare their sentence by means of voting tablets. A maiden on the right acts
as 'herald of the court,' and summons any of the public to come forward before the assembled judges
and speak on behalf of the defendant. This strange interpretation, however, rests at its very outset
on a false conception. How could Asia enter a complaint against Hellas with the Greek gods, and
how could these be the impartial judges of their own land ? This is quite opposed to ancient thought.
The Hellenic gods can only be on the side of Hellas.5 Asia is hurled to ruin through her arrogance
and blindness,6 through her contempt for the Hellenic gods and their laws. Even before the battle,
according to the Greek conception,7 the decree of the gods against the Persians is fixed, and,
according to Herodotos, Mardonios himself knew of an oracle to the effect that the Persians must
be overthrown in Hellas.8 Equally impossible are all the details of the interpretation. The name of
Hellas is given to a figure who turns her head away from the gods in the centre of the frieze,
whereas the only possible representation of her would be as suppliant and protegee of Zeus and
Athena, as she appears on the vases. And Asia is supposed to be sitting enthroned among the Greek
gods as though she belonged to them. And where is there a trace of an indication that the gods
intend to advance to the urn and deliver voting pebbles ? Of the names which Sauer proposes for
the individual gods, there is especial reason to doubt those of sea and earth, Thalassa and Gaia,
which he gives to two corresponding and entirely similar female figures : they must at least have been
different from each other ; and, as a matter of fact, Gaia was at that time regularly represented as
rising up from the earth.9 It is, however, a downright mistake to divide off Fig. 8 in the left half of
the frieze from the group to which she necessarily belongs in view of the motive and the corre
spondence with the other side, and to make her act Ate or Eris to the supposed Asia (Fig. 7). The
centre has three figures (Poseidon, Athena, Zeus) ; on either side of these is a group of five standing
deities, each group divided into two pairs and a single central figure : the symmetrical arrangement
is very clear : it is therefore inadmissible to dissever a member of the group on either side. Then
follows in each half a sitting personage, and several maidens in rapid motion towards the centre ; on
the left are three ; on the right only two are preserved ; 10 but symmetry, which has been regarded
hitherto, demands three maidens here also. It is clear that in this Olympian assemblage we have
before us two of those triple sisterhoods of divine maidens which from old times — we may instance
the Moirai, Horai, and Charites of the Francois vase — artists were fond of introducing into
1 Beovcri 5e uirXiapevoi irpb rov 0apov, Paus. ix. 2, 6.
2 Cf. e.g. the sacred table with a krater in front of Dionysos (Mon. d. Inst. 6, PI. 5 b). At the yearly
feast of the dead held by the graves on the battle-field of Plataia, a krater was set up, from which the Archon
poured libations (Plut. Aristeid. 21).
3 I can think of no argument to be brought against this natural supposition, and do not understand why
Wolters, Gipsabg. p. 284, assumes that the east side is ' certainly' to be separated from the others.
4 Aus der Anomia, 1890, p. 96 sqq., 'das Gdttergericht iiber Asia und Hellas.' Kopp (Arch. Anzeiger, 1892,
p. 128) adopts his interpretation 'in the main.'
6 Only the deities Helios and Selene, whom the barbarians worshipped, are conceived in the Peace of
Aristophanes (406 sqq. ) as conspiring against Hellas. The Peace is of about the same date as the Nike frieze.
6 Apate on the Darius Vase, and in Aischylos.
7 Clearly seen in Herod, ix. 16.
8 Herod, ix. 42. Mardonios does indeed sacrifice before Plataia according to the Greek rite (ibid. 37), but he
does not trouble himself with the way in which the sacrifice turns out (ibid. 41).
9 The Greek representations of Gaia as a whole figure, conjectured by Kuhnert in Roscher's Lexikon, i.
'575 sqq-, are all very uncertain.
Sauer (loc. cit. p. 102 seq.) has shown from their motives that they have no specially close relation to the
sitting figures.

FRIEZE OF THE TEMPLE OF NIKE 449
processions of the gods. The swift dance-like advance would be specially appropriate for Nymphs,
Horai and Charites, whom we are inclined to suggest as most probable here. The immediate
proximity of Aphrodite makes it likely that the group on the left ought to be identified as
Charites. Starting from our conception of an assemblage of the deities interested in the battle of
Plataia, we venture a step further in explaining details, though it is improbable that the figures
of a work so much mutilated can ever all be named. The principal action, the encounter of
Lakedaimonians and Tegeans with the Persians, which we conjecture to have been represented
on the north side, took place by a shrine of the Eleusinian Demeter ; and this, in fact, as the
' soil of the Eleusinian Demeter ' on which the battle was to take place, plays its part in the
oracle of which Plutarch tells us as having been given to Aristeides (Arist. 1 1). This being so,
there is a special fitness in the group of Demeter and Kore placed in the right half of the
frieze — towards the north.1 But the oracle gives us yet further hints: it promises the victory
when prayers are offered to Zeus, to Hera Kithaironia, to Pan, and the nymphs Sphragitides.
Zeus and Hera are undoubtedly in the frieze. But we can now go further, and pronounce that
the maidens hastening forward on the right are the nymphs Sphragitides. In fact, it was in
this fashion, with hovering advance, as it were, that nymphs generally appeared in votive reliefs
also. Pan is to be looked for quite close to them : we may safely conjecture him to
have been on the right, at the missing end of the frieze, where he formed an appropriate
pendant to Aphrodite on the left. Sauer indeed imagined that the second figure behind Zeus
was Pan, but this is impossible from the whole type of the figure.2 On the other hand, neither
this nor the other two standing male figures of the frieze can be reconciled with the types of
the greater gods — Apollo, Hermes, Ares, Dionysos, and we must apparently suppose them to be
heroes. Now it just happens that the oracle we mentioned commanded sacrifice to be made to
various Plataian heroes, first and foremost to Androkrates, by whose shrine the Greeks had
taken up their position at the beginning.3 We may probably suppose him to be the man near
Hera. The corresponding hero on the other side Sauer was probably right in identifying as
Theseus. And he is also right no doubt when he regards the object visible in front of Zeus
as a table very much foreshortened ; but for a voting urn, such as he goes on to suppose,
there was evidently no place upon that table. It was a sacred table, like that brazen trapeza
which stood- in front of Zeus Polieus on the Akropolis, and upon which the sacrificial ox was
offered at the Bouphonia.4 On the north side also we saw, as we thought, a sacred table of
Zeus Eleuthereus ; we may suppose garlands for the victors in the contests to have lain upon
it. Thus a similar table would be very suitable here on the east side, as an indication of the
victory bestowed by this Zeus the Deliverer.
But why was the battle of Plataia in particular chosen for our temple of Athena Nike? It seems
to me that the tendency of Nikias and his peace party manifests itself nowhere more clearly than in
this choice. Plataia was just that battle in which Athenians still fought side by side with Lakedai
monians in friendly union. And this was the political ideal of Nikias's party— agreement with the
foes, in order, as Aristophanes expresses it in the Peace (1082),5 to share with them the rule of Hellas
(koivjj rr/s 'EXXdSoj apxeiv). Thus, instead of representing the victorious battles against Pelo-
ponnesians and Ambrakiots, the frieze transports us back to the good old days when Athens and
Sparta fought as comrades at the head of Hellas against the national foe.
If we suppose that the building was not very rapidly executed, the frieze, which probably was not
taken in hand till towards the end of the work, may quite well belong to as late a date as about 423,
i.e. to the time when (in the spring of 423) Nikias and his adherents brought about a year's armistice
with Sparta, of which use was made for serious peace negotiations. The meaning of our frieze would
be especially clear if it originated then.6 But even in 425 the inclination for peace was sufficiently
1 Quite correctly identified by Sauer, p. 104, following Kekule.
2 Sauer is inaccurate in his statement (p. 101) that only the upper part of the leg, not the lower, is attached
to the background ; a good way below the point at which the knee must be placed, as far down as the place
where the calf begins to grow thinner in a human leg, it is attached to the background ; this is irreconcilable
with the supposition of a goat's leg. Moreover, the attitude, and above all the long robe on the left arm, are
quite unsuitable for Pan.
3 Herod, ix. 25.
4 Porphyr. De Abslin. 2, 30. An Athenian bronze coin shows a trapeza before the image of Dionysos
enthroned: Imhoof-Blumer and Gardner, Num. Comm. PI. CC. 4. _
5 It is the old principle of the party, pregnantly expressed in the well-known aphorism or Kimon s, that Athens
should not draw at the yoke without her yoke-fellow (Plut. Kim. 16.)
6 The first Peace of Aristophanes is supposed to belong to 422 ; that which is now extant dates from the
3 M

450 THE TEMPLES OF ATHENA ON THE AKROPOLIS
powerful to make these sculptures quite intelligible ; the Acharnai of Aristophanes, performed in the
spring of that year— with Dikaiopolis and his thirty years' private peace with Sparta— is an eloquent
proof of this.
When finally we pass on to the artistic character of the frieze, we find that a specially noticeable
point is the pictorial treatment of the relief, which contrasts strongly with the style of the Parthenon
frieze,1 and connects it closely, on the other hand, with the frieze-reliefs from Asia Minor belonging
to about the same time, in which we presumably have specimens of Ionic art.2 These pictorial
peculiarities consist in the indication of several planes one behind another, in the oblique arrangement
and the foreshortening of objects and limbs, and in other points of the same kind. The Nike frieze
especially resembles, both in motives and in stylistic details, the best reliefs of the Nereid monument
from Xanthos. Yet there are equally manifest differences from Ionic work, which serve to charac
terize the frieze of Nike as Attic, such as its Xen-rdVr/r, the elegant delicacy which prevails amid all
the vehemence of the fighting groups.3 Among Attic works it has long been recognized that the
frieze most resembles the sculptures of the Erechtheion.4 The similarity between the standing
goddesses of the east frieze and the Caryatids of the south porch is at any rate complete. The
latter were executed before the interruption of the work at the Erechtheion ; they therefore belong
before 413, and come near in time to the temple of Nike. The figures of the Erechtheion frieze are
not earlier than 409 — 408 ; they are inferior work by stone-masons, and do not lend themselves to the
more delicate comparisons of style. The magnificent Caryatids, on the contrary, may with probability
be assigned to Kallimachos. We have here, therefore, a strong ground for conjecturing that
Kallimachos was also the artist of the frieze of the temple of Nike, whose image within we have
ascribed to him.
Kallimachos, as we saw, introduced from Ionia into Attica the so-called ' akanthos decora
tion ' ; moreover, it was probably he who fashioned the charmingly graceful Torus-capital of
the Erechtheion after Ionic models.5 It seems probable that a third innovation — namely, the
introduction into Attika of the Ionic pictorial style of relief — must also be referred to him.
Speaking generally, when we consider in connexion with Attic art the peculiar style of floating,
clinging drapery both of the Xanthian Nereid monument and of the Gjblbaschi reliefs, we are
constrained to suppose a second period of Ionic influence upon Attic sculpture in the time after
Pheidias. Now Kallimachos seems to have been especially prominent as a representative of this
Ionic style, though he refined it considerably, and freed it completely from that monotony, amounting
to hardness, which it still displays at Xanthos.
The most dazzling perfection of this manner, however, is seen in the splendid reliefs with which the
balustrade round the temple of Nike was adorned. These were added later, very probably after the
great victories of Alkibiades — about the same time that the resolution was passed for the completion
spring of 421. The gods themselves have moved away because they are wroth with the Hellenes (v. 203 sqq.),
who are making war upon each other ; and Homer is summoned to curse the ir6Xepos eiriS-qpios (v. 1097). These
were also the views of the negotiator of the peace, Nikias.
1 Cf. with regard to this, especially Bruckner (Ath. Mitth. xiv. p. 403 seq.) The relief of combatants
published by him there seems, in fact, to be remarkably near akin to the Nike frieze, and might be the work of the
same artist. In this connexion it is not without significance that the Pythodoros, son of Epizelos, who, according
to the excellent conjecture of von Wilamowitz, was the dedicator of the relief, was a thorough adherent of the
oligarchical party led by Nikias. He was one of those who swore to the Peace of Nikias ; he proposed, as we
now know from Aristotle ('AB-qv. iroX. 29, 1), the bill for the establishment of the Four Hundred, and was himself
one of them ; later also he was one of the Thirty.
2 That the Nereid Monument of Xanthos and the Heroon of Gjolbaschi belong to the same period, and that
this is in the fifth century, I have noticed Arch. Ztg. 1882, pp. 359, 368 ; Wolters ( Gipsabg. ) and Benndorf (Heroon
von Gjdlb.) have subsequently followed me in this view The opinion which I also expressed there, that these
works are of original Ionic style and do not imitate Attic models, I likewise still believe to be correct. One may,
however, go higher up in the fifth century for the date than I did then. The Dioskouros of the one Akroterion
from Xanthos (cf. loc. cit. 347) whose head is extant shows a style which can hardly be placed after 440,
as it resembles the metopes of the Parthenon. In/. H. S. xiii. 132 seq. J. Six places the Nereid monument
in the time of King Perikles, i.e. in the first half of the fourth century.
3 The frieze is, when closely examined, of excellent workmanship, fresh and delicate. At a distance it is
not effective ; this fact and its seriously mutilated state are probably the causes of its having repeatedly been
unjustly estimated. An intrinsic difference between the east side and the others (Kekule, Balustr. p. 27) I cannot
perceive. 4 Cf. recently A. IT. Smith, Catal. of Sculpt, in the Brit. Mus. i. p. 241.
5 The observation that the capital of the Erechtheion is an old Ionic type ' ennobled by the Attic sense of
form ' is due to Puchstein, Das Ionische Kapitdl, p. 29. The inscription of 409—408 names the architect Philokles
as a member of the commission elected by the people for the sake of the resumption and completion of the
work at the Erechtheion. The building was however by that time completed in the main, notably the columns
were standing. The design for them and for the whole ornamentation is certainly with more probability to be
assigned lo the gifted Kallimachos, the maker of the lamp within, and inventor of the Corinthian capital, than to
the obscure Philokles, of whom we do not hear till the work is almost complete.

THE PEDIMENTS OF THE PARTHENON 451
of the Erechtheion (409— 408).1 Any one who compares, for instance, the charming figure in the left
corner of the east frieze of the temple (Peitho) with certain fragments of the balustrade will be
inclined to ascribe the latter also to the same artist. But this we will leave undecided. In any case
the artist belonged to the same Ionicizing school, which tended to a wide divergence from the Pheidian
style,2 inclining on the one hand to greater impulsiveness and effect, on the other to greater elegance
and subtlety in expression. Truth is now but slightly regarded ; the difference between the linen stuff
of the chitons and the woollen stuff of the mantles falls quite into the background ; in fact — in the
famous ' Sandal-binder,' for instance — the chiton and mantle are so little differentiated that the edges
of the two pass into each other. The untruthful manner of this style led to the reaction at the beginning
of the fourth century, when the artists returned to nature in a new fashion.3
We have tried to reach a comprehension of the temples of the Akropolis from an historical point
of view. We have found them to be in very close connexion with the struggle of the great parties
which extends throughout the whole of the fifth century. The magnificent productions, the Parthenon
and the Propylaia, were the creations of the men, Themistokles and Perikles, who gave Athens a
standing of her own independent of Sparta, and brought into existence the splendid Attic Empire.
The elegant small buildings, the Erechtheion and the temple of Nike, arose from that reactionary
school which for a time, after the death of Perikles, came again to the front. The old-fashioned piety
of this party found its expression in the Erechtheion, their efforts after peace with Sparta in the
temple of Nike.
The tendencies of this school also had a broad basis in the people. Comedy is the best witness
to this : she was wholly on the side of this party. Aristophanes was not a great patriot, as is often
asserted, but a pure partisan. Nay, it was simply unpatriotic to goad on the allies against Athens as
he did in the Babylonians (426 B.C.) No doubt the whole Attic Empire was repugnant to the
oligarchical party : the treatise On the Athenian Constitution admits one deep into the embittered
soul of a politician of this school.4 Aristophanes was also a worthy colleague of Hermippos, who
brought Aspasia to trial for free-thinking (do-e/3eca) ;° he was not ashamed to sacrifice a Sokrates to
the laughter of the mob, and to drag him in the dirt because his teaching was enlightened. At
the same time he preached peace without regard to what was demanded by the dignity and
power of Athens.
The Akropolis, however, received artistic adornment even from this party, and as Perikles had
Pheidias, so Nikias and his adherents found their artist in Kallimachos, a genius in what was decorative,
elegant, rich, and dazzling, rather than great. And his archaistic manner enabled him to satisfy the
piety of those who gave him commissions. But the symbols of the true greatness of Athens are
and will ever be the creations of Perikles, the Parthenon and Propylaia.

VI. The Meaning of the Pedimental Sculptures of the Parthenon.
The explanation of a number of the figures from the pediments of the Parthenon is still
completely arbitrary and uncertain ; and we are obliged again and again to make fresh efforts to
reach firmer ground in this respect. This is more especially the case with the west pediment, with
which we will accordingly begin.6 We must first of all be clear with regard to the basis of our
knowledge. This is afforded, over and above the extant remains, first of all by Carrey's drawing.
For the so-called Anonymous of Nointel is valueless ; his drawing is nothing but an old and very
unskilful copy from Carrey. It tries to unite into one general picture of the temple facade what is
given in the two separate sheets of Carrey's original. This was rightly perceived some time ago by
Petersen,7 and the published facsimile of the drawings by making comparison possible seems to me
to exclude all doubt.8 In Carrey's work every stroke shows that he had the actual building and its
sculptures before him ; in the drawing of the Anonymous it is equally clear that he worked simply after
Carrey's model, without ever having set eyes on the Parthenon. We will call attention to one or two
1 The deductions of Michaelis (in the Ath. Mitth. xiv. 364 seq. ) with regard to the date of the balustrade
seem to me inevitable. That it is later than the temple has been shown by Bohn on decisive technical grounds.
2 Cf. also Michaelis, loc. cit. 364.
* Cf. also Samm. Sabouroff on PI. 20, p. 2 ; Skulpt. Introd. p. 14 seq.
4 Cf. also what Diimmler has written in Hermes, 1892, 260 sqq., with regard to the treatise ot l<.ntias.
5 Plut. Perikl. 32. , , . . , . , c„ • .„ ¦
6 I have given a short sketch of the views set forth below, at a meeting of the Archaeological Society in
Berlin (Arch. Anz. 1891, p. 70).
7 Fleckeisen's Jahrbiicher, 1872, 296, 307.
8 Antike Denkm. i. PI. 6, 7. Cf. Deutsche Litteratur-Zeitung, 1887, p. 1313.

452 THE TEMPLES OF ATHENA ON THE AKROPOLIS
points only.1 The strange and ugly capitals of the Anonymous are completely explained by Carrey,
who drew them from a high point of view ; thus the Anonymous mistook the echinus for a
thick round roll. Carrey's hasty two lines on the neck he turned into two more rolls. The arbitrary
and unmeaning lines which the Anonymous uses to indicate the remains of the metopes could never
be the work of one who had the original before him, but are completely explained by the failure of the
Anonymous to get sufficient information out of Carrey's indications, which are everywhere slight, and
moreover extended only to the left half of the facade. In figure A the curious form of the belly which
projects to the left is clearly due to misunderstanding of the indefinite shadow given by Carrey in this
place. The effort to make the pediment look as much filled up as possible led the Anonymous to do
away with the gap between A and B. But the relation between the two drawings is especially clear in
the case of the female figure C ; the absolutely unmeaning shape given by the Anonymous to the lower
part of her body could only have arisen from a misunderstanding of Carrey's work ; with the original
before him the draughtsman could not have represented it so. In B, again, the Anonymous gives
only the upper part of the right arm, whereas even Pars saw part of the forearm still left ; Carrey
saw the whole forearm, bent towards the head, still complete with its hand ; he indicated it
however but slightly, and from a point where it was parallel to the line of the upper part of the arm.
Thus the stump of arm given by the Anonymous is again to be explained by a misunderstanding of
Carrey's work. The same has happened with A ; that figure too still preserved in Stuart's time part
of its right forearm ; in Carrey's view, taken from the centre, the arm was very much foreshortened,
and moreover in shadow ; Carrey gives only a slight and not clear indication of the arm. This the
Anonymous has once more interpreted wrongly, and drawn clearly a short stump of arm, whereas the
original at that time possessed a great piece more.
But enough of these proofs, all of which are taken from the three first figures alone, and which
could be added to by comparison of almost any one of the personages that follow. Thus the
divergence in D between the Anonymous and Carrey arises from the fact that the former did not
understand the rapid movement of this figure from Carrey's sketchy drawing, and therefore turned
it into a regular sitting figure quite at rest : and many other cases might be instanced. We need now
only mention the arguments adduced in favour of the Anonymous by his latest supporter,
H. Bliimner.2 They were, it is true, written before the publication of the drawings, and can be
disposed of by a mere glance at the facsimiles. It is inconceivable that the Anonymous took
original sketches from various points and subsequently combined them, as Bliimner imagined. Had
he for instance drawn A independently from the front, the lower part of the leg, which, as Stuart
shows, remained, must have been visible ; as a matter of fact, it disappears, exactly as in Carrey's
drawing, where the phenomenon is explained by the point of view selected. As for Blumner's
assertion, that the Anonymous draws the right foreleg of the hinder horse, which Carrey has not got,
on account of his different point of view, the inexact publication in Michaelis was solely responsible
for this error, since the facsimile shows that the Anonymous was simply copying Carrey. The only
divergence in the horses is that he thought well to add a second hind leg in the place of the slight
and obscure lines which Carrey gives at this point. The different and more correct attitude of the
head in B and C ascribed to the Anonymous is again a mistake arising from Michaelis's plate, and
refuted by the facsimile, which shows his dependence on Carrey in this point also. The second
dolphin and the left leg of W were evolved by the Anonymous himself, Carrey giving only the slightest
indications at these points, and he drew V more upright because he could not reproduce the strong
foreshortening of this figure as it appears in Carrey's drawing. In the group P Q the Anonymous
has especially gone astray in his interpretation of Carrey ; he draws the torso of a boy sitting
on the lap of a woman, whereas the boy is in fact standing on the ground behind her right leg ; this
mistake could not have been made by any one looking at the original, but might very well arise from
a misunderstanding of Carrey's indications.
Thus we must definitely strike out the Anonymous as a source of our knowledge of the west
pediment. The first inference from this is that we have to follow Carrey in the gaps that he gives
as well as in all else. There is, then, a figure missing between U and V as well as one between
A and B. Hitherto it has, as is well known, been customary,3 in spite of the great inconsistency
of such a proceeding, to reject the Anonymous for the left angle of the pediment, and accept the
x My comparisons naturally refer only to the facsimiles of the drawings ;" the reproduction given by
Michaelis is well known to be inaccurate.
2 In Gesamm. Stud. z. Kunstgesch., Festgabe fiir A. Springer, 1885, p. 240 sqq., 252.
This has been the prevalent theory at least since Welcker : on the other hand, Brdndsted, O. Miiller, and
Leake, and more recently Waldstein also (Essays on the Art of Pheidias, the reproduction ibid. p. 133, is given
without discussion), have followed Carrey in supposing two gaps.

THE WESTERN PEDIMENT 453
gap given by Carrey after A, but on the other hand, in the right half of the pediment, to put faith
in the Anonymous alone. And yet every one must admit — and in fact even Michaelis does so1 —
that wherever Carrey's drawings can be tested by the originals they show complete correctness
precisely ' in the arrangement of the figures and the observance of the proper distances between
them,' so that there is no excuse for disregarding the gap which Carrey saw and marked in his
drawing after U.
Moreover, in this case also the extant remains furnish a corroboration of Carrey's evidence.
Bruno Sauer, in the commentary to his highly meritorious survey of the traces still remaining on the
floor of the pediments,2 does indeed base his remarks on the prevalent theory, which in the right
half of the pediment follows the Anonymous and denies the gap. But the very traces he has
studied refute his own theory.
In Slab 20, on the left (see Fig. 183), by the southernmost of the two dowel-holes situated here,
Sauer has correctly observed the mark of what must be the limit between T and U.3 On the other
hand, the position of V is fixed by the mark of the left hand as recognized by Sauer, and by traces of
the chisel and two dowel-holes on Slab 22. The figure can only have projected a. little with its right
knee over Slab 21, if the existing torso is combined with that mark of the hand ; and this shows that
the figure was brought up quite close to W, which accords with Carrey's evidence.4 Thus there
remains, if we suppose no gap between U and V, a space of nearly one slab and three quarters in
breadth for U. This would do very well for a recumbent figure with outstretched legs, but for
a seated one— and such was U, according to Carrey — it is considerably too much, however broadly
the figure sat. What we must therefore assume here is a group of two closely united figures. And in
fact the attitude of U in Carrey's drawing itself points to this, being quite unintelligible if the figure
were not resting the upper part of the body against another figure to be supplied in the gap, and
which we shall call U*.
It is true that the corresponding group on the other side (B C) takes somewhat more space
(about the width of two slabs), but to make up for this the group E F on this side is satisfied with
only the width of one slab, whereas the recumbent group S T is considerably wider, which must
necessarily have made the space for UU* more confined. The clear space belonging to V,
which breaks off to the left on Slab 22, must have passed into that of U*, on Slab 21, just as the
empty spaces of D E F pass into each other.
Thus the existing traces serve only to confirm the correctness of Carrey's drawing. We should
further remark that the condition of the slabs of the floor may even disclose the cause which brought
about Carrey's two gaps. In Slab 4, where is the gap of the left half of the pediment, the larger
half in front is wanting. This was, as Sauer rightly observes, broken at an early period, probably
by the fall of the block of the cornice above it, and this also knocked down the figure, which was
already missing in Carrey's time. Now the same is the case with Slab 21, on which we must, as we
saw, place the second gap given by Carrey ; here also the front half of the slab is wanting, and no
doubt heie also it has been carried away with the figure by the block of the cornice above. Carrey,
on the one sheet where he gives the pediment, draws it complete, no doubt from aesthetic considerations.
But Dalton shows the slabs on which the figures stood as broken away in front just at the places
which we have indicated as those of the two gaps. Dalton is inaccurate simply in the matter of the
arrangement and drawing of the figures.6
It is only by means of that second gap, which we may now consider as proved with certainty,
that we can possibly make out that the most necessary requirement of symmetry was observed in the
pediment— namely, equality in the number of figures in the two halves. The two children in great
part hidden behind Q can of course not be counted in : they are only equivalent to attributes. The
ten persons of the left half of the pediment must have ten likewise in the right half answering
1 Parthenon, p. 102 seq. 2 Antike Denkm. i. 58 ; Ath. Mitth. xvi. PI. 3, p. 59 sqq. (our Fig. 183).
3 Sauer has called my attention to the fact that the dowel-hole naturally marks the lower limit of the figure
T, not the upper, which lay farther out. But this comes to the same, since, as Carrey shows, the lower part of
U's right leg went out in an oblique direction. Thus U followed immediately upon the lower limit of T, not merely
upon the upper ; and on the other side the upper limit of U, according to Carrey, projected but very little over
the lower. Sauer is therefore wrong in thinking that a quite unusually large space may be claimed for U. _
4 Carrey shows also that V is to be turned even more towards the front than has been done m the drawing by
Sauer ; the vacant space also agrees with this ; the knee of the figure, stepping out obliquely towards the left
(north), projected into Slab 21. , „, , , . , .
5 Dalton brought S T far too much to the right ; they would, according to him, stand on Slabs 20 21, which is
impossible. The torso W, which even now remains in the pediment, he left curt entirely. ^ That the figure which
he plac
Room,unreliability 1

1A. *H^-I

i."te

t5.^

TTvl

IS £1-

A* B C D E F G H J K L M N O pQr S T U U* V
Fig. 183.— Western pediment of the Parthenon, showing the traces on the floor of the pediment.1 (From B. Sauer's drawing, Ath. Mitth. xvi, plate iii.)

1 [The letters corresponding to the figures are given to facilitate following Furtwangler's arguments ; they must, however, be taken as rough indications, and not as marking accurately the
place occupied by the figures on each slab. — E. S.]

THE WESTERN PEDIMENT 455
to them. The three women C D F on the left with the youth E correspond without doubt to the
three women Q T U on the right with the youth S.1 Accordingly, B also must have had a pendant
on the right (U*), against whom U leaned as C against B. We must therefore conceive U* as a man
sitting on the ground like B.
Sauer by his excellent investigations has arrived at the result that the east pediment had a com
position similar in the main to that of the west pediment. Now Sauer has proved an exactly
corresponding number of figures for the two halves of the east pediment, and this number is ten
each side. The inference with regard to the west pediment is inevitable : not till we supply the man
who is wanting, U*, does the correspondence and symmetry of the two pediments become complete.
It only remains for us to say something with regard to the places of figures N— T in the west
pediment, since the way in which Sauer has determined them will not quite stand. It cannot be right
to regard the dowel-hole occurring about the middle of Slab 18 as the limit between O and P O R, as
he does : it is more likely the limit between P O R and S T. For his supposition makes the space for
the wide recumbent figure S with T far too scanty, since Q with S T together would have scarcely the
width of two slabs at their disposal. On the other hand, towards the left the space for O would be
far too big, and symmetry would be seriously outraged, since the corresponding charioteer G, whose
limit towards F is determined by the dowel-holes to the left on Slab 9, would stand a good deal nearer
to the centre, and that in spite of Sauer's statement that Poseidon and ' the figures about him '
stood about 33 centimetres nearer the centre ; 2 at any rate, Poseidon stood somewhat nearer
to the centre than Athena, as is shown also by Carrey's drawing : but the teams of horses seem,
according to the indications given by Sauer, to have stood at equal distances from the centre, and
this was certainly the case with the drivers also. Tbe base of O therefore ended in the last quarter
of Slab 17, answering to G on Slab 9 ; likewise N and H corresponded ; the 'ridges ' (Randbanke) on
the wall of the pediment which indicate the end of their plinths are equidistant from the centre.
On the right, the plinth of Q — which is somewhat narrow at the base — with P R, extended from the
end of Slab 17 as far as the centre of 18; the holes on the left in front of Slab 18 are no doubt
due to the affixing of the rock under Q, of which Carrey gives a slight indication. Only in this way
do we obtain the necessary space of one slab and three quarters for the widely outstretched figure T
with S (namely, the right half of Slab 18, Slab 19, and the first quarter of Slab 20). With Sauer's
arrangement there would remain for T only a quite insufficient space.
Finally I have a criticism to make as to the figures that make up the extant remains of the west
pediment ; there can in my opinion be no doubt that we possess the N of the west pediment in the
torso I east, usually assigned to the east side, as it is also by Sauer.3 Its agreement with Carrey's N
is so striking and so complete, extending even to the fracture of the right arm, that if they are not
actually identical we must suppose that the artist repeated the same figure, both at the east and at
the west end. It has recently been thought that a difference could be perceived — that the extant left
shoulder of the torso showed that the arm was lifted higher than it is in N.4 But even this solitary
distinction I cannot concede, for it rests on a mistaken observation — the left arm, as the cast shows,
can by no means have been raised higher than it is in Carrey's drawing. On the other hand, there
is yet a further corroboration of the identity remaining on the torso itself ; at the back, below the
buttocks, is a broken surface where a piece of drapery has fallen off, of which a fragment, with strong
large folds, is still extant. Now the place is just that where the drapery drawn by Carrey as falling
over the left arm would necessarily touch the torso. It has further been asserted 5 that there would have
been no room for the wings, though this has not been established by any attempt to restore them. The
wings0 were certainly expanded sideways, and a glance at Carrey's sketch shows us how suitably the
1 That S is male cannot be doubted any longer, according to the facsimile of Carrey's work, here rightly
understood by the Anonymous. Sauer thinks that he has found two fragments of the figure (loc. cit. p. 79 seq.) ;
of them I cannot judge without actual inspection. In /. H. S. xiii. p. 92, Cecil Smith stated that S of the west
pediment was female in Carrey and Dalton. In Carrey, as the facsimile shows, this is notoriously not the case ;
in Dalton's poor drawing the forms are vague ; neither breast nor hips however are decidedly female.
2 Sauer, loc. cit. 78. In a letter to me, however, he concedes that only Poseidon stood somewhat near the
centre. 3 For the ascription to the east opinions have been expressed notably by Petersen, Pheidias, p. 144, note 1 ;
Michaelis, Arch. Zeitg. 1882, p. 376, note 11 ; Newton, Guide to the Elgin Room ; A. H. Smith, Catal. of Sculpt.
i. 112 ; Overbeck, Arc/idol. Miszellen in the Leipziger Reuuntiatiousprogramm, 1887, p. 14 sqq. For assigning
it to the west: Brunn, Bayr. Sitzungsber. 1874, 24; Trendelenburg, Arch. Ztg. 1880, 131 ; Loschcke ¦ Dorpater
Programm, 1884,-p. 11, n. 14; Waldstein, Essays on the Art of Pheidias, p. 150^/. ; Collignon, Phidias, p. 44,
60 ; E. Sellers, Class. Review, vi. 369.
4 Michaelis, loc. cit., and recently also in the Berl. Philol. Wochenschr. 1892, 11 70.
5 Overbeck, loc. cit. . . ,
6 No doubt is possible as to their existence. Waldslein's idea (loc. cit.) that the large holes in the back were
for metal clamps to support the figure is quite inadmissible.

456 THE TEMPLES OF ATHENA ON THE AKROPOLIS
wings must have served to fill up the empty space between N and O. That the wings were lost even
before Carrey's time is certainly not astonishing, in view of the Byzantine wall against which the
figure came. Everything, then, is in favour of ascribing it to the west pediment, and nothing against
it. On the other hand, the attempt to range it at the east end meets with the most serious obstacles.
Carrey has no figure of the kind there, and Nike if represented must surely have been nearer to
Athena. Sauer has to do great violence to the torso in order to bring it into accord with the marks
on the slabs of the east pediment. He is obliged to place it wholly in profile, and to make it advance
straight inwards, towards the wall of the tympanon ! This is ugly and impossible ; the back— the
left buttock especially — is distinctly characterized by its workmanship as the back side ; the figure
therefore did not stand in profile, which moreover would be a very unfavourable view for it. If we
have won this torso for the west pediment, we must on the other hand reject altogether from the
Parthenon the ' Venice fragment ' x which Waldstein attributed to this same pediment. Its style
and manner of execution show it to be a work of a far later period, as I have repeatedly convinced
myself in looking at the original.
Now that we have cleared the ground by this critical investigation, we may proceed to the
discussion of the doubtful points of interpretation, where almost everything is a matter of controversy.
Even with regard to the significance of the central group and of the whole action there is a divergence
of opinion.2 Some suppose that the two deities are just in the act of producing the tokens of their power,
Olive-tree and Salt Spring, in rivalry, before the eyes of the assembled judges who are to decide which
is the better. Others imagine a conflict between the two gods, concerning the Olive and Salt Spring
as tokens of possession, and the respective authenticity and priority of the tokens. Poseidon is thought
to be raising his trident to a blow, directed either against the Olive, or the snake supposed to be upon
it, or else against Athena herself. Hermes and Iris are approaching, to announce that the Athenians
are to judge,3 or that according to the decree of Zeus and the gods the rivals are to come to terms.4
None of these conceptions however seems to me to be quite correct. They all assume too much,
and see things that, if not impossible to represent, are at least certainly not represented here. Un
doubtedly that view is correct which, in accordance with the older tradition, regards Olive and Salt
Spring not as ' competitive products ' but as ' symbols of taking possession.' 5 But of a ' wild annihilat
ing' combat between the two deities0 I can perceive nothing. If that were intended, they would at
least necessarily be rushing to the encounter. The supposition that Poseidon's trident is being aimed
at the Olive, or at the snake conjectured to be upon It, or at Athena, has become untenable, since
Sauer has shown that the Olive stood exactly in the centre of the pediment behind the deities, and
that the arrangement of the deities given by Carrey is correct, so that the legs of the two crossed. It
would, moreover, as it seems to me, be inconceivably harsh to conceive Poseidon thrusting directly at
Athena. And as a general consideration we must remember that the Parthenon was certainly intended
as a successor to the old temple of Polias, where Polias was worshipped in peaceful union with
Poseidon-Erechtheus, and the tokens produced by the two deities were held equally holy. And finally
that conception leads to impossible conclusions. If a wild combat, threatening to Athena, was repre
sented, we wish to see an indication of its outcome also. Now it is asserted that this is given by the
presence of the messengers of the gods, Hermes and Iris. But this was evidently not so. If it had
been, those messengers would necessarily have been characterized as principal personages, and would
be interfering directly to compose the quarrel. But, on the contrary, Hermes and Iris are clearly
marked as secondary figures by their positions behind the chariots and horses. And what business
has this Hermes to be turning his head round and gossiping, when his duty is to prevent with all
speed the harm that must ensue when Poseidon carries out the thrust with his trident ?
We shall get on better by keeping purely to the artistic representation, and by combining this with
the tradition of the earlier authorities concerning the main import of the myth.
The two deities Athena and Poseidon, enamoured of the small spot of ground, the rocky
Akropolis of the Athenians, have both come thither from Olympos, in their chariots (as is seemly for
great deities).7 Each chariot is driven by a female personage. The charioteer of Poseidon is certainly
1 Arch. Ztg. 1880, PI. 7 ; Waldstein, Essays on the Art of Pheidias, PI. v. p. 121 ; Collignon, Phidias, p. 62.
2 Cf. especially Michaelis, Parthenon, 179 sqq. ; Petersen, Pheidias, 1$, sqq.; Brunn, Bayr. Sitzungsber. 1876,
i. 477 sqq. ; Robert, in Hermes, 1881, 60, and Ath. Mitth. vii. (1882), ^9 sqq. ; Preller-Robert, Gr. Mythol. i.
203; Petersen, in Wiener Studien, v. 42 sqq. ; Hermes, 1882, 124 sqq. ; Loschcke, Dorpater Programm, 1884,
I sqq. ; E. A. Gardner,/. H. S. iii. 244 sqq. ; Friederichs-Wolters, Gipsabg. p. 259 sqq.
3 So Robert, loc. cit. 4 So Loschcke, loc. cit. 5 As Robert (loc. cit.) has shown.
6 Loschcke, loc. cit. ; similarly Robert and Wolters, loc. cit.
• Similarly Themis and Ens have arrived on Ida together from the two sides in their chariots, on the
beautiful vase of the Judgment of Paris, Stephani, Compte Rendu, 1861, PI. 3.

THE WESTERN PEDIMENT 457
not his spouse, the great goddess Amphitrite, but some ministrant Nereid ; for Athena the service is
probably performed by her attendant Nike. Each god has further, as befits their dignity, brought
with them a iropiriis as escort.1 Athena has Hermes, a god friendly to her — worshipped with her in
one cella on the Akropolis ; Poseidon has Iris, who is closely connected with him.2 On the rock of
the Citadel the two gods have met together, both have taken possession, each by a token of power —
Athena by the Olive, Poseidon by the Salt Spring, which was indicated on the right, extending as far
as his chariot, under which Carrey saw a dolphin as a symbol of the salt water. The arrival of the
two deities on the same spot, their collision with one another, both making the same claims — this and
nothing else was represented m the clearest and most striking wise. Like two balls that collide, the
two recoil from each other, while the intersection of their legs makes it clear that they are laying
claim to the same spot. The movement is essentially the same in the two, but Poseidon according to
his nature is wilder, more violent in bearing, Athena more dignified.3 What happened further, after
the meeting of the deities and the setting up of their tokens, the artist does not say, and did not need
to say, since every one knew. The Athenians determined to found a holy place for the two deities
on the Citadel which they had honoured by their presence, and to reverence the tokens of both, but
to worship Athena as their real patron goddess. The presentation of the myth in the form of a
trial, with judges, judicial proceedings, and decision, belongs to a later period ; at any rate, no trace
of it is to be perceived in the pediment. The figures behind the chariots are assuredly not an assem
blage of judges : they are, in fact, mostly women or children.
Now as to the names of these figures the greatest uncertainty prevails ; all the manifold attempts
to assign names are, as their authors must surely admit, only a blind groping in the dark. I believe,
however, that all means for attaining surer results have by no means been yet exhausted. To tear up
by the roots the weeds of old prejudices must, however, be the first part of our task. And the most
deeply rooted and perverse of these prejudices is the notion that the figures in the angles represent
river gods and fountain nymphs.4 It is a wholly baseless explanation, that collapses into nothing as
soon as it is critically examined. Such an examination was first made by Walz in an excellent
though too little known essay.5 The only reason worth mentioning which led to the interpretation,
the tradition handed down by Pausanias, that the angle figures of the east pediment at Olympia
represented river gods, is untenable, the tradition in question being demonstrably wrong.0 Whoever
may have been represented by those recumbent youths at Olympia, they could have been regarded as
river gods only by later times, when art swarmed with personifications of nature, such as
recumbent figures of rivers and the like. Apart from the analogy of this interpretation given by
Pausanias at Olympia (which moreover is applicable only to the male figure on the left, not to
the female figure of the other angle), all other arguments adduced for the ' river gods ' on the Par
thenon are absolutely worthless. Recourse has even been had to the favourite metaphor of language
which describes the forms of the man in the left angle as ' flowing,' by way of support for the inter
pretation of him as a ' stream ' ; and in the ' undulating ' lines of his mantle at the back a hasty fancy
has seen an indication of actual waves.7 Finally, it was also believed that the man was lying on a
1 stream bank,' because under the upper part of his body a rock-basis is given, while his legs rest on
1 This is customary in the Homeric epos, as on the vases of the earlier period ; the usual leader of the gods'
chariot-teams is of course Hermes. We should mention, too, the leaders who are advancing at a quick pace
beside the teams in the frieze of the Parthenon : these are akin to those of the pediment in the motives also.
2 Cf. Trendelenburg, Arch. Ztg. 1880, p. 132 seq. For the parallelism of Iris and Hermes, cf. also the vase
given in J. H. S. xi. PI. 11 (Decking of Pandora) ; it belongs to the period circa 450—440.
3 It is a mistake to consider, as is generally done, that Poseidon is drawing back, Athena triumphing. Their
movement is quite similar, only diversely coloured in accordance with their diverse characters. The well-known
vase at St. Petersburg must be left out of consideration in interpreting the pediment ; it is clearly inspired by these
sculptures, but displays quite different motives. Cf. especially Brunn and E. A. Gardner, loc. cit. Still less
allied is the representation on a fragmentary Pyxis on the Akropolis (unpublished). Athena and probably also
Poseidon are at rest ; between them are the salt ' sea ' (with fish) and the olive. Dionysos is approaching tumultu-
ously behind Poseidon. The lamp with the epis of Athena and Poseidon published by Cecil Smith (/. H. S.
xiii. p. 93) throws no light on the question, as the group only bears a distant resemblance to the group of the
west pediment. The only interesting point is that there is certainly no struggle between the two divinities.
4 With reference to their names there is much variation ; the favourite names are kephissos, Ilissos, and
Kalirrhoe ; recently Eridanos also was proposed (Robert, Nekyia des Polygnot, p. 71 i and so already before him
Weber, 1821). The question as to how these various streams could have been made recognizable to the ancient
spectator is touched by none of the interpreters.
6 Walz, Eckfiguren am Ostgiebel des ol. Zeustempels u. am Westgieb. des Parth., Programm des Evang. Theol.
Seminars zu Maulbronn, 1887 (Progr. No. 543). ,,,,,, o a 111 Ts^,
6 Cf. my discussion and reply to the objections of others in the Jahrb. d. Inst. 1891, p. 87 ; Arch. Anz. IS91,
p. 94; and Berl. Philol. Wocheus. 1892, p. 1316.
7 Cf. Michaelis, Parthenon, p. 193 i Petersen, Phidias, p. 195 i Newton, Guide, p. 28 ; A. H. Smith,
Catal, i. 120. . 3 N

458 THE TEMPLES OF ATHENA ON THE AKROPOLIS
the floor of the pediment.1 But how is the floor of the pediment to be made to suggest water ? How
the artist did represent water is shown by the figure of Helios rising from Okeanos in the correspond
ing angle of the east pediment. Moreover, according to this theory, the kneeling girl C would be
holding her left knee, which is on rock, upon a ' stream bank,' and her right in the water, in which
likewise the left knee of B would be lying, since these two knees, cut off below, rest directly on the
floor of the pediment like that leg of A. But in fact, when we look carefully, it becomes quite clear
that below the figures towards the left angle the rocky ground is in every case given where the scanty
space allowed, and in every case omitted where its introduction would have required that the figure
should be cut lower. It was the effort of the artist in the restricted angles of the pediment to make
the figures as large as was in any way possible. And to any one looking from below, as the floor of
the pediment was not at all visible, this method of giving the rock produced the same effect as though
the figures were lying, kneeling, or sitting entirely upon the rock.
And this rocky surface can be no other than that of the Akropolis, on which the scene in the
centre is taking place. It is quite inadmissible to separate off the figures in the angles, who are
manifesting a lively interest in that scene, and to transport them in imagination to some other spot.
They also are necessarily beings of a like kind with the rest assembled here. And these must be
persons who are at home on the Akropolis, where are no rivers ; they must be the original inhabitants
of the Citadel rock, who witnessed that visit of the two deities, and who first instituted their worship.
It is they who possess the warmest and most immediate interest in the scene, and have the first claim
to be represented here.
In opposition to this simple and natural postulate, various Attic deities have been sought out, now
one now another, for the most part in arbitrary fashion ; some have even supposed the artist's
archaeological and topographical views to be symbolized. The only consistent explanation has been
that of Brunn, who, making the river gods the basis of his theory, regarded the remaining
figures as nothing but a living map of Attika. That this was impossible was soon perceived ; but it
was only the consistent development of the mistaken original idea that there were river gods in the
angles representing a part of the country of Attika. In the artistic products of the fifth century there
are no instances of any figures serving merely as indications of locality.2 Such personifications were
foreign alike to Polygnotan painting and to the extant vases.3 All analogies demand that on the
Parthenon also we should have only personages of flesh and blood, with purely human relations to
the scenes represented.
A safer starting-point is given by the snake on the powerful coils of which the man B is seated,
and on which he supports himself with the left hand. When the original inhabitants of the Akropolis
—the witnesses of the e'pts of the gods — are in question, this man, as has been long recognized, can be
no other than Kekrops, the autochthonous snake-man, the ApaKovrihrjs, after whom the Citadel itself
was supposed to have been in earliest times called Ke<poiria. It is true that he was properly conceived
as ending below in the form of a snake, and was generally represented so, but this old-fashioned
representation would not meet the artistic demands of a Pheidias ; the artist would think it a sufficient
substitute to indicate the inner union of hero and of snake by representing Kekrops merely resting
on his snake. Moreover, on a vase of the more severe period Kekrops is represented in purely
human form, without the snake at all.4
Kekrops is appropriate too on Athena's side, since he stands in the closest connexion with her.
It is customary for him to be present even at the intimate scene of Athena's receiving the infant
Erichthonios from the hands of Ge. And his daughters Aglauros, Herse, and Pandrosos are, as is
well known, the closest companions of Athena; in fact, Aglauros and Pandrosos are actually
surnames of Athena. That the three girls who come next after B can be none other than these three
daughters, the irapdivoi 'Ay\avpi8es, has long been perceived (first by Leake, 1821). They are all
pressing towards their father ; the last, probably the youngest, is kneeling by him, and holds him in
her embrace. The Ephebe who is with them, and takes part in the same movement towards
the left,5 must be their brother, Erysichthon ; in him the relations of Athens to Delos were
1 This has been quite recently brought forward again as an argument for the interpretation of them as river
gods, .by Michaelis (Berl. Philol. Wochenschr. 1892, col. 11 70).
,„. ' 9f; *e Berf Philol. Wochenschr. 1892, col. 1316.— The 'Hellas' and 'Salamis' of Panainos, the
Eleusis, Thebe,' 'Asia' of the vases, are of course not at all mere designations of place, as the supposed
figures in the pediment of the Parthenon would be.
3 Cf. especially Gerber in Fleckeisen's Jahrb. Supp. xiii. 277 sqq. He has, as is well known, also proved
(p. 300 sqq. ) that personifications of mountains are unknown in Greek poetry and art before the Roman period.
4 Munich, 376, Rape of Oreithyia.
0 I conjecture that he is to be conceived as sitting down ; I cannot explain on any other supposition the
rawing up of the right leg so high. According to the theory that the figure had fallen in Carrey's time, and

ERECHTHEUS AND KEKROPS

459

embodied,1 since he led the first festal embassy there, dedicated a carved image to Apollo, and
brought back a sacred image of Eileithyia. Mythology made him die in his father's lifetime
before he had begotten offspring, and so he is fitly presented as an Ephebe. The children
of Kekrops are terrified by the violent encounter of the two great gods, and press timidly towards
their father.
We have seen that the arrangement of the two halves of the pediment was completely symmetrical.
Over on the right side also there were, as Carrey shows, three women with an Ephebe ; the last of the
women is turning, just like D on the other side, only with a still more violent action, away from the
centre towards the angle of the pediment. The next figure, which was no longer there in Carrey's
time, was, as we have shown, a seated man answering to B ; the maiden above mentioned leant in
terror against him. The groups, which corresponded artistically with such completeness, must have
corresponded in meaning also. Here too we must recognize a father with his three daughters. He too
must have belonged, like Kekrops, to the original inhabitants of the Citadel.
From these premises the interpretation follows with simple necessity : we cannot any longer go
astray, since there is only one possibility : we must recognize in these figures Erechtheus with his
daughters. He alone forms a complete counterpart to Kekrops : he is as closely united to Poseidon
as the other is to Athena. He was the second great original inhabitant, the competitor and rival
of Kekrops in the old Attic religion.2 He was conceived as living on in the heart of the Citadel
rock. Erechtheus and Kekrops share together in the west half of the old temple of Athena, and of its
successor the Erechtheion,3 although Erechtheus took precedence in ritual, owing to his identification
with Poseidon. Mythology, however, distinguished the hero from the god. As a hero and one of the
oldest inhabitants of the land, he appears united with Kekrops and his daughters at the birth of
Erichthonios, on a beautiful Attic cylix of the Pheidian epoch ; 4 and another somewhat older vase
painting5 shows Kekrops and Erechtheus again associated as the two ancient kings of the land — a
representation of special importance for us, since those two are there pictured together with their
daughters, who appear as playmates to each other. These monuments show that the artists of the
period regarded Kekrops and Erechtheus as contemporaries, both of whom are alike interested
witnesses of the first acts of the gods on Attic soil.
The daughters of Erechtheus also form of course an exact counterpart to those of Kekrops.
According to the older tradition of which we have evidence for exactly the period of the pediments
of the Parthenon, they likewise were three, a £evyos rpnrdpBevov, nymphs of precisely similar nature to
the Agraulides.6 By the same meadows and caves on the north slope of the Citadel where the
daughters of Kekrops dance to the flute of Pan, one of the daughters of Erechtheus had yielded to
Apollo.7 had previously been upright (Petersen, Phidias, p. 182, No. 1), that leg would be quite inexplicable. The move
ment of the Ephebe is quite similar to that of the girl D beside him.
1 Cf. particularly Milchhdfer, Der Attische Apollon, p. 11.
2 Cf. Immisch in Roscher's Lex. ii. 1023. 3 Cf. supra, p. 435. 4 Mon. d. Inst. x. 39.
5 Munich, 376; a contemporary replica, which only differs in not having all the inscriptions, Berlin, 2165,
Rape of Oreithyia, the Daughter of Erechtheus. On the copy at Munich, Kekrops, Erechtheus, Herse,
Pandrosos, Aglauros are designated by name. The fifth maiden without a name (the 'indistinct traces of letters'
of which jahn speaks here I have made out on the original to be a koX&s) must be a second daughter of Erech
theus, sister of Oreithyia.
6 (evy. rpiir. Eurip. frg. 359. Cf. frg. 362, 1. 36. The tradition that there were six, in the Atthis of
Phanodemos (Suidas, irapBevoi — Photius, p. 397, 7), is obviously later ; the comic poet Phrynichos (ibid.), who
mentioned the irapBevoi, probably did not state their number. In Apollodorus (3, 15, 1 and 3, 15, 8) we find two
traditions, both later, and without value for us ; both give four maidens ; one makes them daughters of a Hyakin-
thos (because they were called Hyakinthides). The same two parallel versions occur in Hyginus, Fab. 238 ; their
cultus-name was simply irapBevoi, Suidas (cf. Eurip. Ion. 278) or 'TaicivBiSes, Lykourgos apud Harpocr., Pseud.
Demosth. eirirdip. 27 (p. 1397). Speech of Phokion apud Diod. 17, 15, 2. There were various versions concerning
the origin of this name. According to Phanodemos, loc. cit., it was derived from a hill, Hyakinthos, on which the
maidens were slain in sacrifice. Others invented a father Hyakinthos, and he naturally had to be represented as a
Lakedaimonian who had migrated to Athens. At the same time it was necessary to invent other names than those
established for the daughters of Erechtheus. In reality the name merely designated the maidens as nymphs to
whom the hyacinth flower is sacred. In the Erechtheus of Euripides (frg. 359) they are called "TdSes : this
again is only a name typical of them as nymphs. The name Hyades was given also, as is well known, to the
nurses of Zeus and of Dionysos, the nymphs of Dodona and of Nysa. As in the case of the other Attic nymphs,
the sacrifices made to them were naturally vqipdxia. This is expressly stated by Philochoros, frg. 31 (Schol. Oed.
Col. 100, Aiovvcra re Kal 'Ep. Bvyarpdaiv, as well as to other gods, r-n^aAm lepi are offered) ; he does not however
at all assert, as is generally stated (Preller-Robert, Gr. Myth. i. 201), that the daughters of Erechtheus received
sacrifices together with Dionysos (Bdhlau, in Bonner Stud. p. 137, actually makes it into ' together on the same
altar') ; they only had in common the characteristic of receiving v-qfy&Xia, the places of their worship were of
course distinct. It may be safely assumed that the place of their worship was on the Akropolis, though this is
nowhere expressly stated.
7 Eurip. Ion. 492 sqq.

460 THE TEMPLES OF ATHENA ON THE AKROPOLIS
The manifold myths with regard to the daughters of Erechtheus enabled the artist to characterize
them more particularly. The figure that sits in full front view next behind the charioteer of Poseidon
must be the Oreithyia so renowned in legend. The story of her being carried off by Boreas was an
especial favourite in the art of the fifth century. According to the oldest tradition the rape took place
on the Akropolis itself,1 to which the girl was going as Kanephoros of Athena Polias. In the far
north, on the Sarpedonian rocks surrounded by the roar of tempests, she bore to the wind-god two
sons, Kalais and Zetes. Thus we reach, without seeking it, at once the explanation of the peculiarities
of that figure O which hitherto have been quite a riddle. On the remaining lower part of the body the
drapery appears in the most striking manner seized by the wind, which troubles its folds. The figure
sits with closed feet, as though to resist the assault of the storm. According to Carrey, she
is on a raised seat of rock, at the foot of which the waves of Poseidon's salt water cease their
flow. She has by her two boys, the bigger of whom presses close to her,2 while the smaller stands on
the rock near her left shoulder.3 The attempts hitherto made at explanation do not at all suit these
facts, and least of all that interpretation of her as Gaia Kourotrophos 4 which has recently come
into favour again. On the other hand, Oreithyia could not be more appropriately characterized than
she has been here. It is true that this could only be done by a certain anachronism, a prolepsis —
incidents being indicated which happened later than the scene represented. But before the cele
brated rape Oreithyia was just no more than a girl, with nothing to distinguish her from others. She
could be characterized only by means of an allusion to her subsequent fate. It is well known 6 that
ancient art made free use of prolepsis, especially in accessories like attributes, and such are the
children of Boreas here ; the characterizing of mythological figures by an allusion to the principal
incident of their fortunes was a specially favourite device of Polygnotan painting. Finally, if it
should be objected that the little Boreadai would necessarily have had wings, we may point in answer
to Ovid, Metam. vi. 713 sqq., where it is expressly stated that the wings only grew with their beards.
In view of Ovid's special connexion with the arts of painting and sculpture ° we may conjecture that
he had in his mind some representation of Oreithyia with her children not yet winged,7 and in fact
art necessarily represented them without wings that they might not be confused with Erotes.
The storm-possessed figure of Oreithyia with the little Boreadai, moreover, awoke in the Athenians
the pleasant memory of their special friendship with Boreas, who as their good 'son-in-law' gave
them such valiant support in the Persian crisis.8 And his home in Thrace also was in fact in the
Periklean period a land of special importance to the Athenians, and a prop of their power.
The woman next to Oreithyia, lying in profile with the Ephebe on her lap, must be the other
famous daughter of Erechtheus, Kreousa, with her son Ion.
We first of all will point out that here again the artistic motive is satisfied by our interpretation.
A youth sitting in this way on the knees of a woman must be either son, or in some relation equally
near. This is proved not only by the nature of things, but by all analogies in art. Parents take their
children on their laps — this movement, as old as the human race, is the root from which all similar
groups arise. It is true that the lover may hold his beloved upon his knees — the group occurs
repeatedly in ancient art °— but not vice versa : the wife does not take her husband on her lap.10 It must
always be the stronger and elder of the two who in this intimate fashion unites with himself the younger
and subordinate member of the pair. Thus, for instance, in the beautiful terra-cotta groups of Asia
Minor, which represent two women thus grouped together,11 the one who holds the other on her lap is
1 Akousilaos apud Schol. Horn. Od. 14, 533 (Miiller, Fragm. Hist. i. 102, 23).
2 The upper part of his body has been quite recently identified (see Cecil Smith's note in Class. Rev. 1892,
(vi.), p. 475).
3 The Boreadai are first called twins by Ovid, Metam. vi. 712.
4 First advocated by Brdndsted and Millingen, recently by Robert (Hermes, xvi. 60), Walz (Maulbr. Progr.
1887, p. 32), and Kuhnert (Roscher's Lexikon, i. 1577). Loschcke calls her Demeter Kurotrophos (Dorpater
Progr. 1884, p. 10).
5 Cf. especially Stephani in the Comptes Rendics (Sal. Reinach, Index des C.R., sub v. Prolepsis).
Cf. Ribbeck, Rdm. Dichlung, ii. 309.
7 Since Ovid visited Athens, the Parthenon itself may have been his authority. In fact, the description of
the contest between Poseidon and Athena (vi. 70 sqq.) also agrees in many points with the west pediment, and
especially the motive of Athena's calling forth the olive-tree by a thrust of her lance might have originated
merely from an interpretation of the pediment. 8 Herod, vii. 189.
8 Cf. e.g. the vase Antiqu. du Bosph. PI. 62, 2. Berlin vases, 2904.
10 Thus the motive of itself tells against Ldschcke's interpretation of the group as Melite and Herakles
(Doipatcr Progr. 1884, p. 10, ' the father sits on the knees of the mother ').— A variant of Ldschcke's interpretation
has been recently proposed by Robert (Nekyia des Polygnol, p. 81) : Eleusis with Herakles. It is at least as good
as Loschcke s, but a whole set of others just as good may be made up— a true indication that no progress is to be
made on that path.
11 Frohner, Catal. Griau, Terresc. PI. 48; Arch. Anz. 1892, p. 159.

DAUGHTERS OF ERECHTHEUS 461
characterized as the elder and more dignified. So it is also in the picture of Aristophon, which
showed Alkibiades sitting on the knees of Nemea ; 1 the beautiful young victor was marked as a nursling
of the goddess. In the frieze of the Erechtheion we find not only a woman with a boy 2 on her knee
but also one with a nearly full-grown youth,3 just as in the group on the Parthenon. Of course both
are to be similarly explained : in both cases we have a mother with a son. Since the Erechtheion
certainly dealt with the myths attaching to Erechtheus and his family,4 the agreement of one of
its groups with that on the Parthenon is specially noteworthy, as we have referred the latter like
wise to the family of Erechtheus. A small marble group at Eleusis ;> shows Kore sitting on the lap of
Demeter, who is seated as usual on her round cist. I may finally mention the relief on a beautiful
bronze mirror from Eretria, where Aphrodite has on her lap Eros, represented as almost a full-grown
boy, a peWecprjPos." And from still later times Domitian, who had himself represented sitting on the
knees of Jupiter,7 affords an argument for the correctness of our conception. Later art also knows a
group of this kind, which again serves as a corroboration— the type which represents St. Anna and
the Virgin Mary, in which the full-grown daughter sits on her mother's lap, as in that group of the
goddesses at Eleusis.
The motive seen in the Parthenon is therefore characteristic for the relation of son and mother,
and our interpretation of it as Kreousa with Ion in her lap fulfils this condition. It fits also excellently
in every other respect. The youthful Ion, the only male scion of the house of Erechtheus, is the most
complete pendant to Erysichthon on the other side, the only son of Kekrops. And the two corre
spond well in another respect also — namely, that they both have close relations to the Ionic Apollo —
Erysichthon, to the Ionic Kyklades, and their worship of Apollo on Delos ; Ion, to the Ionians, and to
Apollo in general. The graves of the two were shown and reverenced in two neighbouring denies of
the east coast — that of Ion in Potamos, that of Erysichthon in Prasiai. But with Ion were connected
yet more important memories than with the other, as we learn from the words which Euripides makes
Athena utter at the close of his Ion. To Ion was referred in Athens the first political organization on
Attic ground ; he instituted the four ancient Phylai (tribes), called after him ' Ionic' 8 His sons are
the Eponymoi of the people which dwells at the rock-citadel of Athena. Hence the Ionians went
forth to the Islands and to Asia Minor,0 and thus Ion is the representative of the ideal Attic Empire —
namely, of the dependence of all Ionians upon Athens, which was the basis of the Athenians' power.10
Thus he is a figure truly fitted by his nature for representation on the pediment destined to glorify
ancient Athens.
That Ion belongs to the house of the Erechtheidai as Kreousa's son is an Attic legend ; the
oldest witnesses to it are Euripides in his extant play, and, apparently, Sophokles in a corresponding
piece now lost.11 But the legend must possess a high antiquity, and must be as old as the introduction
of the worship of the Ionic Apollo Patroos at Athens, and the dedication of a grotto on the north
slope of the Citadel to this Apollo under the name Hypakraios.13 His place of worship is appointed
to him on the Citadel rock, by the playground of the daughters of Kekrops and Erechtheus, clearly
because he was connected with the autochthonous royal house of Attika, as lover of Kreousa, one of
those nymphs.13 Still older, however, is Ion as son of an uncertain mother and Apollo. Apollo and
1 The analogy between this and the group of the west pediment was first pointed out by Loschcke in the
Dorp. Progr. 1884, p. 8.
2 Friederichs-Wolters, Gipsabg. 815 ; Overbeck, Gesch. d. Plastik, 3rd ed. i. Fig. 80, i. A reproduction of this
group was found at Eleusis ('Etp-qp. dpx- 1890, PI. 13), together with a copy executed on the same scale of
Kekrops with the daughter leaning against him from the west pediment (ibid. PI. 12). The series of these figures
at Eleusis probably represented the families of Kekrops and Erechtheus, after the model of the sculptures of the
Akropolis : they seem to belong to what was still a very good period.
3 Friederichs-Wolters, Gipsabg. 817 ; Le Bas, Voyage Arch., Mon. Fig. PI. 16, 1 ; Schdne, Griech. Reliefs,
PI. i.— iv. No. 6.
4 Robert has made a clever attempt (Hermes, xxv. (1890) p. 431 sqq.) to reconstruct the representations : the
details are of course highly problematical.
5 In the Museum at Eleusis, unpublished : it is not even mentioned by Kern in Ath. Mitth. 1892, 125.
6 Athens, in the Arch. Society (Photograph in German Institute, Arch. Anz. 1891, p. 80, No. 81).
7 Tacit. Hist. iii. 74. Cf. Sittl, Gebdrden, p. 35.
8 Herod, v. 66; Aristot. 'ABqv. iroX. 41 ; Eurip. Ion, 1575 sqq. 'J Eurip. Ion, 1581 sqq.
w Eurip. Ion, 1584 : o oBevos riipfj xBovl SiSairiv.
11 The supposition that the two titles ' Kreousa ' and ' Ion ' given by tradition denoted the same piece is
probably correct.
12 Ion instituted the worship of Apollo Patroos: Aristot. 'AB-qv. iroXir. frg. 1 (Kaib.-Wilamow.) With
regard to Ion and Apollo Hypakraios, cf. Milchhofer, Al/ischer Apollon, p. 48. Inscriptions of Roman date
(Ath. Mitth. iii. 144 sqq. ; U. Kdhler) inform us that il was customary for the Archons and the officials under them
to set up votive offerings to Apollo Hypakraios when they retired ; perhaps precisely because this Apollo was the
father of Ion, founder of the first political institutions of Attika.
13 The assertion is often made that Euripides was the first to link Ion to Apollo and sever him from Xutithos
(so even Tdpffer, Alt. Geneal. p. 268) ; this is not only quite groundless and intrinsically improbable, but also in

462 THE TEMPLES OF ATHENA ON THE AKROPOLIS
he came as an established pair from the Tetrapolis to Athens,1 and there the two together were
connected with the Akropolis and the Erechtheidai. The Ionic Phylai, which were traced back to
Ion, existed in Asia as in Attika side by side with the worship of Apollo ;2 they belonged, with the
Ionic Apollo, to the Ionic race. But the connexion with Erechtheus through Kreousa is old Attic.
On the other hand, the myth of Xouthos son of Hellen as father of Ion betrays a later origin. Its
tendency is to represent the stock of Ionians as brother to the stocks of the Dorians and Aiolians, and
to derive them all from a common home ; it probably belongs to the circle of genealogical poetry in
which we first find the three races thus traced back to Hellen as a -single ancestor.3 The two fathers
thus attributed to Ion, Xouthos and Apollo, afforded to the tragedians a welcome subject for spinning
stories of conflicting claims, as has been done so beautifully by Euripides. The artist of the pediment,
however, started directly from the old Attic myth.5
The third daughter of Erechtheus, who is clinging in such terror to her father— now lost — is
clearly conceived as the youngest, and as the subject of the famous myth according to which she
voluntarily suffered death at her father's hands for the weal of her country.6 This too we know to
have been a story of the past on which the Athenians prided themselves.
Again, we now obtain the interpretation of the angle figures also. On either side we find a pair
consisting of a man and a woman — in the gap between A and B a woman is certainly to be supplied.
For the sake of variety their relative positions are reversed at the two sides : on the one the man is
the figure in the angle, on the other the woman. In each pair the figures are closely connected. In
the case of V W this is shown by their attitude, turning towards each other, and in the case of
A A* it is to be presumed, as A is turning himself with a very lively movement in the direction in
which was the woman now lost. Probably her body was turned so as to come close to the man,
while she also directed her eyes in astonishment and fear towards the centre. The couple on the
right, by Erechtheus, is to be interpreted as Boutes and his wife, an explanation which is simple and,
after what we have previously said, almost inevitable. Boutes was, as we have seen (p. 435), one of
the inhabitants of the Erechtheion and the old temple ; he held the cella next to Poseidon-Erechtheus,
and had a priest of his own. He is one of the most important of the ancient heroes of the Citadel,
ranking beside Kekrops and Erechtheus.0 He is very closely connected with Poseidon and
Erechtheus, and at bottom is of the same essence as they. He is the ancestral hero of the
Eteoboutadai, in whose family the priesthoods of Poseidon-Erechtheus and of Athena-Polias were
hereditary. According to the myths of their clan, he was — at least in the fourth century— traced up
to Erechtheus ; but the older myth,7 which no doubt was followed by the artist of the pediment, made
him a son of Poseidon himself. The later myth gave him a place in the old royal house of Attika,
as brother of Erechtheus and son of Pandion. His wife was a somewhat vague figure in the myth :
tradition names her Chthonia, and traces her back either to Erechtheus or to Boreas. It is doubtful
whether the artist of the pediment had any more definite ideas with regard to her personality.
The corresponding pair on the left must be Bouzyges with his wife. For the Bouzygai, with their
great ancestor Bouzyges, correspond in antiquity and importance to the Eteoboutadai with their hero
Boutes.8 The two priestly clans together represent the oldest forms of worship on the Citadel rock ;
and as Boutes is closely connected with Poseidon-Erechtheus, so is Bouzyges with Athena and
Kekrops The Athena, planter of the Olive who sheds her glory on the pediment, is the goddess of
agriculture, and the earth-born Kekrops is the king of Athens in that most ancient time when she was
still purely agricultural. But to them belongs, as an addition necessary to their completeness, the
contradiction to the few facts that we know. Ed. Meyer (Forschung z. alter Gesch. p. 142) is in the main correct
in his judgment ; but he should have laid more stress on the ancient connexion of Ion with Apollo Patroos and
Hypakraios, with his grotto in the Citadel rock, and the daughter of Erechtheus established there. Meyer is
doubtless right in making Ion and Xouthos come from outside ; but Xouthos-Ion is later than Apollo-Kreousa-Ion,
who are connected with the introduction of the cultus of Apollo Patroos into Athens ; the cultus may however
have been introduced at a comparatively late date.
1 With regard to the Tetrapolis as the proper home of Ion, cf. Kirchner, Attica el Peloponnesiaca,
Greifswalder Dissert. 1890, p. 16. 2 Cf. Busolt, Griech. Gesch. i. 392, n. 6.
3 Hesiod apud Tzetzes ad Lycophr. 284 (frg. 27, Rzach).
4 I have tried to show in Samm. Sabouroff, i. Vasen, Introd. p. 14, that the myth of Ion, Kreousa, and Apollo is
represented also on Attic vases of later style (which however may well belong to the end of the fifth century) ; but
by that time they are probably dependent on Sophokles and Euripides.
5 Her name varies in the tradition ; there is also a version that more than one daughter was sacrificed, or
that her sisters voluntarily followed her in her death. In the Erechtheus of Euripides only one daughter was
sacrificed (Lykourgos against Leokr. 98) ; in the Ion, on the contrary (277 seq.), the poet represents the daughters
of Erechtheus as sacrificed ; only Kreousa escaped. The myth fluctuated therefore even in the time of Euripides.
" Cf. Tdpffer, Alt. Geneal. 113 sqq.
7 Hesiod, frg. 124, Rzach. Cf. Tdpffer, loc. cit. 114 ; also Bdhlau, in Bonner Studien, p. 126 sqq.
b Cf- especially Tdpffer, Alt. Gen. p. 136 sqq.

THE EASTERN PEDIMENT 463
hero Bouzyges,1 the first in Kekropian Athens to lay the yoke upon the oxen and to furrow with the
plough the fields which spread around the rock of the Citadel. His plough was shown on the Akropolis,
which had been his abode. In the ordinary mythology, it is true, he was thrust into the background
by the Eleusinian Triptolemos. But in the worship of the state this was not so : there Bouzyges
alone was regarded as the first plougher, and the priestly family of the Bouzygai maintained itself as
high in reverent estimation as that of the Eteoboutadai. Every year on the sacred plough-land at the
foot of the Citadel the solemn ploughing, the lepbs aporos, was performed by the priests, and the
produce of this field was dedicated to Athena,2 who thereby appears as the supreme patroness of agri
culture. In other respects also Bouzyges is specially connected with Athena : it was he, according to the
old myth, who received the true Palladion from Demophon.3 Moreover, the Bouzygai held the priest
hood of Zeus ev naXXafii'w, in whose temple Athena was also worshipped. A statue of Polias, dedicated,
according to the inscription, by one of the Bouzygai,4 seems to have stood in this very place.
Tradition does not tell us the name of Bouzyges's wife, whom we conjecture to have been in
the gap. But we do know that the conjugal relation formed a necessary part of the conception of
Bouzyges, for the Bouzygai likewise held the priesthood of Zeus Teleios, the tutelary god of wedded
life. In fact, the ancient view, as is well known, treated marriage as the social outcome of the con
dition of human life regulated by agriculture ; and Bouzyges himself seems to have had the credit for
introducing wedded life, or, as the old symbolic expression ran, the yaprjXtos aporos.5
At the time of the sacred ploughing, the successors of Bouzyges every year proclaimed at the foot
of the Citadel the curse against those who violated the most general laws of morality, the injunctions
comprised by the German poet in the simple sentence, ' Man must be noble, helpful, and good.' Thus
the hero Bouzyges is at the same time the representative of all true humanity and goodness. This
the artist expressed in the special and singular beauty of his form, which is neither highly developed
by athletic exercise nor powerful in divinity, but gentle and of that true human beauty which makes
the Bouzyges of the pediment, even in his present mutilated state, dear to us above all the other
remains of ihe Parthenon.
We have now finished our interpretation, which, when once we had set the problem before
ourselves correctly, followed point by point almost as inevitably as in a mathematical proof ; and
this is certainly the best warrant for its correctness. It has now brought us to this result : we see
grouped around the contending deities on the Citadel none but persons who in all the phases of their
lives are connected with this Citadel rock, and who, above all, must have been intelligible to every
Athenian, since they expressed, not far-fetched wisdom, but only that which was known to all and
believed by all. Now for the first time we understand how Pheidias, besides all his other qualities,
was an artist for the people, since he was able to give form to that which was in the heart of
every Athenian.
The east pediment is far easier to explain than the west.
Sauer's excellent investigations (Fig. 184)6 have brought us certainty with regard to the disposition
ot the central group.- What formerly could only be conjectured may now be asserted as sure. The relief
at Madrid giving the birth of Athena7 agrees with the composition of the east pediment, as preserved
by the traces on the floor of the pediment : Zeus enthroned, in profile towards the right ; Athena,
in full-grown form without any external connexion, standing in front of him or advancing. The style
of the Madrid relief also points to a Pheidian original of the time of the pediment. The motive
of Zeus resembles that of the statue by Pheidias at Olympia ; the cloak in the arrangement and
treatment of the folds, the form of the throne, and equally the type of Athena are quite in the style
of the frieze and pediments of the Parthenon. Thus there can be no further doubt that the relief
at Madrid is really dependent on the east pediment. The Nike also, who hovers between the two
deities, and fills the gap with her significant figure, must be derived from the original, where she
would have been made fast to the background of the pediment, now lost.8
It is further to be inferred that the figure of Prometheus or Hephaistos with the axe, which
stands in the relief behind the throne of Zeus, is likewise taken from the pediment ; it is, in fact, quite
in the style of the Parthenon frieze (cf. e.g. Michaelis, North, 44, 58). Since the process of birth is now
fully complete, and Athena is in the act of moving rapidly away from her father, the personage who
1 Cf. also Preller, Demeter und Persephone, p. 290. 2 Tdpffer, loc. cit. 137, n. I.
3 Polyaen, i. 5. Cf. Tdpffer, loc. cit. 146. 4 C. I. A. iii. 71. Cf. Tdpffer, loc. cit. 145 seq.
6 Tdpffer, loc. cit. 147. 6 Ath. Mitth. xvi. 68 sqq., 81 sqq.
7 Schneider, Geburt d. Athena, PI. I; Wiener Vorlegebl. viii. PI. II.
8 An attempt to reconstruct the central group with Nike has recently been published by Jan Six (Jahrb. d.
Inst. 1894, p. 84).

Fig. 184.— Eastern pediment of the Parthenon, showing the traces on the floor.
(From B. Sauer's drawing, Ath. Mitth. xvi. plate iii.)

THE EASTERN PEDIMENT 465
has been assisting in the birth cannot possibly have stood with his weapon still uplifted for the blow.
Cn the other hand, the dealer of that blow must have stood immediately behind Zeus. We require
in the pediment, therefore, exactly what the relief gives us. But this makes it impossible to regard
the torso H in the right half of the pediment as Hephaistos. And in fact this common hypothesis
shows great want of reflexion, since it is quite impossible that Hephaistos with uplifted axe should
stand behind Athena instead of behind Zeus. We must rather accept the view already put forward
by Friederichs ' and Michaelis,2 that H is simply a god who draws back in astonishment before
Athena, lifting up his right arm3 in a motive that is closely connected with the Marsyas of
Myron, and also with the Poseidon of the west pediment.4 The name of the god we may leave
undetermined. Next to the four central figures comes on each side, as Sauer has shown, a seated personage
seen in profile, and of course turned towards the centre. We may conjecture them to be Hera and
Poseidon.6 Behind these sitting figures come on each side two who were standing (Nos. 4 and 5 of
either side, counting from the middle) ; of these we know nothing.0 Apollo, Artemis, Ares suggest
themselves. Then follows on the left as No. 6 the female figure moving rapidly towards the left
which is commonly called Iris. The pendant to her is lost. The name of Iris for this hastening
maiden has recently been abandoned, and rightly.7 As she has no wings, Iris is impossible.
But in itself the idea that this maiden is hastening to carry a message from Olympos — the
announcement of Athena's birth — is a mistaken one. The whole space enclosed within the border
of the pediment is in fact Olympos : no one is hastening forth from it. But the gods were assuredly
in great astonishment and excitement, as is shown by the torso H. As he is starting back from
the sudden radiant apparition in the centre, so the maiden is fleeing away in fear, although at the
same time, spell-bound by the apparition, she turns her head back towards it. The forms are so very
youthful and immature that Brunn's interpretation of the figure as Hebe no doubt hits the mark.
The presence near at hand of Hera on her throne, whom we suppose to have been on this side,
would agree well with this hypothesis.
The two groups in the angles, which follow next, are completely preserved. The name of Helios
for the man in the left angle is however the only one wholly beyond doubt. The goddess corre
sponding to him is now ascertained to have been driving away on a quadriga, and therefore does not
display the ' rider type' hitherto known for Selene in Pheidian monuments.8 It is therefore more
probable that the figure should be called Nyx,9 whom some beautiful lines from the Andromeda of
Euripides depict for us driving her team of horses through the aether of lofty Olympos.10
On Olympos, Day and Night alike move on their way ; they are among the most natural
and necessary witnesses and participators in all that happens in heaven. Therefore Helios and
Nyx are not mere indications of place, not mere accessories in the pediment ; they belong to the
persons most closely concerned. Even Nyx, who must withdraw before the rising Helios, turns
back her head to look at the action in the centre. Only figures that have their place on Olympos
can come next to Helios and Nyx, and most probably divinities bearing a certain relation to those
two, especially since they are turning towards them and away from the middle. Nor must we
conjecture them to be great deities, since these would be taking a stronger interest in the central event
-the birth of a new comrade, a goddess equal to themselves. The group of the great deities was
1 Bausteine, p. 143. The passage has not really been improved by the changes introduced by Wolters, p. 253.
PttYth&tioti p 1 7 K
3 Only the right arm was lifted up, not both as is generally stated. The difference in the marking of the
muscles on back and breast between the two sides shows this indubitably.
4 The affinity of the Poseidon to Myron's Marsyas has already been pointed out by E. A. Gardner in the
J' **'* Cf.'Brunn, Bayr. Sitzungsber. 1874, P- 19 seq. There is a fragment of a head that Sauer, (Festschrift fiir
Overbeck, p. 74 sqq. ) t hinks may be ascribed to the sitting figure on the right : she would then be Hera. Between
her and Athena he supposes a goddess bearing a torch (on account of a fragment of a hand) : if this is so, the
torso H would be most suitably placed farther to the right behind Hera. ,,-.,,
8 That the torso I, which Sauer also sets on the right in the fifth place, does not belong here but in the west
^^(K^ffiS netas Hebe, Murray as Eileithyia, Wolters left her name undetermined. Cf.
A' MTott types' of f^e a's ^already enumerated by Furtwangler, Samm Sabouroff, text to IM. 63>
¦and bv Cecil Smth J H S. ix. p. I seq. , must be added a charming red-figured leky hos -from U^dhia in Boeotia,
now in the Possession of. Mr. George Macmillan. Selene riding* represented quite ra the motive of the Selene on
irweU*no^to«n«"se ; shAoIds a two^ronged wand ; above the wand a single star. Height 6J m.-E.S.]
9 As is riahtlv observed by E. Sellers in the Classical Review, vi. 370.
iu a ¦ f i mVZ rati (Eurip fre 1 14)- Cf. Robert in Hermes, xix. 467 ; Samm. Sabouroff, 11. Supp. p. 5.
'« Anstoph. Thes* . 1065 (Eunp _irg. HI- ^ ¦ as drivj Qn a ^^ (cf_ Samm_ Sabo!lrof
Since however the poetry of the earliei penoct lepiesems oc . , °- th • d : ft 11 th "
i on PI €X\ the decision is not certain : it is, however, of little importance, since the idea is attei an tne same.
30

466 THE TEMPLES OF ATHENA ON THE AKROPOLIS
clearly bounded on the left by the terrified Hebe, on the right by the pendant to her, now lost. The
cosmic conception, made clear by the figures of Helios and Nyx, excludes deities of merely local
Attic significance, limited to a definite spot.1 Athena is not being born for Attika alone, but for the
whole world, far as the rays of Helios pierce and the veil of night obscures.2
Beyond this, the arrangement of the figures in question gives us yet another definite basis of
investigation. As was already shown by Carrey's drawing, and as is confirmed by Sauer,3 the three
women on the right formed a closely united group : the first leans on the lap of the second, and her
right arm appeared supported on the lap of the third.4 We have to keep in view in our interpretation
the certain fact that we have here a triad of goddesses.6 It is not so on the left, since there we find a
pair of goddesses on thrones, and by them a youth on a rock, who is turned toward Helios, and is not
in any specially close connexion with the two figures.
Those three goddesses turned towards the darkness of night can be no others than the Moirai— and
in fact they have been so explained by most of the earlier interpreters, with Visconti at their head. They
are doubtless those three sisters, who could not possibly be absent at a birth, least of all on Olympos.
The goddesses of fate were in fact at all times conceived as present at births, which is the reason that
in Pindar Eileithyia appears in company with the Moirai. They are the daughters of Night in Hesiod,6
whom Pheidias has followed elsewhere also (as is shown by the birth of Aphrodite and that of Pandora,
in the reliefs on the pedestals of Zeus and the Parthenon) ; this is the reason that they are here turned
towards Nyx : eidiKevoi Kovpai Nuktos, they were called likewise by an unknown poet of the fifth
century.7 They are goddesses of the widest cosmic importance, just such as we should wish to find
in this position. They weave and spin the world's future ; already in the Homeric poems they
are the K\5>8es, the spinning sisters. As Sauer has observed,8 the tranquil recumbent figure M holds
in her raised left hand a staff-like attribute, the end of which was loosely placed among the drapery of
the upper part of the leg. This agrees excellently with the supposition that she held a distaff in the
left hand, while the right drew the thread ; the others would have been without attributes, since the
shears and the globe belong to an allegorizing fashion of much later times. In the old days all three
were spinners only.
But besides their widespread importance, they have also a specially close connexion with the
two principal personages of the scene, — with Zeus, who was worshipped as Motpayerrjs, and with the
Athena of the Akropolis. An old decree concerning the worship of Polias on the Citadel shows, as it
seems, that the Praxiergidai— whose duty it was to put on and take off the Peplos of Polias— had, in
connexion with this, to pay honour to the Moirai and to their head, Zeus Moiragetes.0 If any one
finds a difficulty in the fact that the spinning sisters of heaven, the daughters of Night, are here
represented as so beautiful, he is certainly in error. It should be remembered that in Athens
Aphrodite Ourania was held to be the eldest of the Moirai, which shows that they were not conceived
as ugly. Pheidias has represented them as eiaXevot, the name given them by the poet mentioned
above. And moreover the motive which so strongly suggests Aphrodite, the chiton slipping down
from one shoulder (on Fig. M), is known as occurring several times in representations of the Moirai,
and also in a beautiful statue of Dike, a goddess nearly allied to them.10
Finally, we have further corroboration in the fact that the Madrid relief likewise gives us the
three Moirai present at Athena's birth. It is true that they have nothing in common with the figures
of the pediment ; types of much later date have been chosen here. But this is easily explicable by
the fact that the groups disposed so as to suit the angles of a pediment were not available for the
artist of the relief on a circular pedestal ; he adhered in subject only to the east pediment, and took
the motives from elsewhere.
The Moirai of the right side help us to explain also the two women of the left half: they must
be the Horai, and they too have been already recognized by earlier writers.11 These are the most
1 Therefore the 'Dew Sisters,' the daughters of Kekrops, are impossible : they would be as unsuitable here
as they are necessary in the west pediment.
2 Cf. also Brunn, Bayr. Sitzungsber. 1874, p. 14.
3 loc. cit. 83.
4 So represented by Carrey.
5 This excludes Petersen's interpretation of them as Hestia, Peitho, Aphrodite.
6 Theog. 217. The Charites have recently also been suggested (Collignon, Phidias, p. 46), and in many
respects would be suitable ; but they are excluded by the fact that they would have no connexion with a scene of
birth, or with Nyx.
7 Bergk, Poetae Melici, frg. adesp. 140.
8 loc. cit. 83.
9 C. I. A. i. 93. In line 12 it seems to me clear that Moi]pais, Ad Moipayer-q... should be restored.
10 Milchhdfer, in the Jahrb. d. Inst. vii. (1892), p. 206 seq.
11 First by Brdndsted, Millingen, Lloyd ; later notably by Brunn.

MOIRAI AND HORAI 467
complete counterpart to the Moirai, very closely akin to them, and yet on the other hand sufficiently
distinct ; from old times poets and artists had been fond of uniting and contrasting the two groups.1
The Horai also are pre-eminently appropriate at a birth, since they bring on the due time for the
'accomplishment of the birth, and like the Moirai they were conceived as protecting births.2 They
also were represented in early times as weaving or spinning.3 But while the Moirai embodied
the unknown dispensations of the powers of destiny, and are therefore represented by Pheidias
as resting on rocks in careless self-abandonment, the Horai personified the conception of
conscious adherence to law ; this contrast is especially brought out in the names which are given
to Horai and Moirai in another passage of the Hesiodic Theogony,4 and which afterwards
became very popular. It represents the Horai and Moirai as sisters, alike the children of Zeus
by Themis ; but while the names Klotho, Lachesis, Atropos indicate the blind decrees of fate, the
Horai are called Eunomia, Dike, and Eirene : it is in law and right that they have dominion.
Conscious of this contrast, Pheidias made the Horai sitting on thrones and maintaining a tranquil
dignity. The old statues of the Horai by Smilis in the Heraion at Olympia in like manner sat upon
thrones : \tirap60povoi they are called by the poet above mentioned, who invokes them with the
Moirai, and Pindar designates them as evdpovoi.5
Finally, the position of the two enthroned women in the pediment agrees excellently with the
interpretation of them as Horai. They are on the side where we have placed Hera and Iris. The
close connexion of the Horai with Hera is well known ; they already appear in the Iliad as attendants
assigned to her. On the other hand, the Horai are as closely associated with Helios, as the Moirai with
Nyx. It was precisely in Athens that took place the processions called Eiresione, which, according to
the testimony of Theophrastos, were in honour of Helios and the Horai.0 And lastly Brunn has
already pointed out that the Horai, who, according to the Iliad, open and shut the gates of Olympos,
are most suitably placed just here at the end of the pediment near Helios.
As for the number of Horai represented, there is as much evidence for a group of two as for one
of three. The Horai as two are known on works of art of various places from early times, and even
as late as the fourth century ; 7 at Athens 'notably we know from express testimony that two Horai
were worshipped under the names Thallo and Karpo, designating the two principal seasons of the
year — the time of blossoming and that of fruit.8 This of course does not at all exclude the represen
tation of the Horai by Attic artists also as three, in accordance with the more generally accepted Hesiodic
tradition.9 When Pheidias, on the contrary, gave the preference to the group of two, no doubt artistic
considerations had most weight with him in this decision. The third figure must be recumbent ; but
if he represented two Horai enthroned he could not make the third lying down ; moreover, it must
have accorded much better with his efforts after a pleasing variety to have a male figure here
answering to the outstretched female figure on the other side.
This male figure is sufficiently characterized to make it possible to assign him a definite name. The
powerful youth with muscles steeled by exercise and with short close-cropped hair 10 reclines upon a
rock over which he has spread the skin of a wild beast, and over that his cloak ; the left hand held a
bronze attribute,11 and his feet had a covering of the same metal.12 Herakles has been suggested on
account of the skin, but the monuments show us that he would not have spread a garment over it :
moreover, the skin would need to be a distinct lion's skin ; and the covering of the feet excludes
Herakles, apart from the fact that he would be a quite inappropriate figure here.13 Dionysos again is
entirely impossible, on account of the powerful form and the smooth short hair. Although Dionysos
1 Cf. e.g. the passage of the lyric poem cited above, where Moirai and Horai are invoked together
(Bergk, Poetae Mel. frg. adesp. 140). Cf. also the Amyklaian Throne (altar-relief), and the Zeus of
Theokosmos at Megara (Paus. i. 40, 4). An equally favourite contrast was that of Horai and Charites.
2 Cf. Rapp in Roscher's Lex. i. 2738 seq.
3 Cf. ibid. 2715.
4 Hes. Th. 901 sqq.
5 Cf. supra, n. I. Pind. Pyth. 9, 60.
6 Apud Porphyr. De Abstin. ii. 7, Schol. Aristoph. Equ. 729 ; Plut. 1054. Properly it was Apollo,
who was identified with Helios. Cf. Mannhardt, Antike Feld- und Waldkulte, 217 seq.
7 Cf. the old Ionic Phineus Kylix (Mon. d. Inst. 10, 8), the table of Damophon at Megalopolis, Paus.
viii. 31, 3, and the Tarentine vase of the fourth century, Stephani, Compte Rendu, 1862, PI. 4.
8 Paus. ix. 35, 2.
9 Cf. the Francois Vase and the Cylix by Sosias (Berlin, 2278).
10 Cf. the more accurate account given by me in Arch. Zeitg. 1881, p. 304, from the original, and the
correction of the mistaken assertion made by Overbeck (and lately repeated by A. H. Smith, in the Catal. of
Sculpt, i. 107) that he had plaits behind.
11 Sauer, loc. cit. 82.
12 Cf. ibid.
13 Cf. Petersen, Phidias, p. 118.

468 THE TEMPLES OF ATHENA ON THE AKROPOLIS
in the time of Pheidias was certainly represented as beardless and with short hair,1 and in fact
appears in this guise in the frieze of the Parthenon (cf. supra, p. 431), yet there are no analogies at all
for the representation with such smooth uncurled hair as we see here. On the contrary, the only
parallels for this are found on heads of athletes and similar figures. In addition to this the forehead
too, with its marked protuberances towards the middle,2 resembles the forehead of athletes and
heroes, especially those of the Myronian school.
In fact, there is only one interpretation which satisfies all requirements : the man is Kephalos?
the beautiful hunter.4 Everything is explicable on this interpretation ; the skin is just the
characteristic of a hunter, and it is precisely to a hunter that the covering of the feet also is appro
priate. To him also are suitable the strong beautiful form, the athletic head, and the rocky seat.
And Kephalos is also the only man who is fitting in this position close to Helios ; and he is the only
person with regard to whom it is not disturbing, to whom it is even quite suitable, that he (as was
required by parallelism with the Moirai on the other side) should take no interest in the event going
on in the centre, but should sit quietly turned towards the angle and Helios. When the sun rose
yonder behind Hymettos, then the Athenians of Pheidias's time, as we are shown by a beautiful
painted vase, bethought themselves of Kephalos, who was carried off thence by Eos.5
It is precisely on vases of the Periklean period that the myth of Kephalos is specially popular.
Eos carries off the beautiful youth whom she loves to the place where her home is on the boundary
of heaven, by the stream Okeanos ; there he begets by her the dazzling Phaethon, the morning
and evening star.0 Thus Kephalos sits in the pediment close to the edge of Okeanos, the waves
of which reached to his feet.7 The bronze attribute which he carried in his left hand is naturally to
be explained as the lance which is in fact characteristic of Kephalos as huntsman, and without which
he hardly ever appears on the monuments.8 The lance rested on his left shoulder ; the right hand
we may conjecture to have been without attribute, and raised in greeting to Helios.
It was a specially skilful idea of Pheidias to place a figure so well known, and so dear to every
Athenian as Kephalos was, between Helios and the Horai at the entrance to Olympos, where
Okeanos flows round it. At the same time he has made, in the strong and youthful figure who bathes
his breast' in the ruddiness of morning, a splendid counterpart to the deceptive beautiful Fates, akin
to the darkness of night, who spin the mysterious future.
Thus in the eastern pediment also, as in the western one, Pheidias represents only easily
intelligible and truly living figures : at this end as at the other, his creations are both poetical and
popular. VII. The Earth Goddess entreating for Rain, near the Parthenon.
In front of the north side of the Parthenon, cut in the living rock, runs the inscription, belonging
perhaps to the time of Hadrian, rfj? Kapirocpopov Kara pavreiav? Here must likewise have stood
the image described by Pausanias in his Periegesis of the Akropolis (i. 24, 3) as rfjs ayaXpa
iKerevovo-ris vaai oi tov At'a : it showed the goddess Earth entreating heaven for rain. A drought
: Some early examples of the beardless Dionysos I have mentioned in Samm. Sabouroff, on PI. 23, n. 6 (cf.
besides, Back, in Fleckeisen's Jahrb. 1887, 445 sqq.) The type of the golden mask from Kertsch there cited by
me was no doubt created in the fifth century, in the Pheidian period ; it gives the god curly hair, abundant
but not long. On the Attic vases, also, of the time of the pediment, the beardless Dionysos is not
unfrequent ; but he has always abundant and rather long curls (cf. the Melian ' Giant ' Vase ; also Samm.
Sabouroff, PI. 55, 57).
I In the original, and even in good photographs, very plainly to be seen in spite of the weathering.
» lhe_ interpretation has been already brought forward by Brdndsted. Brunn's explanation of the figure as
Olympos is at once excluded by the fact that neither Greek poetry nor Greek art knows any personification
of the mountain Olympos (cf. supra, p. 458, n. 3).
4 That he is beautiful and that he is a hunter are his two principal characteristics in the myth.
- Ihe so-called Blacas Krater,' Mon. d. Inst. ii. 55 ; Gerhard, Gesamm. Abh. PI. 5, 2 ; Roscher's Lex. i.
2010. The rape of Kephalos was localized on Hymettos.
6 So Hesiod, Tkeog. 986. Cf. particularly von Wilamowitz, in Hermes, xviii. 421 sqq. Since Kephalos was
of course conceived as dwelling and living with Eos in heaven, I do not understand Petersen's objection (Phidias,
p. 119) to his presence in the pediment. With regard to the original identity of Kephalos and Orion, see von
Wilamowitz, loc. cit. 425 ; Rapp in Roscher's Lex. ii. 1097.
7 According to the position of the figure established by Sauer.
Especially similar to the representation in the pediment are the coins from Pale in Kephallenia (Head, Hist.
Num. p. 35S), to which Miss J. E. Harrison has courteously drawn my attention. Kephalos, known with certainty
from^ the inscription, is seen sitting on a rock and holding a spear.
, Jr-J: A\}n- l66- Cf. Heydemann, in Hermes, iv. 381 sqq. Petersen (Lanckoronski, Stddte in Pamphylien
undPisidien, 11. p. 5c 1 seq) proved the worship of Ge Karpophoros at Tennesson in the neighbourhood of the shrine
of Zeus Solymeus and of another belonging to Zeus and Dione. He there recalls how at Athens likewise Ge
Karpophoros was close to the place where Zeus (i.e. Polieus) was worshipped.

EARTH GODDESS

469

Fig. 185.— Attic
seal.

had of course been the occasion of its erection, but Pausanias gives no more exact details. The
supposition is certainly right that only the upper part of the form of the goddess was represented,
and that being placed immediately on the rocky ground she seemed to be rising out of it.1 Even if
the image was not older than the inscription (doubtless however merely a restoration
of an older one), yet the subject and the type certainly belong to a far earlier time.
For determining the way in which this type is to be conceived, and whence it
was derived, the interesting little monument given in Fig. 185 seems to me to be of
importance. It is the impression of a seal, preserved on one of those common little
pyramids of terra-cotta. The object is in the Berlin Museum,2 and comes from a
private collection at Athens. The qualily of the clay seems to show that it is also of
Attic workmanship, and it might perhaps belong to the fourth or third century B.C.
We see the upper part of an unclothed female form, bending far back, and looking
up towards heaven. The left hand is raised before her face, the right grasps her long dishevelled hair,
which streams down behind. The figure expresses an urgent and pathetic entreaty to heaven. The body
is cut off in its lower part, and rests— this is especially noteworthy— on a kind of car, with a wheel of that
old-fashioned sort which, in place of radiating spokes, has bars in the form of a double cross. Wheels
of this kind3 occur only on vehicles such as carts, upon which the occupants sat, and to which were
yoked mules or oxen,4 such as the carts used in the fields ; thus the winged chariot on which, in
an old Attic vase-painting, Dionysos is driving, as does Triptolemos, to extend the blessings of
agriculture, has this same shape.6 The cart of our seal is, like that of Triptolemos, wholly without
a pole, but it has round it a peculiar fringed rim ; there is also something projecting at the top
before and behind the woman's body. It looks exactly as though
the cart were covered with cut grass, corn, or the like.
This strange representation must clearly be connected with certain
religious usages, of which we meet with occasional notices. At
Krannon, for instance, the Thessalian town which lay in the fruitful
Pelasgiotis, there was a sacred chariot of brass : if a drought occurred
in the land, this was moved rapidly backwards and forwards, while
prayers were offered over it for rain. There were moreover here
two sacred ravens, and, as it was said, no other raven in the territory
of the town. Now chariot and ravens were so highly reverenced that they were used as the coat of arms
and seal of the town.0 Thus they appear for instance on a coin which has come down to our time
(Fig. 186).7 The vehicle is again a mere cart, as on the Athenian seal ; but it appears to have had four
wheels, at least there are two (presumably on the same side) represented. On these sit the two ravens
facing each other. In the middle, on the cart, stands a large amphora. This last clearly indicates what
it was desired to secure — heaven's moisture, rain. We may conjecture that when in time of drought this
rain-charm was used, the vase was filled with water, so that it spirted out when the chariot was shaken to
and fro. Dipping in water and sprinkling with water are among the most favourite forms of rain-
incantation in the circle of Indo-Germanic religions.8 The two ravens, again, become intelligible, if we
recall the Greek legend, occurring in diverse variants, to the effect that the raven in summer, at the time

Fig. 186.-

-Two bronze coins of
Krannon.

1 So Heydemann, loc. cit., and Dittenberger in the C. I. A. Kuhnert is mistaken in his conjecture (Roscher's
Lex. i. 1 581) that it was a relief showing Zeus and Gaia.
2 Antiquarium, T.C. Inv. 6787. It is pierced at the top. With regard to these pyramids of terra-cotta,
cf. Olympia , vol. iv. Die Bronzen, p. 206, on No. 1331, and the works there cited; moreover, Annali, 1872,
PI. M, and the interesting notice, Ann. 1884, p. 237 seq.
3 Cf. with regard to this the proofs given by me in Olympia, vol. iv. Die Bronzen, text, p. 69, on No. 510.
4 An ox-cart of this kind is seen, e.g., in the old Makedonian coin, Berlin Catalog, ii. PI. 7, 67. Mules are
more frequently drawing them. 5 Gerhard, Auserl. Vasenb. 41.
6 So Antigonos of Karystos, Histor. Mirab. xv. (ed. Keller) : 'Ev Se Kpdvvavl rijs ®ertaXtas Svo ipaalv pivov
elvai KSpaxas- Sib Kal eirl tav irpo£eviav tav avaypaipopevav rb irapdo-qpov trjs ir6Xeas . ..{nroypdtpovrai Svo Kopaxes
ecp' apa£iov x«A.koC, Sia rb p-qSe irore irXtiovs rovrav SitpBat. r) Se apa£a irpomapdKeirai St* roiavr-qv alriav
\evov yap loas av k*\ tovto ipaveiij- eativ aiitoh avaiceipev-q xa^'ci), V 'irav aiixpbs ft aeiovres vSap alrovvrai rbv
Beov, Kai <paai yiveaBai. There follows an excerpt from Theopompos lo the effect that the two ravens, when they
have reared up young to take their place, always fly away again.
7 Cf. Brit. Mus. Catal. Thessaly, PI. II. 14, 15 ; Eckhel, Doclr. ii. 36 ; Head, Hist. Num. p. 249. Two
types are to be distinguished, both of which we give in Fig. 186. One, which is the older, shows on the obverse
a head, on the reverse a chariot with hgavy spokeless wheels, on which sit the ravens. The other has on the
obverse a horseman, on the reverse the chariot with wheels well executed, but for the most part without the
ravens A specimen at Athens and one at Klagenfurt (I have seen impressions, through the kindness of Imhoof-
Blumer) have in each case one raven only on the right-hand wheel. Fig. 186 is taken from two specimens at
Berlin, but the inscription on the older coin has been completed with the help of the impression of a better-
preserved specimen belonging to the collection of Imhoof-Blumer. The inscription is Kpavvovviovv (cf. Collitz,
Dialektinschr. i. 364). Cf' Mannhardt, Baumkultus, passim.

470 THE TEMPLES OF ATHENA ON THE AKROPOLIS
of the greatest heat, when the corn or the figs are ripening, is tormented by an unquenchable thirst,
and that his croaking is evidence of this. It is supposed to be a punishment inflicted by Apollo because
the raven when sent out by him to fetch water neglected this commission.1 We may conjecture that
this myth arose in the first place out of the ritual, which at the time of the greatest heat attributed
an important significance to the raven. Of what particular kind this was is indicated in the feature
of the story which says that the bird was sent to fetch water. Ravens were considered in antiquity
as heralds of impending rain.2 In German mythology also and the two ravens of Odin, it is still
clearly perceptible that they are the dark storm-birds of the rain-clouds.3 They are to bring down
the water of heaven ; that is why they sit on the wheels of the chariot of Krannon, which produces
rain at times of drought.
Many traces from allied spheres of civilization point to a same high ritual importance attaching
to a sacred chariot, as at Krannon. An instance is the sacred chariot set up in Phrygia in the
principal shrine at Gordion.4 And from Phrygia, with the worship of the mother of the gods —
the great earth goddess— was also transplanted to Rome the usage of driving about her idol
on her chariot at her festival, and then bathing it. The bathing of the idol was probably here and
elsewhere originally intended merely as a kind of rain-charm, like all dipping into water. In the
expedition of Xerxes there were in the centre ten sacred steeds, and behind them the sacred chariot
of the supreme god, drawn by eight steeds, whose driver went on foot beside the chariot ; next
followed the king.6 Among the Germans, according to Tacitus, quite the same usage seems to have
obtained ; 6 a sacred car with horses was led by priests and princes. Again, in the North of Germany,
as Tacitus tells us, at the beginning of spring, Nerthus, whom he designates as terra mater, was driven
on a sacred car drawn by cows through the country, and finally car, robe, and nwnen were bathed in a
lake.7 Traces of similar customs in German ritual have been repeatedly pointed out.8 We should
mention too that in several European countries it is still customary to bring in what Mannhardt
calls the ' Daemon of Vegetation ' on a car. This deity is sometimes represented .in vegetable
form, as may-pole, harvest tree, or the like, sometimes in human shape, being then a figure
thickly enveloped in foliage.8 From the regions of the Balkan and from modern Greece we hear
of the following custom.10 When there is a prolonged drought, in the height of summer, a young
girl, completely covered up with greenery and flowers, is driven about and sprinkled with water ; this
is intended to elicit rain,11 and at the same time a song is sung entreating for rain. The girl covered
with foliage represents the spirit of growth in earth, which desires to be sprinkled with water.
The last-mentioned usages explain to us the grass-like covering of the car on which the goddess
entreating for rain appears upon our Athenian seal.
But what is the meaning of the car? Originally it is no doubt a pure symbol. In the ritual
usage at Krannon, in which the car is pushed, rolled, or shaken to and fro, the water meantime, as
we may suppose, spirting out of the amphora, the significance is still recognizable — namely, the
car betokens the cloud, the roaring and rattling cloud of thunderstorm and rain. In the German
mythology this symbolism is still particularly plain. Thor and Odin ride both of them on
roaring cloud-chariots ; but especially in the beginning of spring the great cloud goddess makes
1 The simplest and probably oldest conception is given by Aelian, De Nat. Anim. i. 47 : <S>pbyerai Sia toC
Bepovs 6 K6pa\ rip Siif/ei KoXa(6pevos, Kal ,8ocj tV ripapiav paprvpopevos. Apollo sends him to fetch water ; he
comes to a green cornfield, and in order to be able to eat the grain waits till it has become quite dry with the
heat ; this makes him forget his commission, and accordingly as a punishment he has to be always thirsty in the
driest time of the year. A somewhat different and later form occurs, Eratosthenes, Kataster. 41 (Robert, p. 188),
following Aristotle : The raven is to fetch water for a sacrificial libation ; he comes to a fig-tree, and waits till the
figs are ripe, then brings n water-snake and the Krater to Apollo, and excuses himself by saying that the snake
always drank up the water ; as a punishment he was obliged to thirst, irepl tr)v a.Kp'qv trjs biriipas (Schol. Aral.) ;
Raven, Krater, and Hydra occur among the constellations. According to the poem of Dionysios, 7repl opvlBav
(excerpt Cramer, Anecdota Gr., Paris, i. p. 25), lust was the cause of his not executing the commission. In the
treatise of O. Keller, mentioned in the following note, no account is taken of these interesting myths.
2 Cf. O. Keller, Rabe und Krdhe im Alterthum (1. Jahresber. d. wiss. Vereins f. Volkskundeund Linguistik
in Prag, 1893), p. 8.
3 J. Grimm, Deutsche Mythol. 4th ed. p. 559 ; Elard Hugo Meyer, German. Mythol. p. 112.
4 Arrian, Anab. 2, 3, where the temple is designated as that of Zeus Basileus.
5 Herod, vii. 40.
6 Tacit. Germ. 10. Cf. Mannhardt, Baumkultus, 580.
7 Tacit. Germ. 40. Cf. especially Mannhardt, Baumkultus, 567 sqq. ; Miillenhoff, Deutsche Alterth.-Kunde,
2, 28 ; E. H. Meyer, German. Mythol. p. 287 seq.
8 Cf. Mannhardt, loc. cit. So the Goths in the fourth century drove about an idol on a baggage-wagon
(appdpa^a), loc. cit. 578.
9 Mannhardt, Baumkultus, p. 154 seq.
10 Ibid. p. 328.
11 Similarly it is customary to sprinkle the 'harvest tree' with water, to guard against drought in the
following year (ibid. p. 214).

EARTH GODDESS 47 1
her progress in a chariot, or else on a ship, for the latter is merely another, not less favourite,
symbol of the cloud.1
The ship of the Panathenaic festal procession also had in the beginning no other significance
than this. It pertained to the old agrarian Athena who made the tilled land fertile, who nourished
the olives, in the product of which consisted the prizes given at the games held in honour of her
festival. The progress of her ship is wholly similar to that of the German goddess of clouds and
of the earth. Athena is just as closely connected with ship, car, and plough,2 the old symbols of the
clouds, as is that German goddess. Pallas was also bathed at Athens, it is only uncertain at what
festival.3 Like Athena, Dionysos also — not the Thrakian Dionysos, but the old Greek 'god of
vegetation'4 — was driven about in Attika upon a ship that moved on wheels.6 It is an old mistake
to try to explain these progresses on a ship by a supposed immigration from abroad.
Of the two monuments — the type of the coin from Thessaly, and the impression of the seal from
Athens — the former undoubtedly represents an earlier stage of ritual than the other. For the former
shows as yet no daemon in human form upon the chariot, but only the symbolical amphora and the
two ravens. The Phrygian, Persian, and German sacred chariots in early times were likewise without
human image.
The chariot of Krannon, again, so strongly recalls certain still extant chariots of the Bronze age
and the earliest part of the Iron age, found in Mid and North Europe, that we must necessarily speak
of them here. There are the ' cauldron chariots,' as they are called,6 small four-wheeled contrivances
of the nature of chariots, on each of which rests a cauldron, exactly as at Krannon. Sometimes
heads of birds are affixed. The celebrated chariot of Judenburg is most richly fitted up ; 7 it has a set
of figures denoting a sacrifice of a stag and a festal procession ; in the centre stands a woman
supporting a cauldron. The religious significance of this utensil has for a long time been con
jectured ; but interpreters have gone far astray in connecting it with the great cauldron of Solomon's
temple, moving upon wheels, and which was intended for the purification of the sacrificial gifts.8 The
use of wheels under utensils of all kinds, as here below the cauldrons of Hiram, has nothing at all to
do with those northern ' cauldron chariots ' : we find, it also in the Homeric epos, and old Etruscan
and Greek discoveries have brought it to our knowledge.9 In my opinion — for which I will give
the reasons more in detail elsewhere — it is one of the most serious errors of prehistoric archaeology,10
that it has been led by the superficial resemblance to suppose a ' Semitic Oriental' influence on the
early European artistic industry just in the point where it least of all, according to my conviction,
existed. Those 'cauldron chariots' are no more Semitic than the chariot of Krannon. On the
contrary, they probably had the same significance : they were properly, I think, religious symbols
of the fruit-bringing rain-cloud ; and we may conjecture that they, like the other, were shaken in
times of drought.
The utensils in the form of birds mounted on four wheels11 which are found in Italy, and also in
Bosnia and Hungary, likewise remind us of the chariot of Krannon with its ravens, and probably had
a similar religious significance.
The representation on the Athenian seal is far in advance of this stage of primitive symbolical
suggestion. It is true that the car covered with grass and foliage is still quite a popular symbol,
which was no doubt used in country processions and supplications at times of drought in Attika,
as we may now conclude. But here there rises above the car the image of the earth goddess
herself, in the type established by art, with only the upper part of her form projecting ; in a lifelike
and strikingly beautiful fashion she makes entreaty to heaven, that it should please it to send rain
upon the thirsty ground. In a like pathetic attitude must we conceive that image to have been
represented which once rose to the north of the Parthenon directly from the rock.
1 Cf. Elard Hugo Meyer. German. Mythol. pp. 90, 232, 239, 281 sqq., 290 seq.
2 Athena is the inventress of the ship, the plough, and the chariot, or-in the last two cases-at least of
harnessing animals to them. Cf. Roscher's Lex. i. 680 seq. _ . .,,.,.,,,
3 Cf. Preller-Robert, Griech. Mythol. i. 209, n. 3. The procession of the Epheboi with the idol to
Phaleron is generally ascribed to the Plynteria or the Oschophoria ; either ascription is quite uncertain.
4 Cf. Back in Fleckeisen's Jahrb. 1887, 444 sqq.
5 Inghirami, Vasi Fittili, i. 33. an Attic black-figured vase.
5 What is known on this subject is collected and reviewed by Undset in Zeitschr. f. Ethtwl 1890, pp. 56-61.
7 A good reproduction in Much, Vorgeschichtl. Atlas, PI. 41- Undset, loc. at.
9 Cf. the instances collected by Undset, p. 71 sqq. .
" Undset's conclusions, which are connected with his general views-erroneous as I am convinced-with
regard to Oriental influences on the style of early European, and specially of early Italian, art, are indorsed by
H°™ecV.^Undfet^^.P«V4I49 sqq. The birds sometimes have ox-heads, which certainly has a daemoniac
significance.

INDEX

INDEX

(Numbers following after a comma refer to the notes. )

I. HISTORICAL

AEGINETAN ART 7. 19. 416, 8
ARCHAISTIC 42. 441.
FINE PERIOD UP TO ALEXANDER.
a SCULPTORS.
HAGELAIDAS AND THE OLD ARGIVE
SCHOOL. Male type 25. 49^/. 182. 191 sqq.
212 sqq. 226. 285. 287 sqq. 312. 354. 359.
Female, type 23. Further development after
Hagelaidas 196. 212. 226. Introduction of the
walking motive 213 seq. 226. School tradition 290
seq. Argeiadas 196. Dionysios and Glaukos 215.
Offering of Smikythos 215.^7. Unknown master
connected with the Argive school 290 seq. Apollo
52. Dionysos 42. Zeus 212 sqq. Youth 353, 2.357.
HEGIAS 54. Vjiseq. 194. 197.
KRITIOS AND NESIOTES. Style 46. 53. 171
seq. 194. 197. 223, 1.
Works in style of 7. 19. 49. 52.
KALAMIS. Style and works 24. 42. 81 seq. 171.
175, 2. 182. 191. 196, 2. 197. 298. 437.
SCHOOL OF KALAMIS (Praxias and others) 55.
81. 197. 298.
PYTHAGORAS. Style of works 24. 108. 172.
Female types 24. Influence on Sicilian coin-
types 108. 172. Perseus (?) 197, 3. 199, 1.
TELEPHANES 57.
MYRON. Period 182 seq. Pupil of Hagelaidas (?)
196 sqq. General character 201 seq. Ancient
criticisms 173 seq. 182 seq. Treatment of hair 170.
173. 181. 199. Style of drapery 212. 188,5. Forma
tion of eyes 170 seq. Formation of the forehead
169. 173. 178. 180. 182. 199. 210. 468. Expression
of countenance 173 seq. Myron and Hagelaidas
175. 182. iqoseq. 196. 210. 212*7.7.; and Kritios
and Nesiotes 194 ; and Kalamis 190 seq. ; and
Pythagoras 172. 175 ; and the Olympian sculptures
188 ; and Pheidias 202.

Works. Diskobolos 168. 173. 180. 191 seq. 193.
199. 204. 287. Marsyas 126. 180 sqq. 193. 196. 259.
Perseus 197 sqq. Apollo 190 sqq. Zeus 186 sqq.
Herakles 178 sqq. 202. Erechtheus 204, 205.
Goddess 202. Medusa 201 seq. Heroes 165. 169.
181. 193 Hermes (?) 182. Poseidon (?) 184
Athletes 175 (Youthful v ork). 172 sqq. 204.
Portraits 175 seq. 191. Works falsely attributed to
46, 3. 287.
LYKIOS, son of Myron 182. 259. 296.
PHEIDIAS. Period 25 seq. 48. Pupil of Hegias
53. of Hagelaidas 53. Painter 45. 48. Kimonian
period 33 sqq. Law-suit and death 36 sqq.
Apocryphal portrait of 48. Descendants 39.
Pupils and colleagues 32. Pose of statues 25. 27.
30. 78. 244. 249. Form of ear 18. Style of relief
43 sqq. ; Ph. and Hagelaidas50 sqq. ; and school of
Kalamis 81 ; and Olympian sculptures 23 ; and
Alkamenes 20. 84. 90 seq. 95 ; and Kresilas 136 sqq
and Polykleitos 137. 244. 291 ; and Perikles 34. 36,
Works. Athena Lemnia 4 sqq. ; related to 26. 49
66. 77. 100. 106 seq. 138. 291. Athena Parthenos
10 ; (head) 107 ; (date) 33 sqq. ; relation to the
Lemnia 43 sqq. 47. 99. 108 seq. ; (shield) 45
99; (basis) 21. 26 seq. 29. 31 seq. 66. 90. 94
426. Athena in Rome 78. Athena at Pellene 36
at Plataia 36. Promachos 9 seq. 31 sqq. 46. 100
sqq. Delphic group 34 seq. Olympian Zeus 36
sqq. 43 seq. (Throne reliefs) Parthenos 38. 45 sqq.
Pantarkes 39. Anadumenos 39. 244 sqq. 298.
Aphrodite Ourania 71. 249. 399. Amazon 136 sqq.
Colossus of Monte Cavallo 95 seq. 421. Cliduchus
32> i-
Works in style of Pheidias. Early period.
Apollo 49. 197. Bearded head 55. Female
heads 58 seq. 138. Middle and later period
26 seq. Athena 60. 76. 90. Aphrodite 66.
140. Eros 69. 140. Mousaios (?) 64. Hero (?)
90. Female head 66. Anakreon 60. 244.

476

INDEX

Works falsely attributed to Pheidias 32 sqq.
57. 144-
Influence and School of Pheidias. Aphro
dite 71. 399. 400. Apollo 410 sqq. Ares 92 seq.
Asklepios 55. 227. Athena 78. 144. Herakles
298. Leda 395. Female types with coif 66, 2.
71. '¦ 395- Goddess in animated movement 104.
107. Zeus or Poseidon 104. Zeus Meilichios (?) 212,
1. Coin-types 104 seq.
ALKAMENES. Period 90; and Pheidias 84. 90 seq,
95. Head types 84. Aphrodite 19 seq. 82. 275,
10. Ares 89 sqq. 95. 155. 298, 3. Athena 76 jw/.
84. 141 ; Without a helmet (?) 85. Athlete 90.
137 seq. 296. Dionysos 90. Hekate 90, 3. 260 seq.
Hephaistos 88, 89. Hera 82. 84. 90, 3. Herakles
298, 3. Orpheus relief 88.
AGORAKRITOS. Nemesis 8. 85. 400. Hades
and Athena 80, 1. Apollo 88. 305, 6. Athena 88.
Aphrodite 400.
PRAXITELES THE ELDER 32 sqq. 99 sqq.
102 sqq. 296. 412.
KALLIMACHOS 437 seq. Hera 437, 5. Lak.
438 seq. Other works ibid. 450 seq.
TEMPLE SCULPTURES, Stylistic. 'Theseion'
46.99. 182, 5. 448. Sunium 46. Parthenon: metopes
29 seq. 43, 6. 45 seq. 62. 97. 99. 140. 182. Frieze
and pediments 19 seq. 30. 43, 7. 44. 46 sqq. 97.
106. 136. 140. 244. 254. 399. 449. Nike temple
21. 245, 2. 439. 445. 449. Erechtheion 11. 450.
Phigalia437, 4.
KRESILAS. Period and Life 115*??. 165. Develop
ment 165. Kr. and Myron 165. 168. 170 seq. 180.
188. 196 seq. 201 seq. ; and Polykleitos 134 sqq. 144.
154. 243. 254 seq. 265 seq. 268. 298 ; and Pheidias
136 seq. 144. Imitated in later times 165.
Works: Perikles 9. 62, 8. 117 seq. 145. 178. Dii
trephes 122 sqq. Amazon 128 sqq. 145. 154 seq. 254
seq. Athena 141 sqq. 188. Diomede 146 sqq. 154
seq. 165 seq. 201. 265. 298. Medusa 156 sqq. 201.
Athlete (Doryphoros) 161 sqq. The Runner 126
sqq. 178. Works falsely attributed to 62, 8.
MASTER RELATED TO KRESILAS. 247.
248, 4.
STRONGYLION 140.
PAIONIOS 41, 6.
DEMETRIOS 122, 2. 321, 1.
MYS 34.
POLYKLEITOS. Period 130. 140. 196. 223.
Birthplace 225 sqq. Family 226. Pupil of Hage
laidas ? 196. 214 seq. 223. 226. Works which can be
dated 223 seq. Bases in Olympia 224. 249 sqq.
Quinctilian verdict 237. 243. Beardless figures
236. 292. Delicate gradations from boyhood to
adolescence 252. 257. 264 sqq. 281 sqq. 292.
Standing motive 214. 227. 242. 248 seq. 254.
260. 266. 274. Characteristic defect 134. 266
seq. 243 seq. 254. 261. 265. Attic influence
upon P. 243 seq. 261. 266 seq. 274. 277. 291 seq.
298. P. and Pheidias 138. 245. 292 ; and
Kresilas 134. 144. 154. 243. 254. 265. 268. 298 ;
and Myron 236 ; and Patrokles 226 ; and the Attic
masters of the fourth century 298 sqq. 312.316 jy??.

Works: Doryphoros 138. 140 seq. 151, 5. 163. 227
seq. 242 ; adaptations from 229 sqq. ; Canon 196.
229. Amazon 128. 134. 138. 140. 247^. Diadu
menos 238 .r«/. 242 sqq. 247. 268. Herakles 234 sqq.
266. Hera 223. 242. Kyniskos 214. 224. 249 seq.
Pythokles 224. 262 sqq. Xenokles 224. 279. Aristion
225. Destringens se 262. Nudus talo incessens 249,2.
Pueri astragal. 292, 1. Youth, hand on back 238.
Boy crowning himself 249 sqq. Adaptations 255 seq.
Boy crowned 256. Boy athlete 265 sqq. Boxer
246. Oil-pourer 257 seq. Athlete with aryballos
264. Standing still 279 sqq.
Falsely attributed to 92, 1. 144. 214. 223, 1.
290, 4.
POLYKLEITOS, SCHOOL OF 225 seq. 229.
249. 256. 286. 289. 291 seq.
Works : Adonis (? 'Narkissos') 272, 5. 357. Ares
271. Aristaios 278. Asklepios 277, 6. Dionysos
272.350^. Boxer 246. Hermes 288 sqq. Youth
standing still 279 seq. Idolino 283 sqq. 290. 350.
Other works : 265. 268 seq. Pan 229. 269.— Dis
tantly related 290 sqq. — Falsely attributed to 223.
243 (Heraion at Argos).
PHRADMON 128 seq. 140.
PATROKLES 226. 286. 289. 290. 292.
POLYKLEITOS THE YOUNGER 224 sqq. 288.
302.
NAUKYDES 225. 249. 288, 1. 313. Immolans
arietem 231. Hermes 288.
DAIDALOS 225. 249.
KLEON 225. 249. 278.
KEPHISODOTOS THE ELDER 146. 296 seq.
SKOPAS. Origin and school 302. Earlier period
under Polykleitan influence 296. Stylistic develop
ment in the bodily forms 305 seq. 396. Marble
technique 310. 391. Works in bronze 394. Female
bodies 392 seq. Type of face 391. Ear 391 seq.
395. Sk. and Polykleitos 235. 302 seq ; and Praxi
teles 391 seq. 392. 395 seq. ; and Lysippos 301.
363. 394 seq.
Works : Aphrodite 391 seq. 394. 396. A. Pandemos
at Elis 394. Apollo Smintheus 303, 2. 336. 388.
Kitharoidos 305. 397. Ares 304. 394. Asklepios
300. Athena 305. 439. Column at Ephesos 307.
Goddess, head in Athens 309. 334. 344. 391. Others
392. Herakles 66, 5. 235. 29(1 seq. 301, I. 307. 341,
1. 396. Hypnos 303. 395 seq. Youth, foot raised
on eminence 302 seq. Kanephoros 337, 1. ' Leda'
394 seq. Maenad 397. Meleager 184. 304 seq.
Pan 303. Adaptations 392, 5. 397 seq.
Falsely attributed to 337.
PRAXITELES. Period 295. 309. Sojourn in Asia
Minor 322, Phryne 318. Earlier period under Pelo
ponnesian influence 318 seq. 324. Skull 312. Hair
309 seq. 334. 344. 391. Eyes 334. 345. Ear 345.
Mouth 345. Female type of face 345. 391. Breast
392. Marble technique 309 seq. 334. 344. Bronze
casting 391. Pr. and Polykleitos 312. 317; and
Skopas 344. 391 sqq.
Works: Aphrodite, earlier period 387. 391. Knidian
319 sqq. 324. 387. 391 seq. Koan 322. 385, I. Origi
nal head of A. 320. 334. 343 sqq. Apollo, earlier

INDEX

477

period 314 seq. 337 ; later 337. Sauroktonos 336.
388. Artemis, Brauronia 102, 3. 323. 388. 426, 3.
431, 4; another of an earlier period 324 seq. ;
another 326 sqq. ; at Antikyra 328. Athena 326.
Bonus Eventus, Bona Fort una 349 seq. Dionysos
311, 2. 314 seq. 323, 5. 337. In Elis 308, 8. Infant
Dionysos 296. Eros 313 sqq. Thespian 271, I. 317.
355; of Parion 336; of KaUistratos 314, 1. 340.
Eubouleus 103. 330. 331 sqq. 340. 391 sqq. Group
in Mantineia 295, 1. 309. 313, in Megara 313, 1.
Herakles 340 sqq. Hermes in Olympia 190. 302.
307 sqq. 330. 334. 337 sqq. 342. 344. 345. Another
Hermes 338. Kanephoroi 337, 1. Muses on the
Mantineia basis 309. 319. 325. Leto 326. Opora
337,1. Phryne 320. Pseliumene 387. Satyr pour
ing wine 310 sqq. 319; leaning 329 sqq. Trip
tolemos 334. Tyche 325. Type of Zeus 190.
' Aberdeen head ' 346 seq. Falsely attributed to
337. 5- 354- 356, 7-
SONS OF PRAXITELES 309. Hermes 338. 362.
Aphrodite 345.
EUPHRANOR. Period, school 348. General
character 363. Influence in Athens 356. Develop
ments, offshoots of his tendency 363 seq. E. and
Praxiteles 355 seq. 363 seq. Skopas 363 seq.
Lysippos 363 seq.
Works : Aphrodite 359. Apollo 352 seq. Athena
359 S1J- Bonus Eventus 349 seq. 353. Dionysos
3^0 seq. 356**7. Herakles 357. Leto 349. Pan
359. Paris 357 seq.
SOSTRATOS, son of Euphranor 349, 3. 363.
SEILANION 71, 1. 153, 4.
BRYAXIS 230, 2.

LEOCHARES. Relation to older prototypes
410 sqq. Alexander 408. Apollo Belvedere 328.
4085-^,7. Artemis 328. 409. Ganymede 408. 410.
Zeus 190, 3.
LYSIPPOS. Date 302, 4. of Sikyon 225. Character
of arts 363 seq. Motive 300 seq. 348. 363 seq. In
fluence of Skopas 300 seq. 363. L. and Polykleitos
363 seq. ; and Euphranor 363 seq. ; and Polykleitos
the Younger 224, 2.
Works: Apoxyomenos 300 seq. 304. 338. 341, 1. 351
seq. Athletes, standing still 364 ; in movement
364, 2. Alexander with spear 364, 2. Eros
394-seq. Herakles 301, 1. 341, 1. Hermes 290. 300.
338. Kairos 396. Warrior (Alexander?) 301, 1.
Pulydamas 302, 4. Troilos 302, 4. Falsely attri
buted to 96. 304 seq.
EUPHRANOR, painting of Cavalry Engagement in
Stoa Basileios 348.
POLYGNOTOS 36. 41. 43, 8. 44. 303.
PANAINOS 40 seq. 48.
PARRHASIOS 34. 34§-
EUPOMPOS 256. 330, 3.
APELLES and Phryne 318.
PROTOGENES 330.
ZEUXIS 348. AFTER ALEXANDER.
PERGAMENE ART. Female torso 392, 3. Female
head 398 seq. Sappho adapted 66, 2. Athena
after fifth-century original 7. 16, 5. 27.
RENASCENCE IN SECOND CENTURY B.C.
399. Eubulides 141, 2. 144. 146.

II. MUSEUMS

ATHENS
Akropolis : Athena on relief, severe style 14, 5. 20
sqq. 25. 144, 7- Statue of Athena, severe style 20
sqq. 25. Barbarian statue, archaic 35, 5. Bronze
female head 52. Bronze statuette of boxer 246.
Charites, relief 20. 23, 1. 29. Statue of boy (style
of Kritios) 7. 19. S3 sqq. 'Oil-pourer,' relief 260,
5. Sandal-binder 364, 2. Armchair 441. 'Altar
to the Four Gods ' 439.
Central Museum. Amazon from Luku 138, 2.
Aphrodite, torso 319, 1. Apollo, archaic statues:
so-called 'on the Omphalos' 81 seq. 175. 182. 190.
298 ; so-called of Cassel, torso 191, 1 ; head 191,
1. Asklepios, beardless 277, 5. Athena of Epi
dauros 412. Athlete of Skopasian style 162, 2.
Doryphoros motive, Argive relief 230. Eleusinian
relief 336, I. Eubouleus, head 103 seq. 333 sqq.
340. 344 sqq. Monument of Eubulides 144.
Hermes from Aegion 289, from Atalanti 291, 6.
Kanephoroi of Praxitelean style 337, I. Boy,
severe style, from Olympieion 196, 1. Head,

Skopasian, from south slope of Akropolis 309. 334.
344. Grave-relief from Ilissos 296. 302. 307. In
fant Ploutos 296, 1, 2. Poseidon from Melos 376.
Altar of the Twelve Gods 431, I.
Polytechnicon : Athlete, bronze statuette 265.
Peiraieus : Head of athlete 296, 6.
BALE
Apollo Steinhauser 409. Head of Diskobolos 168, 3.
BERLIN
Sculptures : Anakreon 60 sqq. Pheidian Aphrodite,
statuette 249. Statue 249. 400, I. Apollo, Sa
bouroff bronze 184. 352 ; marble statue 184. 299,
3. 407, 5; resting 337, 2; torso, 'Adonis' type
354, 4. Artemis, statue 324, I ; head, replica of
Gabii type 323, 5. Asklepios or Zeus 188, 3.
Asklepios, Pheidian 277, 2. ' Aspasia ' head 81,
1. Athena from Pergamon 7. 16, 5. 27 ; Skopa
sian 305, I ; head ibid. ; replica of Parthenos 100,
2; replica of Ath. Giustiniani 359, 4. 361, 5;

478

INDEX

Praxitelean 326, 2 ; replica of Ath. Velletri 141, 2.
athlete, Lysippian 364, 1. Bust of Caracalla 156.
' Demeter ' with veil 82. Dionysos, Polykleitan
school 272, I. Diadumenos, small torso 239, 5-
Diomede, statuette 150, 3. Diskobolos of Myron,
head 148 seq. Eros of Centocelle, replica 316, 5.
Herakles, head of severe style 82 ; head, Poly
kleitan 234, 1. Hennes 289, I ; do. 289, 3 ; do.
by Antiphanes 289. Youth, head 291, 8. Replica
of hero Riccardi 168. Replica of 'Dresden boy,'
head 266, 3. Boy crowning himself, two torsos
251. Female head, style of Alkamenes 85. Fe
male head from Pergamon 398. Meleager 184, 2.
' Narkissos ' 272, 4 (a), (11), (p). Female portrait-
statue, Roman 328, 2. Relief of Polystrata 224, 4.
Relief from Sparta 209. Reliefs of dancing women
438, 3. Satyr pouring wine 312, 1. 'Sappho'
head 66, 2. Head of strategos 120 seq. Zeus,
head 186. 190, 2.
Antiquarium : Bronzes : Statuette of youth, Argive
226. 285, 3. Hermes, Lysippian 300. 338, 2.
Athlete, Skopasian 303, 8. Portrait, Hellenistic
155. Apollo Greau 353. Asklepios, forgery 208,
5. Terra-cotta : Aphrodite from Myrina 384.
Gems: Cameo, head of Eros 69. Paste, dancing
woman 319, 3. Aphrodite, Hellenistic 380. Vases :
from Magna Graecia 109.
Collection Pourtales : Torso of Doryphoros 428.
Collection Kaufmann : Head of Knidian Aphro
dite 322, 3. 343, 5.
Schloss Glienecke : Head of Athena 305, 1.
BOLOGNA
Museo Civico : Head of Lemnian Athena 2. 7. 13

sqq. 49 sqq.

BONN

Double terminal portrait-bust 309, 3. Praxitelean
Eros, bronze statuette 314, 2. 340.
BRUNSWICK
' Sappho ' head 66, 2 (b). Mercury, bronze 232, 3.
BRESCIA
Head of Athena, Pheidian 90. Head of youth, se
vere style 175. Victory, bronze statue 386.
CASSEL
A polio 49. 190 sqq. Artemis 324, I. Asklepios or
Zeus 188, 3. Athena Lemnia 5. 361, 4. Head of
Diadumenos 240. Boxer 245 seq. Hephaistos 88
seq. Youth 354, 2. ' Narkissos ' 272, 4 (q).
Oil-pourer 260, 3.
CARLSRUHE
Aphrodite from Myrina, terra-cotta 384, 2. Youthful
Asklepios, bronze 300, 2. Forged bronze, resem
bling Diadumenos 247. Doryphoros motive 231.
' Narkissos ' 272, 4 (g).

CATAJO
Head of Diskobolos 169 ; of Hermes 300, 4 ; statue
of youth 353, 2. COLOGNE
Mask of Medusa 156.
CONSTANTINOPLE
MUSEUM: 'Sappho' head 66, 2(q). Zeus, bronze
342-
Coi l. of Sir Edgar Vincent ; Boy crowning
himself, head 251. COPENHAGEN
Coll. Jacobsen (Glyptothek Ny Carisberg) : head
of Amazon. Addenda 2. Anakreon 60 seq. 149, 2.
Aphrodite, head 392. Apollo, 'Adonis ' type 314,
4. Artemis, torso 328. Head of Athena (Pro
machos ?) 100. Head of athlete, bearded, Lysippian
364, 2. Eros of Centocelle, replica 316, 5. Hera
Borghese 84, 1. Herakles, bronze statue 299, 3 ;
looking upward 341, 1. Head of youth, early
Pheidian style 55 ; of Euphranor 359, 3. Female
head, Pheidian 66, I. Head of Meleager 304, 3.
' Narkissos ' 272, 4 (r), (c). Paris 359 seq. Attic
grave-relief 408, 10. Polykleitan torso 238, 1.
Female torso, Pergamene style 328, 6. Motive of
Venus of Capua 385, 1 (e).
Royal Museum : Hermes, bronze 232. Terra-cotta
relief with figure from Parthenon frieze 431, 5.
CORNETO.
Two replicas of ' Sappho ' head 66, 2 (/ and m).
DRESDEN
Aphrodite, bronze 388 seq. ; with Triton 305, 3 ;
head of 345. Ares Borghese, replica 148. Artemis,
Praxitelean 324. Asklepios ('Zeus') 55. Athena
Lemnia 4 sqq. Head of Athena, cast 73 sqq.
Athlete statues, Lysippian 364, 2. Athlete head,
Polykleitan 282, 8 ; from Perinthos 1 70. Dia
dumenos head 240 sqq. Head of Diomede, cast
148. Dionysos, Polykleitan school 272, 1. Tripod basis
441. Herakles torso 248, 4. 299; head, Polykleitan
234, 1 ; resting 274, 2; looking upward 344, 1 ; bronze,
cast of 283, 3. Hermes with winged head 363.
Youth (style of Euphranor ?) 312. Boy, Poly
kleitan 266 sqq. ; crowning himself 251. Oil-
pourer 259, 4; do. 259. Palladion (' Dresden Pallas ')
439. Phaon, so-called 69, 3. Infant Ploutos 296,
1, 2. Satyr pouring wine 310 sqq. Motive of
Venus of Milo 380 seq. Kourotrophos on votive
relief 332, 1.
EDINBURGH
Statue from Kyrene 278 seq.

INDEX

479

ENGLAND
Private Collections: Blacker: Diadumenos, terra
cotta 239, 8.
Broadlands (Hampshire) : Athena of Velletri,
replica 141, 2 (a). Head of Hermes 57 ; of
Herakles, Polykleitan 234, 1.
Deepdene in Surrey (Hope collection) : Athena 73,
4. 76 sqq. Dionysos 323, 5. Resting Satyr after
Praxiteles, head 329, 5.
Ince Blundell Hall (Lancashire) : Head of
athlete 172. Dionysos, Polykleitan school 272, 1 (c).
'Theseus '94, 1. 359, 3.
Lansdowne House (London) : Amazon 134 seq.
Apollo, ' Adonis ' type 354, 4. Athena of Velletri,
replica 141, 2 (d). Statue of athlete 171. Marcus
Aurelius 92, 4. Boxer 245, 3. Statue of Herakles
296 sqq. ; of Hermes, Polykleitan 288 sqq. Head
of Hermes 290. Belvedere Hermes, replica 338,
I. Paris 358 seq.
Petworth House (Sussex) : Amazon 130, 138.
Head of athlete with fillet 161 sqq. Oil-pourer
257 seq. Aphrodite, Praxitelean 343 sqq. (removed
to Lord Leconfield's London residence).
Richmond (Collection of Sir Francis Cook) : Apollo,
'Adonis' type 354, 4. Hermes bronze 233, 1.
Vase with Blinding of Polyphemos 109.
Belonging lo A. H. Smith, Esq.: terra-cotta relief
with figs. 133, 134 of north frieze of Parthenon,
431. 5-
Presumably in England: Diomede 148. Athena
73, 4. Polykleitan boy 266, 1.

FLORENCE
Uffizi : Aphrodite and Ares 384, 6. 392, 7. 394, 1.
Asklepios, severe style 205 sqq. Athena 305, 1 (a).
Athlete (apoxyomenos) 205. 260, 3. 261, 1. 515.
519. Doryphoros, statue 228, I ; torso 228; head
229, I. Boxer (?) 247, 1. Hermes, Praxitelean
338, 4. Leda 392, 394. Niobids 394. Venus of
Medici 345.
Museo Etrusco : Chimaira 134, 5. Idolino 283
sqq. 290. 350 sqq.
Palazzo Pitti: Aphrodite, Knidian 322,3 (5). Apollo,
severe style 81. Athena, Giustiniani type 359, 4 (d).
Doryphoros head 229. Herakles 296, 3. Hermes
289, 2. Praxitelean head (Triptolemos ? ) 334, 4.
Oil-pourer 259, 9. Supposed ' Sappho ' 69, 7.
Giardino Boboli : Aphrodite head 345. Athlete
(' Harmodios ') 171. ' Hera,' replica of Capitoline
type (= Ov. K. M. 461, 6), cf. p. 82. Hermes
and Dionysos 230. Female head, Myronian 202.
Motive of Venus of Capua 385, 1 (d).
Palazzo Cepparelli : Draped female statue 69.
Palazzo Riccardi : Anakreon 60, 7 (^). So-called
'Eubouleus' 331, 1. Bust of hero, Myronian
165^17. ' Sappho' head 66, 2 (e). Head of vic
torious athlete with fillet 161, 3 (c). 169, 5. Head
of Zeus 190, 2.
Palazzo Vecchio : Hermes, Belvedere type 338, 1.

GENEVA.
Ares, bronze 230. Vase from Magna Graecia, after
prototype of Pheidian style 109, 5.
GENOA.
Palazzo Reale : ' Narkissos ' 272, 4 (i)
HANOVER.
Kestner Museum : Pan 270, 1 (i)

Pan 270 seq.

LEYDEN. LONDON.

British Museum : Apollo statue, bronze from
Lower Egypt 353 ; head, of Pourtales 165, I ; do.
from Baths of Caracalla ibid. ; do. ('Alexander')
411 ; do. resting 337, 2. do. Kithairoidos 305, 6.
Ares, bronze 230. Aristaios (?) statue from Kyrene
277. Athena, bronze of severe style 16,1 (1). 23,11;
head of Velletri type 141,2 (e). Athlete, Westmacott
250 sqq. Athlete statuette, Polykleitan 265.
Diadumenos, Vaison 238 sqq. 242. sqq.; Farnese
244. Diskobolos 161, 4. Eros, so-called Elgin 256.
Sculptured drum from Ephesos 301 seq. Herakles
bearded, Myronian 178. Aberdeen head, Praxi
telean 346 seq. Hero, statue of 298. Kanephoros,
Praxitelean 337, I (c ). Bearded head (Asklepios ?)
210; another 205. Mercury, bronze from Gaul 232.
Disc with relief of the Niobids 43, 4. Female
statue from Ostia (' Townley Venus ') 319^. Two
statues of Pan 270, 1 (b), (c). Perikles, terminal bust
177 sqq. Pheidias's portrait on shield 48. ' Sappho '
head 66, 2 (c). Xanthos, three torsos of severe
style 23, 5. Nereid monument 450. Zeus, bronze
statuette 299.
Private Collections : supra, ENGLAND.
LYONS.
Hypnos, two small bronze replicas 395, 8.
MADRID.
Athena, statue related to Lemnia 27.
Athena head, Velletri type 141, 2 (/).
Birth of Athena, relief 463 seq. Diadumenos 240 seq.
Diomedes (?), head 149, 3. Dionysos 337. Ude-
fonso group 257. Female head 60. ' Sappho-
Phaon,' terminal bust 66, 2 (p ). Venus of Capua
motive 385, 1 (b).
MANTUA.
Apollo 19. 25. 52 seq. 194. 197. 354, 5- 359, 4 (A
Eubouleus 331, 1. Narkissos 272, 4 (e), (h)
MUNICH.
Glyptothek : Aphrodite 322 and note 5. Apollo,
Barberini 88. 305, 6 ; head of, ' Adonis ' type
354, 4 (placed on the 'Jason'). Artemis of
Gabii 52, 4. 439 seq. 'Artemis' (Tyche) 325.
Athena 60 and note 6; Albani 141, 2 sqq.
Athlete, black marble 264, 1 (B). Diomedes 146

480

INDEX

sqq. 341, 2. Eirene 295 seq. Faun of Winckelmann
359. Youth, head of 81. Boy, bronze head, 291.
Medusa Rondanini 156 sqq. Oil-pourer 204. 259
seq. 287. 296. 312. Paris, head 358. Head of
strategos ('Perikles') 118, 1. 122, 2. Zeus
statue 212 sqq.
Antiquarium : Zeus, bronze 299.
Brunn, belonging to : Apollo, Cassel type, head
191. 1 (7)- NAPLES.
Amazon, bronze terminal bust 138 seq. ; bronze
statuette 140, 1. Antinous 353, 5. Aphrodite,
torso 359. Apollo, bronze statue 52, 6 ; head of
Cassel type 191, I (8); do. on gladiator 411.
Archidamos 321, 1. Ares, Ludovisi type 304, 5.
Asklepios 89, 6. Aphrodite (' Kallipygos') 395.
Athena, Farnese 73 sqq. 82. 141. 144 ; Giustiniani
type 359, 4 (e) ; head of Pheidian 13, I ; do. from
Herculaneum 60. Athlete, bronze head, Poly
kleitan 283 ; Athlete type used for Roman portrait
282. Dionysos and Eros 311, 2. Doryphoros statue
228, 2. 233, 7. Head of, by Apollonios 229. Electra
351. Eros Centocelle type 315. Flora Farnese 323.
' Gladiator ' Farnese 1 25 seq. Harmodios 148. Hera
Farnese 144, 2. 223, 1. Herakles, head, Polykleitos
95, I. Hermes, statue 289 ; statuette 338, 5. Boy
victor 291. Two bronze heads, Skopasian 296.
' Narkissos ' 272, 4 (0). ' Psyche ' of Capua 395.
Portrait in the Doryphoros motive 231, 3 ; in the
Diadumenos motive 247, 2. ' Sappho ' head 66, 2
(n). 68, 2. 'Sappho,' Albani replica 68, 2.
Dancing Satyr 410, 2. ' Dancing girls ' of Hercu
laneum 24. seq. 291, 7. Vases 109 sqq. Venus of
Capua, 384 sqq. OLYMPIA
Sculptures of the Zeus temple 12, 3. 16. 20 seq.
23. 60. 62. 90, 3. 171. seq. 175. 188. 291, 4.
302. 457. Aphrodite, small head 322. 345.
Apollo (?) 172, 3. Asklepios 55, 6. Bases of
Polykleitan statues 224 seq. 249 sqq. 256. 262 sqq.
279 sqq. 308 ; of the works of Sophokles 307^.;
of the Zanes 249. 308. 343. Basis in form of astra
gal 249, 2. 256. Buildings of the fourth century
308. Basis of Smikythos 214, 4. Ergasterion of
Pheidias 39. Hermes of Praxiteles 307 sqq. 337
sqq. Victor statues, iconic 128, 2. Olympian
games, institution of, on vase 1 10, 2. Torso,
Argive type 215, 2. Doryphoros, copy 228. Zeus
of Pheidias 36 sqq.
OXFORD
' Sappho ' head 66, 2 (r).
PALERMO
Hermes, head 290, 6 (b). Pan, head 270, 1 (/;). Satyr
pouring wine 312, I. Selinos sculptures, Aktaion
metope 223, 1. PARIS
Louvre': Marble sculptures: Amazon 132. Anak
reon 60, 7. Aphrodite 'Genetrix' 19. 82. 275, 10.

323 ; of Knidos 322, 3 (1), (10), (n) ; of Kos 323
seq.; head 391, 1. Apollo of Cassel type 191, 1 (2),
191 seq., torso 191, 1 (2), head 191, 1 (5). A.
' Mazarin,' Mantua type 52, 6 ; at rest 337, 2.
Torso, Argive 52, 2 (a) ; early Praxitelean 314.
Ares, Borghese 89^. 148. Head of Ares, Phei
dian 92 seq. Asklepios or Zeus 188,3(3). Head
of Asklepios (?) 210. 'Aspasia' head 81, 1 (b).
Athena, Farnese type 76, 3 (b) ; of Velletri
141 sqq. Diadumenos, torso 239 ; head 240.
' Diane de Gabies ' 323 ; ' de Versailles ' 328, 409.
Diomede 146 sqq. Dionysos resting (' Bacchus de
Versailles') 337, 4. Dioskouros 231. Eros of the
Palatine 313 seq. Boxer, severe style 171 seq.
' Germanicus ' 57, 2. Genius Borghese 336.
Herakles, Skopasian 301 ; severe style 82 ; similar
to the Ares Borghese 298, 3. Hero, torso 298, 1.
Hermes head 290, 5- Two statues of a youth with
foot supported 303, 6. ' Jupiter de Versailles '
(Poseidon?) 104. Meleager, torso 304, 3. 'Mer-
cure Richelieu ' 289 seq. ' Miltiades ' head, Pheidian
36, 1. 90. ' Narkissos ' 272, 4 (/) and (t). Nike of
Samothrake 310. ' Oil-pourer' 261, 1. 'Sappho'
statue 68, 2 (1); head 66, 2(0). Satyr from the
Palatine 330. Torso allied to the Lemnia 26 seq.
Tyche 325, 1. Aphrodite (Venus d'Arles) 319 seq.
Aphrodite from Melos (Venus de Milo) 367 seq.
Venus Falerone 385, I (a) ; and Mars 384, 6.
Venus of Capua motive 385, I (c). ' Virgil ' 102
seq. Zeus statue 188, 1. Zeus, Chthonic 212, 1.
Zeus Talleyrand 42.
Bronzes and the smaller arts : Apollo, Argive style
('Achille') 52, 4. Athlete, Polykleitan 279. With
fillet (?) 278. Dionysos, supposed Praxitelean
337,5. 354, 1. Wooden female figure, severe style
23, 6. Head from Beneventum 290 seq. Mercury
232. Vases : Krater with Expiation of Orestes 109-
Bibliotheque Nationale : Bronzes : Aristaios (?)
276 seq. Diadumenos 239, 7. Four Mercury
bronzes 232, 3 ; 233, 2 ; 233, 4 ; 233, 6. Hydria
(frieze of Argonauts) 109. Lekythos 124.
Ecole des Beaux-Arts : Torso Medici 27 sqq.

Eros torso 313, 2.

PARMA.

POMPEII

Stucco relief, Eros 316, 9. Wall painting, athlete
162, 2.
ROME
Vatican : Galleria Lapidaria: Herakles 298, 2.
Boy victor, Dresden type 266, 2 (b). Torso of boy
3°?. 3-
Braccio Nuovo : Artemis 57 ; do. 104, 4. Asklepios,
beardless 300, 3. Athena, Giustiniani 359 sqq.
Athletes 260, 3. 262, 1. 264, 1. (A). 302, 3. Horned
river-god (?) 46. 55. 81. 102, 3. Medusa mask
156. Satyr resting 329.
Museo Chiaramonti : Aphrodite of Knidos, statuette
322, 3. Apollo, Argive type 52, 2 (b). Artemis,
torso 329. Asklepios 188, 3 (6). Athena, Praxite-

INDEX

lean 326, 2. A. head, Skopasian 305, 1. Charites
relief 20. 25, I. 29. Doryphoros, torso 228, 7 ;
head 229, 1. Hephaistos head 89. Hera head 84,
1. Herakles with Telephos 341 seq. Lysippian
363, 5. Head, Polykleitan 234, 1. Hermes,
head, Polykleitan 290, 6. Head, Dresden boy
266, 3 (b). Boy victor 291, 5. Youthful god
102, 4 (b) ; bearded, Pheidian (?) 64 ; Myronian (?)
204. ' Narkissos ' 272, 4 (b). ' Niobid ' 394.
' Phaon ' 69, 3. ' Sappho ' 66, 2 (/). Satyr
pouring wine, head 310, 2.
Giardino della Pigna: Torso, Venus of Capua type
385, 1 (*)•
Belvedere and Cortile : Apollo 328. 405 sqq. Athena
'Hygieia' 15, 2. Doryphoros, allied to the 231,
3. Herakles, torso 399. Hermes 238. 338. Me
leager 184.
Galleria delle Statue : Amazon 132. 137 sqq. Athena
Lemnia, head 6. Another A. head 326, 3.
Athlete, Polykleitan 281 seq. Augustus, statue
156, 1. Eros Centocelle 315 sqq. Menander 309,
3. Mercury Ingenui 182 seq. Neptune 184 seq.
Paris, seated 357, 5. Penelope, head (Diadumenos)
245, I. Poseidippos 309, 3. ' Septimius Severus '
282, 2.
Gabinetto delle Maschere : Apollo, ' Adonis ' 354 sqq.
Loggia Scoperta : Hermes 165, 2.
Hall of the Muses: Apollo Kithairoidos 305. Poet,
head 63. Perikles 117 sqq. Themistokles 122, 2.
Rotunda : ' Ceres ' (Nemesis ?) 88. Hera Barberini
82. Herakles, bronze gilt 296, 3. Zeus Otricoli
190. 342.
Sala a Croce greca: Aphrodite of Knidos 322, 3.
Athlete, Polykleitan 282, 3.
Sala della Biga : 'Alkibiades' 126 seq. Diskobolos
173, 4. Diskophoros 90. 137. 287.
Galleria dei Candelabri : Eros 316, 4. Hermes 338,
5. Youth, Polykleitan 238, I. Pan 270, 1 (d).
Satyr pouring wine 302, 3. Head 310, 2.
Sala Geografica : Hermes head 290, 5 (b)
Vestibule : Athlete, Polykleitan 282, 3. Doryphoros
head 229, 1.
Magazino : Aphrodite of Knidos 322, 3 (1).
Museum of the Capitol : Ground floor : Askle
pios or Zeus 188, 3. Athena (Agorakritos ?) 88.
Athlete, head bound with thongs 204, 6.
Upper floor : Amazon 132 sqq. Antinous 353, 5.
Aphrodite head 392, 4 ; of Knidos 322, 3 (13).
Apollo statue, severe style 49. 55. 197, 2. Ter
minal bust, severe style 55. Athena Giustiniani
replica 359, 4 ; a terminal bust 60, 5 (3). Well
head, archaistic 441. ' Demeter ' 82. Diskobolos
173, 4. Eubouleus 331, I. 'Giunone' head,
Skopasian 392. Hermes, foot supported 303.
' Hygieia ' allied to Pheidias 27. Youth, statue of
363. Female head, Pheidian 66, I. Bearded head
with helmet 90, 2. Portrait-statue, seated 309, 3.
Satyr resting 329, 5. Venus and Mars 384, 6.
Conservatori Palace : Amazon, head 132, 1.
Anakreon 60, 7 (a). Apollo, resting 337, 2. Ares
(supposed Eros) 271, 1. Athena, Velletri type
141, 2. Eros, Centocelle type 316. Camillus 20.

Head, Riccardi type 165, 2 (a). Runners 128, 1.
Pan 270, I (e). Paris 357, 5. Charioteer 81.
Lateran : Ares 92. Herakles, Polykleitan 234, 1.
Hero, head 298, 1. Youth, torso of (Pythagoras ?)
'72> 3- Head of boy crowning himself 252, 3..
Head of statue by Stephanos, replica 214, 1.
Marsyas 180 seq. Pan, head 270, I (k). Torso,
allied to Parthenos 27 ; allied to Doryphoros
231, 3-
Museo delle Terme : Apollo, Pheidian 49. 197.
Asklepios head 89, 6. Athena, Olympian style
16, 8. 23, 8. Head, Giustiniani type 359, 4 (/;).
Athlete head 296, 6. Diomedes, statuette of 149, 3 ;
torso of ibid. Dionysos of Tivoli 350 sqq. Dios
curi on sarcophagus 100, 4. Eros, Centocelle
type 316. Hera, Barberini type 84, 2. Hermes of
the Palatine 300 seq. Boy victor, statue 291, 6 ;
derived from the Polykleitan boy 255, 6 ; Pan,
head 270, 1 (/).
Villa Albani : Amazon, head 132, 1. Ares or
Hero 230, 3. Artemis, statue 26. Relief 329, 3.
Athena with skin helmet 55. 78 seq. Replica of
the Farnese Athena (now lost) 76, 3 (d). A. head,
bronze 78. Herakles 340. Hermes or portrait
231, 3. Kanephoroi, Praxitelean 337, 1 (b). Boy,
early Pheidian, crowning jhimself 251, 4. 255, 4.
Head, youthful 'Alexander' 103, 1. Female
head, early Pheidian 60. Another, idealized
portrait 328, 3. 'Lysias' 81. Paris, head 358, 3.
'Peisistratos' 175 seq. 'Sappho,' statue 68, 2.
Head 66, 2 (z). 69 seq. Sauroktonos 336, 2. Stra-
tegos, head 122, 2. Venus Torlonia 384, 5 (a).
Museo Torlonia : Aphrodite of Knidos 322, 3
(3), (4). Ares, head 92, 2. Asklepios 188, 3 (9).
Athena, Giustiniani type 359, 4 (c). Athlete,
Polykleitan, torso 282, 4, head 282, 6 ; head, replica
of Florentine athlete 262, 1. Diadumenos type, in
fluenced by 247, 2. Dionysos, severe style 42.
Doryphoros head 229, 1. Eirene 296, 1. Hestia
Giustiniani 81. Bearded head, with winged fillet
63 seq. Female head, Pheidian 66, 1 (b and c).
Boy crowning himself, torso 251, 3. Head of
252, I, modified 255, 3. Dresden boy, replica
266, 2 (a). Torso, Skopasian (?) 302, 3. Medusa
Rondanini, two replicas 156. Meleager 304, 3.
Oil-pourer 259, 9. ' Phaon ' head 69, 3. Torso
299, 2. Wrongly interpreted as Satyr pouring
wine 310, 2. Tyche 325, 1.
Villa Borghese : Artemis 324, 1. Athena 89, 3.
326, 2. Meleager 184, 2. 'Narkissos' 272, 4 (d).
Oil-pourer 259, 9. ' Sappho ' heads 62, 2 (g and
h). Torso, Argive type 299, 1. Venus and Mars
384, 6.
Museo Boncompagni-Ludovisi : Aphrodite, of
Knidos 322, 3 (6). Apollo head 337, 2. Ares 304.
' Hera ' head 326 seq. Terminal figures, series of
248, 4. 299. 357. Hermes 57- ' Sappho ' 66, 2
(r). Female, torso, severe style, 144, 4.
Museo Kircheriano : Asklepios 188, 3 (7). Fico-
roni Cista in.
Mr. Abbati's At.
head 161, 3 (a).

Petworth athlete, replica of

3 Q

482

INDEX

Pal. Altemps : Asklepios or Zeus 188, 3. Herakles
seated 202 seq.
Pal. Barberini : Asklepios and Hygieia 206 seq.
Youth, Idolino type 287. Head, bearded 55.
' Narkissos ' 272, 4 (in). Suppliant 20. 395.
Barracco Collection • Amazon 132, 1. Aphro
dite, head of 391, 1. Apollo, Cassel type 191, 1
(6). Resting 337, 2. Athena head 438, 2. Dia
dumenos, head of 240. Doryphoros, head of 229,
1. Youth, Polykleitan (Hermes?) 238, 1. Boy
crowning himself 250. Dresden boy, replica of
head 266, 3 (c). Head, Pheidian 57. Marsyas
181, 4. Perikles 118. ' Sappho' 66, 2 (k).
Pal. Borghese : Ares 92 seq. Herakles 298, 2.
Pal. Caetani : Aphrodite, head of 389 sqq.
Caracalla Thermae : Torso, Polykleitan 238, 1
(c).
Monte Cavallo : Dioscuri 95 sqq.
Pal. Chigi : Apollo (' Adonis ') 354, 4.
Pal. Colonna: Amazon 135, 1. Athlete, Poly
kleitan 282, 4. Hermes, Polykleitan 289. Head,
with helmet 90, 2. ' Narkissos ' 272, 4 (k). Zeus
head (Myronian) 190, I (b).
Pal. Doria : Meleager 304, 3. ' Narkissos ' 272, 4
(/).
Pal. Giustiniani : Apollo 165, 1. Diadumenos,
torso 239. Head 242. Doryphoros 228, 7. Ask
lepios 188, 3.
Pal. Massimi alle Colonne : Doryphoros, two
torsos 228, 7. Herakles 296, 3.
Pal. Mattei : Apollo resting 337, 2. Oil-pourer,
restored torso 259, 4.
Villa Mattei : Statue related to Parthenos 27, 5.
Doryphoros 229, 1. Torso, Polykleitan 238, 1 (a).
Villa Medici : Head of Meleager 304, 3. 338, 3.
Aphrodite of Knidos 322, 3 (9). Torso, Praxi
telean 338, 3.
Monteverde Collection : Eros, Centocelle type
316, 2.
Pal. Odescalchi : Aphrodite 400.
Villa Pamfili : Amazon 129.
Pal. del Quirinale : Boy crowning himself 250.
Casino Rospigliosi : Athena 305, 1. 'Narkissos'
272, 4 (c).
Pal. Sciarra : Tyche 325, 1. Amazon 132, 3. Cf.
Addenda
Pal. Torlonia : Amazon 135, 1. Athena, head of
76, 3. Youth, head of 81, 3. Meleager, torso
(restored as Hermes) 304, 3.
Pal. Valentini : Aphrodite 400. ' Diomedes ' 204.
Doryphoros 229, 1, modification of (Pertinax)
231. 3-
In the Market : Diadumenos, head 240. Herakles,
Argive type 357, 4.
ST. PETERSBURG.
Hermitage : Amazon 132, 1. Aphrodite, Hellenistic
392,6. Apollo, 'Adonis' type 354,4. Artemis

bust 329. Athena, Velletri type 141, 2 (c).
Giustiniani 159, 4 (g). Athlete head, Polykleitan
282, 7. Florence athlete, replica 262, 1. Athlete
restored as Mercury 268 seq. Doryphoros head
229, 3. Gem with Hippolytos 248. Hermes,
winged Polykleitan 290, 4. Female head, early
Pheidian 57 ; later Pheidian 104, 4 ; head, bearded
Myronian 178 ; of the boy crowning himself 252 ;
Niobid relief 43, 4 ; Parthenos, allied to the 27.
Vase with Adonis 275, 10. Attic krater no, 2.
Tarentine 109 seq. Zeus or Asklepios statue 186
sqq.
Stroganoff Collection: Aphrodite head 391, 1.
Apollo bronze 405 sqq.
Pawlowsk : Eros 315, 4.
SARAGOSSA
Apollo, bronze, forgery 407.
SMYRNA.
Female head from Tralles 398.
SPIRES (Speier).
Mercury, Polykleitan style 283, 2.
STOCKHOLM.
Female head 84, 4. TREVES.
Athena with skin helmet, two replicas 80, 1. Pet
worth, replica of head 161, 3 (b). Venus of Milo
motive 380. TURIN.
Amazon 138, 2. Asklepios and Hygieia 385, 3.
Athena 326, 2. Diadumenos 239. Eros Cento
celle, replica 315. Head, youth with fillet 247.
Hypnos, small bronze 395, 8. Oil-pourer 260, 3.
Marble seat with reliefs 162, 2.
VENICE.
Torso, supposed from the Parthenon 456.
VERONA.
Amazon, forged, bronze 137, 1.
VIENNA.
Aphrodite, head from Tralles 398. Artemis from
Cyprus 326. Doryphoros 228, 7.. ' Hera ' of
Ephesos 84, 4. Herakles, bronze 299. Youth,
from Carinthia 290. Mercury, bronze 233, 5- Pan
270, I (g). Venus torso 384, 5 (b).
WORLITZ.
Amazon 132. 132, 1. 136, 4.

ZURICH.

Hermes, bronze 1 55.

INDEX

4§3

III. GENERAL

Achilleae statuae 233, 7.
Adonis 255. 275.
Aegis, shape of, in fifth century 10. 78. 80. 141 seq.
Transverse 16.
Aegis, scales on helmet 60.
Aegis, Medusa type 158, 3.
Aeschylos, portrait 304, 3.
Agathe Tyche 349.
Agothodaimon ibid.
Agrippina 326, 7.
Akanthos 437 seq. 450.
Alexander, portrait by Lysippos 301, I. 364, 2 (Alex
ander Balas).
Alkibiades 432 ; supposed statue of 127.
Amazonomachia, on shield of Parthenos 44. On vases
109.
Amazons 128 sqq. Legend of 131. Polykleitan type
128. 130. 134. Capitoline 128. Mattei 30. 130.
132. 136. 140; by Phradmon 128. 130. 140 ; by
Strongylion 140.
Anakreon 39. 117, 4. 149, 2.
Anecdotes of artists 21, 2. 84.
Antinous, statue 353. Head, Mondragone 18.
Aphrodite, nude archaic 387, 1 ; nude later art 387 ;
half draped 386 seq. ; with shield 386 ; formation of
eyes 345 ; diadem 389 ; fillet ibid. ; terminal
274 ; A. and Ares 385 sqq. 383 ; and Tyche 381 ;
and Nike 386 ; Anadyomene by Pheidias 43 ; other
statues of Pheidian school 71. 249. 399 seq. ; ' Gene
trix' (Alkamenes) 19, 3. 82. 275, 10. 323 ; by
Praxiteles 318 sqq. 343 sqq.; by Skopas 391. 329.
395 ; by Euphranor 358 ; Kallipygos 395. Medi-
cean type 345. 387. 398 ; of Melos 367 sqq. ; of
Capua'384 ; MrjAe/a 382, 2 ; Ourania 71. 274 ; Pan-
demos 394
Apollo : characterization of head 49. 194. 336. Attri
butes 408. Patroos 461 seq. 363. Hypakraios
461, 12. Smintheus 303. 336. 388. Sauroktonos
336. 388. 396. Argive 52. Style of Kritios 53 seq.
197. Mantuantype 19. 25. 51. 52. 194. 195. Early
Pheidian 49 sqq. ; allied to later 49 sqq. On the
Omphalos 81. 175. 182. 190. 196, 2. 197. Cassel
type 190 sqq. Barberini (Agorakritos?) 88. Later
Pheidian 410 sqq. Praxitelean, Sauroktonos 336.
338 ; at rest 337. 314. Head Pourtales 165 seq.
Palatine 305, 6. Kitharoidos, older type ibid.,
Skopasian 305. By Euphranor 352 seq. 354 sqq.
By Leochares 408 sqq. Belvedere 328. 405 sqq.
Stroganoff 405 sqq. Steinhauser 409.
Aposkopeuon, motive of 254, 3.
Apoxyomenos, Attic 261, 2 ; Polykleitan 262 ; Lysip
pian 300 seq. 304. 338. 341, 1. 351.
Apulia, vases 109 seq.
Archidamos 122, 2. 321 seq.

Ares, Pheidian 92^,7. 155. By Alkamenes 89 seq. 95.
Polykleitan 230. 271. Skopasian 304 seq.
Arethas 31,9.
Argive characters 225.
Argos, and Athens 200 ; and Sikyon 225. Sculptures
from Heraion 243.
Aristaios 278.
Aristandros of Paros 302.
Aristeides 421, 422.
Aristodemos, Doryphoros by 164, I.
Aristophanes 445, 6. 449. 451.
Artemis, statues 26. 52, 4. 57. 323. 324. 326 sqq.
Head of, severe style 223, 1. On Parthenon frieze
431. With cap 431, 4. Brauronia 102, 3. 426, 3.
431, 4. 432, 3 seq. 388. Laphria at Patrai 388.
Praxitelean 324 seq. 326.
Asklepios, types in fifth century 55. 89. 188 seq. 277, 2.
277, 5. Myronian 204 sqq. Types in fourth century
300 sqq. Zeus type adapted to Asklepios 188 seq.
Snake without staff 55, 6. 209. With Hygieia 209.
Beardless 277. 300.
Aspasia, so-called 81.
Athena : Old, agrarian goddess 471; ancient olive-
wood image on the Akropolis 419. 426. 427 seq. ;
with pomegranate 445 ; Polias and Parthenos 427 ;
A. Nike 444 sqq ; A. ritual on the Akropolis 430.
466 ; Palladion 433. 437 seq. ; P. robbed by
Diomede 149 sqq. ; A. on prize-amphoras 439, 6,
A. and Poseidon 456 sqq.; A. and Hades 80.
1 ; Alalkomeneis 80, 1. 305 ; Itonia 80, 1. peaceful,
bareheaded type 13 sqq. 445 ; with fillet 15 ;
short, gathered-up hair 16 ; skin helmet 78 seq. ;
and Triton 305 ; A. severe style, with transverse
aegis 16. 23, 8 ; relief and statuette 21 sqq. 25 ; with
roll of hair in neck, on coins 16, I. Athena
Lemnia 4 sqq. ; Parthenos 3. 10 sqq. 33 seq. 46. 47.
49 seq. 105. 107. 143. Promachos 9. 27 sqq. 46.
100. Farnese type 5. 73 sqq. 82. 141. 144 ; Albani
55. 78 sqq. ; Pheidian type, at Naples 60 ; Hope 73
sqq. ; at Brescia 90 ; (by Agorakritos ?) 88. Velletri
type 141 sqq. 188. By Skopas 305. 394. By
Praxiteles 89, 3. 326 ; Giustiniani type 359 sqq. ;
from Pergamon 7- 16, 5. 27. Types on coins
105 sqq.
Athens : (a) Akropolis : ' Old Temple ' 415 sqq. 421.
424. 426. 432. 433. 435. First Parthenon 419 sqq.
424. Periklean Parthenon 25. 46, 5. 419. 423 sqq.
Opisthodomos 425. Frieze, interpretation of east
side 427 sqq. 16. 41. West pediment, interpreta
tion 451 sqq. East pediment 463 sqq. Propylaia
9 seq. 424. 428. 442 seq. Erechtheion 415. 432.^7.
450 seq. Temple of Athena Nike 442 sqq. 449 ;
frieze, interjiretation 445 sqq. Artemis Brauronia,
statue 102, 3. 426, 3. 431, 4. 309? Treasury 425 ?

484

INDEX

426 ? Parthenoi, cultus of 45S sqq. Erechtheus,
cultus of 416. 418. 419. 424. 432. 434. 435. Ke
krops, cultus of 434 sqq. ; how represented 45S
Snake, guardian of Citadel 436. Ge Karpophoros
421, 4. 468 sqq. Quadriga 10, 4. Votive offerings
of Kleruchs 8. Walls, northern 420, 2. 432 ; south
ern 420. 422. 442. (/') Lower City : Stoa Poikile
35. 41. 'Theseion' 46. 99. 182, 5. 447. 448.
Temple on the Ilissos 444, 3. Cultus of Athena
Itonia 80, I. -Cult of lakchos 104, I. Cult of
Demeter in Agrai 444, 3.
Athlete, head bound with thongs 204.
Bases of statues, pillar-shaped 21 ; for bronzes 286.
299, 3 ; with moulded profile 307 seq. ; with
Latin designation of artists 95 seq.
Blood, drops indicated 126. 134, 5.
Boeotian terra-cotta, severe style 23, 4.
Bonus Eventus 349 seq. 353.
Bronze, colour of 233, 6. Forgeries 406. B. style 8.
18. 134, 2. 391.
Boutes, cultus of 435. 462.
Bouzyges 462 seq.
Camtana terra-cotta reliefs 42, 4. 438.
Campania, Attic vases in 408 sqq.
Canova, his estimate of the Colossi of Monte Cavallo,
Addenda i.
Carrey, drawings 451 sqq.
Carts or chariots in agrarian ritual 469 sqq.
' Cauldron chariots' 471.
Charites, relief, severe style 20. 23, I. 29. On frieze
of Nike Temple 448 seq.
Chiusi, armchair in grave 430, 3.
Coins. Transference of Pheidian style to coins 105
sqq. ; die-cutters in Magna Graecia 106. 107 seq. ;
C. of Alexandria Troas 303, 2 ; Allifa 106, 6 ;
Amisos 80, I ; Amphipolis 410^. ; Argos 152, 5
(Diomede), 200 (Perseus) ; Argos and Elis 217 seq.
223; Athens 31. 105. 143. 196, 2. 356; of
Augustus 190, 3 ; of the Brettii 255 ; Campanians
106 ; Corinth 16, 1. 388 ; Crete 116 ; Elis 217 seq.
Eresos 71, I ; Herakleia Luc. 15. 105, 6. 107 ;
Hyria 106; Kaulonia 408. 412; Klazomenai 410 ;
Kleonai 237 ; Krannon 469 ; Neapolis Camp. 106 ;
Nola ibid. ; Pale in Kephallenia 468 ; Pandosia 106 ;
Phistelia ibid ; 107 ; Phokaia 58 ; Poseidonia 107 ;
Rhodes 410 ; Side 16, I (2) ; Stymphalos 237 ;
Kyme 105, 2. 106, 3; Lokri Epizephyrii 218;
Melos 381. 382, 3; Miletos 410, 8; Mitylene 71,
1 ; Naxos Sic. 108 ; Sybaris 105, 2 ; Syracuse 107
seq. 218; Tarentum 15. 107; Terina 106; Thurii
105 seq. ; Troezene 249. 255, 9. ; Athena type of
Thurii, spread of 106 seq.
Copies. Relation between several copies of the same
original, 5. 12, I. 60, 7. 117. 132 sqq. 137, I. 141,
2. 165 sqq. 181, 4. 188, 3. 191 sqq. 198 sqq. 2IO.
212, 2. 229 seq. 234. 238, 3. 239. 240. 249 sqq.
259 sqq. 263 seq. 270, 1. 270, 5. 272, 5. 277 seq.
288 sqq. 296 seq. 304. 310 seq. 314 seq. 319.
354 seq. 359, 4. 384 sqq. Copies and originals

compared 334. 343 seq. Copies in relation to works
known from tradition 49 sqq. 82. Copies made
with help of casts 3. Size of copies 3. 138, 2 ;
varying in case of copies from colossal works 4 ;
reduced because of costly material 138, 2. 336, 2 ;
reduced for decorative purposes, to fit into series of
a given size 260, 3. 262, 1. Distinction between
variations introduced by copyists or different originals
76. 82, 9. 92 seq. 165, I. 212. 240 sqq. 246. 259^.
384 sqq. 389 seq. 409 seq. Additions made by
copyists 55, I. 82. 89, 5. 92. 97. 148. 184. 188. 231.
272, 5. 289 seq. 304. 314. 315. 323. 330. 354, 4.
359, 4. 361. 385. 386. 389. Plastic rendering of
objects put on separately in the original 148. Free
copies of the Hellenistic period 27. 141, 2 (g). In
accuracy, especially in the hair 168. 229. 272, I (a).
409. Adaptations 57, 2. 84, 2. 165, 1. 184. 255.
274. 283. 289 seq. 316. 322. 325. 345. Free
adaptations, in bronze statuettes 231 sqq. 239, 7;
in terra-cotta 329. Copies in Asia Minor terra
cottas 239. Groups composed by copyists 208. 230.
384 seq. Artists' inscriptions on. copies 95 seq. 194.
330. 350, 2. 357.
Corinth, Aphrodite of 388. Puteal 441.
Crete 116.
Delos : court for assembly 378. Purification of 444
Delphi : group ded. by Athenians 35 ; by Lakedai-
monians 37 seq. Phryne 318. 320 seq.
Demeter 24 ; on Parthenon frieze 431.
Diadumenos : Polykleitan 238 sqq. ; Farnese 242, 2;
Petworth type 161 ; descr. by Kallistratos 164, 2.
Diitrephes 122 seq.
Diomede with Palladion 152 seq. ; in Argos 152 ;
' Valentini ' 204.
Dionysos : horned 356, 7 ; short-haired 375. 467.
Progress in cart 469 ; in boat 471. Statue, severe
style 42. Polykleitan 255. 272. Praxitelean 308,
8. 311, 2. 314. 323, 5. 337. Tendency of Euphranor
35° SlM- 35^- On Parthenon frieze 431.
Dioskouroi, typical grouping 100. Of Monte Cavallo
95 sqq. derived from Polykleitan Doryphoros 231 ;
aslragalizontes 292, 1.
Doryphoros, meaning of 151. 162. 228. Polykleitan
138. 141. 151, 5. 163. 226 sqq. 242 seq. Kresilaian
163. By Aristodemos 164, 1.
Eirene 295 seq. 319. 324.
Eleusis, cultus 331 sqq. Relief from 336, 1.
Ephesos, sculptures of second temple 301 ; founding
of temple 131. Legend of Amazons 128 sqq.
Erichthonios 436. 458.
Eros : in Euripides 3 1 7- With arrow 3 1 7. Stringing
bow 394. Pheidian 69. 316, 2. Praxitelean 313
sqq. Palaestric 314, 1. ' Elgin ' 356, 5.
Erysichthon on Parthenon pediment 458. Tomb at
Prasiai 461.
Eubouleus 103. 330 sqq.
Euripides in Magna Graecia 1 10. Erechtheus 436, 5 ;
Ion 461 seq.
Ficoroni cista ill,

INDEX

485

Form : (A.) Body : (a) veins on the abdomen 229. 279 ;
(b) eye 66. 97, 2. 103. 120. 134. 145. 153. 160. 165,
I. 170. 171. 179. 334. 345; (c) eyebrows (hair indi.
cated) 179. 210. 236; (d) nose, Polykleitan 17 seq.
154. 243. 291. in works of Kresilas 120. 122. 154.
160. 164. 180; (c) ear, swollen 92. 149, 5. 155, 2.
165. 173. Form in fifth century 19 ; (/) hair: (1)
method of dressing roll 16. 57. 64. 194. 351. Part
ing 19. From the crown to the front 19. 186. Front
hair waved 19 seq. 57- 105- Crown of hair in
divinities 184. 186 seq. 194. Knot on forehead
316. Plait behind 194. Plait on nape, Roman
326. In Zeus 42. 217 seq. In chthonic gods 212,
In Amazons 136. 138. Short hair in gods 42. 118.
217. 356. 468. Loose wisps 19. 389. 398. Ringlets
55. Kerchief or coif 71, I. (2) Style and technique
8. 309. 334. 344. 391. 399.
(B.) Drapery: (a) costume, Doric peplos in fifth
century 21 sqq. ; girt over the diplois 24. 26. 27 ;
ungirt 24 ; Ionic chiton below 24. 27 ; horseman's
cloak worn by Amazon 136; chiton with short
sleeves 309, 3 ; (b) style : archaic 29. Severe
Peloponnesian 23. Attic 24. In works of Pheidias
25. On the Parthenon 30. 46 seq. 62. 399. On
the Nike temple 450 seq. In Myron's works 188.
In Kresilas's works 134 sqq. 140. 144. 188. Ionic,
second half of the fifth century 450 seq. ; in fourth
century 310 ; on the Niobids 394. Linen and
woollen 29. 46 seq. 134 seq. 144. Transparency
84. 85. 400. 439.
Ganymede of Leochares 408. 410. On a gem 275, 1.
Ge 421, 4. 448, 9. Karpophoros 468^.
Gems : Italian, similar to scarab style 124. Hellen
istic 380. Early Roman 386, 7. Emerald plasmas
with copies of statues 137, 1. 163, 1. 261. 262, 2.
277, 1. Amazon 132. 137. Aphrodite 66,
8. Athena 6. 14, 4. 15, 6. 16, 5. Bonus Eventus
350. Diadumenos head 244. Eros head, cameo
69. Herakles 248. Hippolytos 248. Medusa
201. ' Narkissos ' motive 274 seq. Niobids 43, 8.
' Oil-pourer ' 260. Polykleitan motives 248. 255,
10. Apoxyomenos 261. Victory 386, 7. Paste
of Solon 15, 6. Inscription HEIOT 6, 3.
Genius 231, 3.
Gigantomachia on shield of Parthenos 45.
Glaukos, Argive artist 215.
Gold-ivory statues, difficulty in copying 3 ; technique
of drapery 12, 8.
Grave-reliefs, Attic (Telesias) 275, 5. (Hegeso) 19.
275, 10. Reorganization of 309
Groups : of separate statues 140 (Eph. Amazons). Put
together later 208. 230. 384 seq.
Gymnasia, decoration of 375. 377 seq.
Hades, cap of 80, 1.
Hair. See under Form.
Hebe, on Parthenon pediment 465.
Heiou gem 6, 3.
Hekate of Alkamenes 90, 3. 442.
Hekatompedon 418. 419. 424.

Helios on works of Pheidias : and Selene 45 ; and
Nyx 465 seq.
Hellenotamiai 425.
Helmet, Boeotian 446, 4. Corinthian, 361.
Hephaistos, statue of Argive type 217, 1 ; of Alka
menes 88 ; on east pediment of the Parthenon
465.
Hera of Polykleitos 218. 223. 242; of Alkamenes
82. 84 ; at Plataia by the Elder Praxiteles 102 ;
on the Parthenon frieze 431 ; on the east pediment
465 ; on the Nike temple 449 ; Farnese (so -
called II.) 223, I ; Barberini 82 seq. ; Ludovisi
326.
Herakleia in Lucania : coins 15. 107 ; vases no.
Herakles, of severe style 82; characterization in fifth
century 178 sqq. ; crushed ears 155 ; rolled
fillet 95, 342 (Fig. 147) ; resting 274 ; crowning
himself 255 ; Myronian 178. 202 seq. 236; Poly
kleitan 234. sqq. ; Pheidian 298 ; Skopasian 66,
5. 235. 296 sqq. 301 ; Praxitelean, bearded 340
seq. beardless 346 seq. ; with Telephos 341 ; time
of Trajan 204; terminal busts 66, 5- 235. 301, 2.
supposed entrance into Olympos 441, 6 ; not on
pediment of_Parthenon 467 seq.
Hermes : Homeric conception 340. With winged
fillet 64. 363, 4 ; on coins of Phokaia 58 ; H.
Ludovisi 58. 231, 1. Myronian (?) 184 ; of Poly
kleitan "school 230. 255. 288. 290. In motive 01
Diomede 155, of Doryphoros 231 sqq. Leading
ram 23 1 ; in Roman bronzes 231 seq. 283 ; with
infant Dionysos ; Peloponnesian 230 ; Skopasian (?)
300 seq. 338 ; Praxitelean 307 sqq. 337 sqq.
Lysippian 300. 290. 338.
Hermolykos, son of Diitrephes 122 seq. The Pan-
kratiast 123
Heroes, with snake 209 ; in Asklepios type 277- On
the Nike temple (Androkrates and Theseus) 449
Hestia Giustiniani 23. 81.
Hippolytos, statue 249. 277, 6. On gem 248
Hoplitodrome 204.
Horai, on Corinthian puteal 441. On Parthenon
pediment 467
Horses 97. 230.
Hyakinthides 459, 6.
Hypatodoros and Aristogeiton 41.
Hypnos : Cultus 396. Head-wings 396. Statues 303.
395 seq. On lion 396, 5. Wrongly interpreted as
H. 274, 3.
Iakchos, in group noted by Pausanias 102 ; by Elder
Praxiteles 104 ; confused with Dionysos 104, 1.
With Eleusinian goddess 333. So-called, in
Vatican 55. 81. 102, 3.
Iconic statues of victorious athletes 128, 2.
Idolino 283 sqq.
Inscription on statues : in older period 123, 3 ; in
later 373 seq. 400. Artist inscriptions on copies 95
seq. 194. 357-
Ion, on west pediment of Parthenon 461 seq. 461, 12,
13-
Ionian art : reliefs, style and draperies of 450 seq.
Iris on Parthenon pediments 457. 465.

486

INDEX

Jupiter Tonans in capitolio 190, 3.
Kalathiskos dancing girls 438.
Kallistratos, iKippdtreis 164, 2. 340. 397, I.
Kekrops, worship in Erechtheion 434 seq. On
Parthenon pediment 458 seq.
Kephalos on Parthenon east pediment 468.
Kephisodoros 145 seq.
Kimon 33. 34 sqq. 41. 423- 437- And the fivst Ptl1'-
thenon 420 sqq. Kimonian wall 421. 422.
Kimonian 'treaty' 34, 1.
Kallimachos 437. 441. 45°-
Kore 24.
Kreousa 461 seq.
Krison, the runner 128.
Lakrateides, relief of 333.
Leaping on to horse's back 137.
Leda, statues 394 seq.
Lekythos, slave carrying chair 428, 3. In Paris,
wounded warrior 124
Libation 279. 286.
Locality designated by human figures ; not in fifth
century 458.
Lucania, vases from 109.
Lysias, so-called, in Villa Albani 81.
Maenad of Skopas 397. On reliefs 439.
Marathon, votive gifts for 35 sqq.
Mask-cttltus 160 seq.
Medusa mask : types in fifth century 158 seq. 201.
In Hellenistic period 159. With neck severed 159
seq. 201. With wings 158. 160. With knot of
snakes 158.
Meleager, Skopasian 184. 304. 307.
Melos : find of statues 376. Sepulchral caves ibid.
Gymnasium 377 seq. Tyche of the island 381 sqq.
Menander 309.
Mikon 41.
Miltiades 35. 36, 1. Pinax with inscription of 355.
Mitra 161.
Moirai on Parthenon 466 seq.
Motives : standing motive, of the school of Hagelaidas
49 sqq. 191 sqq. 215. 287. 314 ; transitional from
Hagelaidas and Polykleitos 214 seq. ; of Polykleitos
and his school 215. 227 seq. 242, 248. 249 sqq. 254.
261. 264. 266. Attic 296 sqq. Of Myron 190 seq.
Pheidian: II seq. 25. 27. 30. 137 seq. 244. 296.
319. 324 ; of Alkamenes 84; of Kresilas 143. Of
Skopas and Lysippos 300 sqq. 394. Relation be
tween arm and leg 227. 247. Arm resting against
hip 300. 303. 305. Hand on back 238. 273 seq.
Arm raised and supported 248. 329 sqq. On a high
pillar 380 seq. ; the other arm also being raised
248. Hand supported 274. Leaning attitude 329.
Foot raised and supported 302 seq. Arm over
head 162. 248. 337. Clasping knee 304 ; head bent
254 ; sceptre held high 364, 2. Sitting on the lap
of another figure 460 seq. ; long stride in works of
Pheidias 99 ; motives from the flat introduced into
sculpture in the round 304.
Mousaios 64

Muses : basis from Mantineia 309, 319, 325 ; on the
Chigi relief 408, 10.
Narkissos, wrongly interpreted as N. 254, 3 ; on
gem 275, 2.
Nemesis of Agorakritos 8. 88 ; with swan (called
Leda) 395
New-Attic reliefs 43, 4. 439.
Niketas 31, 9
Nikias 432 sqq. 437 ; dedicates Palladion 433 ; and
Demosthenes 445 seq. ; and Delos ibid.
Niobids, reliefs of Pheidias 43 seq. ; statues 394 ;
Niobid Chiaramonti ibid.
Nymphs, Sphragitides 449
Nyx on Parthenon pediment 465
' Oil-pourer ' 257 sqq. ; on vases 259 ; various
statues 259 sqq. ; terra-cottas 260, 4
Oreithyia on Parthenon 460.
Orpheus relief 88.
PAIN, expression of, in fifth century 134 seq.
Pan, Polykleitan 299. Skopasian (?) 303. On frieze
of Nike temple 449.
Panathenaia 427 sqq. 471.
Pandora, on basis of Parthenon 45.
Pantarkes 39 seq.
Pantias of Chios 349.
Parian art 362.
Parthenoi, on west pediment of Parthenon, and cultus
453 sqq.
Parthenon. See under Athens, Akropolis.
Parthenos. See Index I. , under Pheidias.
Peisianax 41.
Perikles, portrait of 10. 48. 62. 117 sqq. 145; and
Pheidias 34. 36 ; and Kresilas 1 16 ; and the Par
thenon 422. 423 sqq. 432. Peace Congress 423.
Perseus, with Hades cap 80, 1. 198 ; on coins and
vases 148 ; running or kneeling 200 ; in tranquil
attitude 200 seq. ; by Myron 197 sqq. ; by Pythago
ras 197, 3.
Personification of mountains 458, 3. 468-, 3.
Phigaleia, sculptures of 99.
Philip, portrait of 321.
Phryne 318. 320.
Plataia, Hera at 102. Battle of 446 sqq.
Pliny, his chronology 40. 309.
Poseidon, Pheidian? 104. Myronian? 186. In west
pediment of Parthenon 456 sqq. Statue from Melos
376.
Prize-amphoras 24, 6. 439, 6.
Promachos. See Index I. , under Pheidias.
Psyche, so-called, at Naples 395.
Puteal, Corinthian 441 ; Capitoline 441.
Pyramids, small, of terra-cotta 469.
Pythodoros, son of Epizelos 450, 1.
Rain-charm 469 sqq.
Ravens, legend, interpretation 469 seq.
River god, horned 55. 81. 102, 3.
Ruvo vases 109

INDEX

487

Sappho on coins of Mytilene 71, 1. Supposed por
trait 69 seq. Head, so-called 66.
Satyr pouring wine 310^^. ; resting 329 seq. ; later
modification 330.
Seilenos, on Magna Graecia vases 108, 7 ; hair 181
Selene riding 465. 465, 8.
Selinos, Heraion at 223, 1.
Shield painted on the inside 45.
Ship, in ritual 471.
Sikyon and Argos 416. School of painting 416. 256.
Sicily : coins 105 sqq. vases no.
Smikythos votive gift 214. 215. 219.
Sunion, temple-sculptures 46, 2.
Tarentum : coins 45. 107. vases 1 10. Dioskouroi
cult of 102.
Technique : bronze 7 seq. 18. 134, 2. 233, 6. 352. 391.
marble 7. 330. 391. Joined pieces 369. 376. 395, 3.
Use of drill 309. 334. 344.
Terminal busts and figures : varying turn of head
66, 5. 132, 1. Shoulder-curls and fillets added
132, 1. 236. 310, 2. Of hip-depth, draped 274, 6.
Ludovisi terms 248, 4. 299. 357.
Terminal Aphrodite 274
Terra-cottas : Boeotian, severe style 23, 4. From
Asia Minor 239. 384. Aphrodite with arm raised
and supported 380, do. with foot raised 400, 1.
Terra-cotta slabs from Parthenon frieze 431, 5.
Tetrapolis 462, 1.
Theseion. See under Athens, Lower City.
Throne of gods 429.
Thurii, coins of. See under Coins.
Thymilos 311, 2.
Treasure of Confederation 425.
Treasury of Athenians at Delphi 35, 5.
Triptolemos 255. 333. 334. 349. 353 ; 0f Praxiteles
334-
Turret-crown worn by hero 277 seq.
Twelve gods, Attic 430 seq.
Tyche, of Melos 381 seq. ; and Artemis 325 ; and
Aphrodite 382 ; by Praxiteles (Isis-Tyche) 325.

Vases : Attic black-figured (bareheaded Athena) 14.
Severe fine style (b.c. 460—450) 14. 25. Dating
of vases in the fifth century 21. Paintings derived
from shield of Parthenos 44. 45. 99. 109 seq.
Helios and Selene as cosmic setting to vase-paint
ings 45. Panathenaic amphoras of later style 24, 6.
439, 6. Late, black-figured, white Iekythoi 124.
Copies of statues on vases 124. 153. ' Oil-pourer '
motive 259. Statuette vases 80, 1. 332. With
Eleusinian figures 332. Other vases with Eleu
sinian representations 332, 8. 333. 5. Zeus on
vases (headdress) 42. 217. Drapery of women 21
sqq. Magna Graecia fabrics, fifth century, 1 08
sqq. Influence of tragedy on vases 1 10.
Venus of Milo 367 sqq. ; of Capua 384 sqq. ; of Aries
319 seq.
Victory, writing on shield 386 seq.
Votive offerings, to insure success 8 ; in thanksgiving
after victory 33. 35. In form of buildings, temples,
and temple images 35. 444 seq. Seats, armchairs,
couches 429 seq. Custom of avdBeais 445.
Wheels, with spokes in form of double cross 469.
Xenokrates, one of Pliny's sources 173. 196. 224
seq- 348. 35i- 356.
Xoanon 444 seq.
Xouthos 462.
Zanes at Olympia 308. 343. 349.
Zeus, of Pheidias 36 sqq. ; his throne 43 seq. ; Z. on
Parthenon frieze 431 ; on east pediment 463 ; in
Olympia east pediment 188. Myronian 186 sqq.
Attic, fourth century 299. Of Otricoli 190. 342.
Talleyrand 42. F52, 4. Eleutherios in Syracuse
and Lokri 218. Meilichios 209. 212, 1. 214
Moiragetes 466. Hair 42. 217. On coins of Elis
218. At Olympia ded. by Mummius 215 seq.
Z. type adapted to Asklepios 190.

Richard Clay and Sons, Limited,
london and bungay.

YALE UNIVERISTY LIBRARY

3 9002 02664 3909