GREEK AND ROMAN 
 VERSIFICATION 
 
 WITH 
 
 AN INTRODUCTION ON THE DEVELOPMENT 
 OF ANCIENT VERSIFICATION 
 
 BY 
 
 LUCIAN MULLER 
 
 TRANSLATED BY 
 
 SAMUEL BALL PLATNER 
 
 Professor in Adelbert College 
 
 , • > ' i' ' y > 
 1 i i > > } > 
 
 Boston 
 
 ALLYN AND BACON 
 1892
 
 Copyright, 1892, 
 By SAMUEL BALL PLATNER. 
 
 c c t ccc *""cc 
 
 c t 
 
 t 1 1 
 
 , c c c ' 
 
 c tt tecs 
 
 Typography ey J. S. Gushing & Co., Boston. 
 
 Presswork by Berwick & Smith, Boston.
 
 o 
 z 
 
 00 
 
 c 
 
 CO 
 
 u 
 
 CO 
 
 o 
 
 f A 
 
 TRANSLATOR'S NOTE. 
 
 LuciAN Muller's " Metrik der Griechen und Romer"' (2d ed., 
 Leipzig, 1885) has met with so favorable a reception in Europe, 
 and is in so many respects a valuable handbook, that it has seemed 
 to be worth while to translate it into English, with tlie author's 
 sanction. 
 
 Almost all students in our preparatory schools and colleges are 
 sadly deficient in their knowledge of Latin and Greek versification, 
 and any help whatever towards remedying this condition of things 
 may not be amiss. Hence this translation, in which no changes 
 have been introduced except the musical notation. 
 
 S. B. P. 
 
 July, 1892. 
 
 44G4'?fi
 
 PREFACE TO THE SECOND EDITION. 
 
 The kind reception of this book, which was pubHshed in 
 the beginning of 1880 in Russian, and lias already been most 
 carefully translated into French and Italian, of necessity 
 spurred me on to make it still better adapted to subserve 
 its intended purpose. Hence the criticisms, offered in 
 the different reviews of the book, as in the " Philologische 
 Rundschau" (1881, No. 38) and the "Revue Critique" 
 (188 1, Nos. 36 and 52), have been conscientiously considered 
 and, as far as possible, made use of. I am especially indebted 
 to Professor A. Eussner, who has called my attention to 
 various inequaUties in the work. The addition of an Alpha- 
 betical Index to this edition, as well as to the French and 
 Italian translations, will please many readers. 
 
 I have not succeeded in accomplishing the desire, several 
 times expressed, to treat exhaustively of the metres of 
 Catullus in this little book. For these I must refer to 
 the " Summarium rei metricae poetarum latinorum," St. 
 Petersburg and Leipzig, 1878, a work which has retained 
 its popularity even by the side of the German ' Metrik,' 
 as is shown by the continuous demand for it. 
 
 L. M. 
 
 St. Petersburg, January i, 1885.
 
 PREFACE TO THE FIRST EDITION. 
 
 The thought which has influenced me most in the compo- 
 sition of this book, is that which I have already expressed 
 on page loi of my Biography of Ritschl, namely, that a 
 knowledge of the most usual classical metres, founded on 
 a developed linguistic sense, is the most important and, in 
 practice, the most necessary requirement — as well for the 
 teacher as for the pupil in the gymnasium. Even among 
 philologists there are few specialists in versification. How 
 can more be asked of the students than is demanded 
 above ? 
 
 It is, however, a matter of great moment that this knowl- 
 edge should not be simply mechanical, a mere exercise of 
 the memory, but that it should be thoroughly understood 
 and felt by the students, so that they, to speak with Horace, 
 not only legitiiimm sonum digitis callent, but, as is most im- 
 portant, ail re. Encouraged by the approval which has been 
 bestowed in Germany, France, Russia, and elsewhere, upon 
 my "Summary of Latin Versification," — a book intended, as 
 is stated in the Preface, for students, teachers, and philolo- 
 gists, who are not specialists in this line, — and by its wide 
 circulation, in spite of the fact that it is written in Latin, 
 I now have decided to write a treatise on the versification 
 of the Greeks and Romans, especially adapted for the 
 upper classes in the gymnasia. To this task I have been 
 repeatedly urged by higlily esteemed teachers. 
 
 5
 
 6 Preface to First Edition. 
 
 The method is exactly the same as that pursued in the 
 " De re metrica poetarum latinorum." Following the 
 examples of Hermann and Lachmann, and still more that 
 of Bentley and Porson, the attempt is everywhere made to 
 explain the phenomena of versification from a linguistic 
 point of view. While there may be a difference of opinion 
 concerning the scientific justification of this mode of 
 procedure, a question that I have discussed at more length 
 in the Biograjjhy of Ritschl, page loo, there can hardly be 
 any doubt among intelligent teachers of its practical useful- 
 ness for the purpose of this handbook. 
 
 The great majority of judges who are qualified to express 
 an opinion on the subject now acknowledge that gram- 
 matical accent is wholly without influence so far as the 
 rhythmical formation of the classical metres is concerned. 
 My own theory, which goes much further and amounts to 
 this, that the main object of the old poets was to produce 
 as great variation as possible between the poetical rhythm 
 and the grammatical accent, and that in general, in the 
 structure of the verse, no regard was had for the accent, 
 but only for the number of syllables, especially for the 
 balancing of monosyllables and polysyllables, still encounters 
 much opposition. My only hope is that after reading 
 the Fifth Section, even my most stubborn opponents will 
 acknowledge that this view can be put to excellent use in 
 actual practice. 
 
 In accordance with the object of this work the Greeks 
 principally considered are Homer, the fragments of the 
 Elegiac, Iambic, and .'Folic poets, as far as they serve to 
 illustrate Homer ; among the tragedians, especially Sophocles ; 
 of the Romans, Vergil, Horace, Ovid, Ph?edrus, Tibullus, 
 and Propertius. Only occasional references are made to
 
 Preface to First Edition. 7 
 
 the chorus and in general to the lyrical parts of the Greek 
 tragedies, for the following reasons : — 
 
 In the first place, the criticism and metrical reconstruction 
 of these parts is very uncertain, just as in the cantica of 
 Plautus. Besides, a metrical scheme of these passages is 
 found in all the editions ordinarily used ; and although 
 I have grave doubts about the accepted divisions, still any 
 change in them would be dangerous without a longer 
 argument, and polemical arguments would expand this book 
 beyond its proper limits. Further, I am of the opinion 
 that the teacher should read metrically the lyrical parts of 
 tragedy, and require the same of the students, without 
 lingering too long over the versification. Every teacher 
 will admit that in the reading of a Greek drama in the 
 gymnasia so many other difficult questions must be consid- 
 ered, that only a small portion of time can be given to 
 metrical questions, if the reading of any particular play is 
 to be finished or even carried to any considerable length. 
 It is the ol)ject of the gymnasium to develop the under- 
 standing and imagination of the scholars, and to inspire in 
 them a love and appreciation of classical antiquity, but 
 not to make of them philologists or specialists in metre. 
 Therefore I believe that I shall have fulfilled my duty if 
 I succeed in bringing the students to know and understand 
 the ordinary metres of those poets usually read in the 
 gymnasia. 
 
 I have therefore treated especially of the two most 
 frequent and noble metres, the Dactylic Hexameter and 
 Iambic Trimeter (together with the Strophes of Horace), 
 with the conviction that one who has thoroughly mastered 
 these measures has already advanced a long way into the 
 knowledge of ancient versification.
 
 8 Preface to First Edition. 
 
 Just because versification is so often unreasonably 
 neglected, a great part of the charm that poetry afforded 
 to the men of olden time is veiled or wholly taken away 
 from students. Every teacher knows that for most students, 
 until they reach the highest classes in the gymnasium, 
 scanning is a veritable " crux,'' which has contributed not 
 a little to keep alive the prejudice against classical authors, 
 while in reality the most beautiful creation of the genius of 
 language is the versification of the classic poets which are 
 read in the schools. To produce a love for this versifica- 
 tion and an understanding of it, which may outlive the 
 years of school life, is the purpose of this book. 
 
 Truly, I must beg for it an indulgent reception. 
 
 It is not easy in a handbook, which should be at once 
 thoroughly scientific, brief, and generally intelligible, to 
 satisfy all fair demands, not to mention the unfair. The 
 task was all the harder because, although for a long time 
 I have had a lively interest in the practical questions of 
 classical philology as well as in the needs of the gymnasia, 
 still on account of my position I have come but little into 
 direct touch with these institutions, and so have been 
 deprived of the equally inspiring and manifold impressions 
 produced by daily immediate contact with youth. So much 
 the more grateful ought I to be, that experienced teachers 
 of different countries have assisted me with their advice. 
 They were also, almost without exception, of the opinion 
 that I should treat in only a cursory way of the lyrical 
 parts of the Greek tragedy. 
 
 For the rest, perhaps this handbook, though primarily 
 intended for the upper classes in the gymnasia, may prove 
 not unwelcome to many students of philology, even the 
 younger.
 
 Preface to First Edition. 9 
 
 The introduction is, of course, principally for teachers 
 and philologists. For the part treating of Greek versifi- 
 cation I must ask indulgence, since in the metrical works 
 known to me, even in the excellent book of Christ which 
 I have often used, there is no simple connected statement 
 of the development of Greek versification. For this part, 
 as well as for the whole book, any corrections or hints of 
 qualified philologists or teachers will be most welcome. 
 
 L. M. 
 
 St. Petersburg, January i, iSSo.
 
 CONTENTS. 
 
 »o« 
 
 PAGE 
 
 Translator's Note 3 
 
 Preface to the Second Edition 4 
 
 Preface to the First Edition 15 
 
 INTRODUCTION. 
 
 Development of Classical Versification. » 
 
 1. General Remarks 13 
 
 2. Greek and Roman Versification compared 14 
 
 I. Greek. — 3. Hexameter and Pentameter 15 
 
 4. Archilochus 17 
 
 5. ^olic Lyric Poetry. — Anacrcon. Ilipponax. Ananias 17 
 
 6. Doric Lyric Poetry 18 
 
 7. The Attic Drama 20 
 
 8. The Alexandrian and tlie Later Cireek Poets. — • Nonnus 23 
 
 II. Roman. — g. The Earliest Period until Ennius 24 
 
 10. Ennius. Lucilius. Accius 26 
 
 11. Contemporaries of Cicero 28 
 
 12. Augustan Age 29 
 
 13. The First Centuries after Christ 32 
 
 14. Antiquarian Tendency in Versilication 2;} 
 
 III. — 15. Final State of Greek and Roman Versification 34 
 
 16. Rhythmical Poetry 36 
 
 FIRST SECTION. 
 
 Gener.\l Introduction. 
 
 1. Rhythm and Metre 37 
 
 2. Long and Short Syllables. Arsis and Thesis 38 
 
 3. Verse-feet. Basis. Anacrusis 38 
 
 4. Verse 40 
 
 5. Cresura 42 
 
 10
 
 Contents. 1 1 
 
 PAGE 
 
 6. Final Syllables 42 
 
 7. System. Strophe. Epode 43 
 
 8. Punctuation in Verse 43 
 
 9. Rhyme. Alliteration 4^ 
 
 SECOND SECTION. 
 
 Ox PeculIxVrities of the Foot. 
 
 10. Synapheia 4^ 
 
 11. Resolution of Thesis and Arsis 4*^ 
 
 THIRD SECTION. 
 Description of the Most Important Metkes, Strophes, and Systems. 
 
 12. Dactylic Metres 5° 
 
 13. Anapaestic Metres 55 
 
 14. Iambic Metres 57 
 
 Iambics of Phredrus 60 
 
 15. Trochaic Metres 62 
 
 16. lonici a Miiiori 64 
 
 17. Logaoedic Metres 64 
 
 18. Asynartete Verse (Mixed Measures) 66 
 
 19. The Elegiac Distich (Callinus, Archilochus) 67 
 
 20. The Lyric Strophes of Horace 67 
 
 21. Epodic Systems , 7' 
 
 FOURTH SECTION. 
 
 On Metrical Licenses. 
 
 22. Preface 74 
 
 23. Metrical Licenses 74 
 
 FIFTH SECTION. 
 On the Rhythmical Strccture of the Verse. 
 
 24. General Remarks 77 
 
 25. Rhythmical Structure of the Hexameter and Pentameter 79 
 
 26. Rhythmical Structure of the Remaining Metres 83 
 
 SIXTH SECTION. 
 
 Enclisis and Tmesis. 
 
 27. Enclisis 86 
 
 28. Tmesis 87
 
 12 Contents. 
 
 SEVENTH SECTION. 
 On the Treatment of Successive Vowel Sounds. 
 
 PAGE 
 
 29. Synizesis, Diaeresis, Crasis, Elision, Hiatus 89 
 
 30. Synizesis in Greelc 92 
 
 31. Synizesis in Latin 93 
 
 32. Diaeresis g^ 
 
 12,- Elision 96 
 
 34. Elision in Greek. Crasis. Aphreresis 97 
 
 35. Elision in Latin 98 
 
 36. Differences in Elision in Greek and Latin \'erse loi 
 
 37. Hiatus 103 
 
 38. Hiatus in Greek 104 
 
 39. Hiatus in Latin 106 
 
 EIGHTH SECTION. 
 
 Lengthening by Position. 
 
 40. General Remarks 108 
 
 41. Greek loS 
 
 42. Latin 109 
 
 NINTH SECTION. 
 Homeric Prosody. 
 
 43. Peculiarities of Prosody in Homer ill 
 
 TENTH SECTION. 
 Latin Prosody. 
 
 44. Peculiarities of Latin Prosody t ' • 
 
 ELEVENTH SECTION. 
 Lengthening. 
 
 45. Lengthening by the Thesis at the End of a Woril 116 
 
 46. Greek 116 
 
 47. Latin 117 
 
 Alphabetical Tndi.x 119
 
 INTRODUCTION. 
 
 0-0>d<CK> 
 
 DEVELOPMENT OF CLASSICAL VERSIFICATION. 
 1. General Remarks. 
 
 In both Greek and Latin poetry versification depends 
 solely upon tlie length of the single syllable ; that is, 
 upon the principle of quantity. The versification of the 
 classic peoples developed in precisely the same way as the 
 plastic art of the Greeks, and for the metrical form of 
 language, originality did not appear to the poets to be 
 the most important requirement. Rather was it the rule 
 that when an exceptional genius had discovered the metrical 
 form best adapted to a particular kind of poetry, this should 
 be preserved ; and poets preferred developing in details the 
 approved invention of another, to supplanting it by a new 
 and perhaps less suitable form. 
 
 Thus the Dactylic Hexameter became, through Homer, the 
 accepted epic verse of all antiquity, and the Iambic Trimeter 
 and the Trochaic Tetrameter Catalectic, both of which had 
 their origin at the country festivals of the lonians, remained 
 the favorite metres of the dialogue of tragedy and comedy, 
 products also of those same festivals, long after these had 
 developed into artistic poetical form. The younger generation 
 of Athenian tragedians followed the metrical example of the 
 great masters — /Eschylus, Sophocles, and Euripides. 
 
 For this reason a continuous, though not always equally
 
 14 Development of Classical Versification. 
 
 active and intelligent, tradition of metrical art runs throughout 
 antiquity. This was greatly furthered by the guilds of poets 
 formed in the centres of ancient culture, e.g. Athens, Alexan- 
 dria, and Rome ; and again after tlie time of Alexander the 
 Great, by the care of the grammarians, who not only carefully 
 analyzed the versification of the classical poets, but also kept 
 up uninterrupted intercourse with poets. 
 
 2. Greek and Roman Versification compared. 
 
 If the Greek versification of the classical age, down to the 
 time of Alexander the Great, is compared with the Latin up to 
 Hadrian's time, it will be seen that the Greek versification is 
 distinguished by originality, boldness, versatility, grace, and 
 variability, in consequence of which characteristics it some- 
 times falls into arbitrary and irregular forms (though the Greek 
 poets, and even Homer himself, are much stricter in their 
 versification than was formerly supposed), while the Latin is 
 distinguished by earnestness, dignity, and strictness of rule, 
 which descends to the minutest detail, as well as by a clear 
 understanding and judicious application of this strictness. 
 This was well suited to the peculiarities of the Latin language, 
 the strong, energetic, and sonorous, but much less rich and 
 variable, sister tongue of the Greek ; although, as a result, 
 Latin verse sometimes suffered from monotony, pedantry, and 
 excessive care. 
 
 In the choice of metres for the different kinds of poetry the 
 Greeks usually surpassed the Romans in taste, as it is generally 
 the case that in the artificial imitation of the metres of another 
 people mistakes in use easily arise. During the classical period 
 of Roman poetry it is especially the polymetrical forms of verse 
 of the contemporaries of Cicero that show uncertainty and 
 misconception in this respect.
 
 Hexameter and Pentameter. 15 
 
 I. Greek. 
 3. Hexameter and Pentameter. 
 
 The oldest verse measure of a people is naturally Kara 
 o-Tt'xov ; that is, it consists of one verse, recurring as often as 
 the poet pleases. 
 
 The first measure artistically developed among the Greeks 
 was the Dactylic Hexameter. Its inventor, who belonged to 
 the Ionic stock, is unknown.^ 
 
 The lightness and mobility of the Ionic dialect ; its richness 
 in short syllables ; the possibility of increasing still more the 
 number of these short syllables, l)y placing long final vowels 
 in a hiatus ; the possibility, on the other hand, of lengthening 
 short vowels by position, or at the close of words by the thesis ; 
 the varying quantity of many syllables, and the substitution of 
 a long syllable for a pyrrhic arsis, — produced the rapid and 
 peculiar development of this metre, of which the oldest repre- 
 sentatives are the Homeric Hymns (about 900 B.C.) and the 
 poems of Hesiod, and of the Homeric and Hesiodic school 
 (about 800 R.c). 
 
 Monotony of rhythm in the Hexameter is avoided by 
 the change from dactyl to spondee, and by the different 
 csesuras. 
 
 This metre continued to hold its place in general popularity, 
 and, until the end of the Middle Ages, was used in the most 
 different kinds of poetry. Sanctioned by the authority of the 
 Homeric poems, it influenced not only the dactylic measures, 
 but also the other different metres. Still the authority of 
 
 ^ The ancients frequently confuse the inventor of a measure with the 
 poet who first introduced it in literature, and so metres are often named, 
 nut after tlieir inventors, hut after tlie poet Ijy whom they were most 
 frequently employed.
 
 1 6 Development of Classical Versification. 
 
 Homer stood very much in the way of the proper development 
 of the Hexameter from a popular to an artistic form, because 
 in later times Homer's metrical rules and licenses were 
 partially imitated, while the linguistic phenomena which 
 occasioned them were in large part not understood ; e.g. 
 there was no knowledge of the digavima ceolicum as used by 
 Homer. The evil became still worse in this respect, that 
 the Alexandrian poets evolved from misunderstood passages 
 in Homer a mass of absurd rules or exceptions, as e.g. the 
 hypermetric Hexameter, and a new metre, the Hexameter 
 my urns : — 
 
 — WW _lww — v_yw — WW — WW WW 
 
 Their theories in turn served as models for the Greek and 
 Roman poets. 
 
 The Pentameter was produced by repeating the first half of 
 the Hexameter as far as the penthemimeral caesura. Hexam- 
 eter and Pentameter together formed the first verse system, 
 the Distich, which appears for the first time in the work of the 
 Ionic poets Callinus and Archilochus (about 700 B.C.). 
 Through this change of metre the verse itself became more 
 lively and passionate than in the unvarying Hexameter, and 
 consequently the Distich resulted in giving more room for 
 the subjectivity of the poet, and paved the way for lyric 
 poetry. 
 
 The Distich was the beginning of strophic forms, and it 
 shows already the harmonious and artistic finish which every 
 strophe of the Golden Age of Greek literature has, even 
 if, in the strophes of the Doric lyric poets and the dramatists, 
 the difficulty of their formation and the great corruption of 
 the text often prevent us from completely appreciating the 
 skill displayed in their composition.
 
 ALolic Lyric Poetry. 17 
 
 4. Archilochiis. 
 
 FroiTi early times it had been customary at harvest festivals 
 and vintage time to recite or sing songs of a joking or mock- 
 ing sort, usually written in alternate form, in which Iambic 
 and Trochaic metres were employed. In these measures the 
 thesis was not fixed, but could be resolved ; and in certain 
 positions of the arsis a short vowel could be replaced at pleas- 
 ure by a long. These metres came forth from obscurity into 
 use at about the same time as the Distich, through the poems 
 of one of the greatest artists of antic^uity, — Archilochus of 
 Paros, — who with perfect artistic knowledge used the Iambic 
 Trimeter and Trochaic Tetrameter Catalectic in their greatest 
 beauty. He also employed various iambic and dactylic metres, 
 dactyls with anacrusis, asynartetic verses, made up by a combi- 
 nation of dactylic and iambic or trochaic measures. At the 
 same time he developed the Epodic System, particularly 
 by a union of iambic trimeters and dimeters, but also from 
 dactylic or asynartetic and iambic (probably also trochaic) 
 verses, and vice versa. 
 
 5. .aiolic Lyric Poetry. — Anacreon. Hipponax. Ananius. 
 
 After Archilochus Greek versification made very rapid prog- 
 ress. Among those especially influential in its development 
 were the .-I'^olic poets Alcaeus and Sappho (about 600 i;.c.), 
 who were the first to compose real strophes, consisting usually 
 of two or four verses, of which two at least were alike, so that 
 the metrical elements of the corresponding verses of the 
 strophes were in general quite alike, except for single licenses 
 in the basis, anacrusis, or middle of the verse. These poets 
 seldom employed purely dactylic verse either in strophic or 
 non-strophic poems, but more frequently such verse preceded
 
 1 8 Development of Classical Versification. 
 
 by a dissyllabic beat of any quantity, even pyrrhic, — seldom 
 iambic or trochaic, — apparently without a resolution of the 
 thesis, but especially logaoedic, ionici a /niiio?i (sometimes 
 also a luajori, in combination with trochees), asynartetic 
 measures, and mixed metres in great variety. The most 
 remarkable of their strophes are the Sapphic, Alcaic, and 
 Asclepiadean. The yEolic poets exercised great influence, 
 especially on the Alexandrians and the Romans ; e.g. the 
 metre invented by Sappho, but named hcndccasyUabus pha- 
 Icecius, after an Alexandrian poet, was very often employed 
 by both. 
 
 Anacreon, an Ionian of Teos (about 550 B.C.), stands about 
 midway between the versification of Archilochus and that of 
 Alcaeus and Sappho. The effeminacy of his nature appears 
 especially in his frequent use of ionici a iiiiiiori, Glyconics, 
 and the Anacreontic measures, named after himself. On the 
 other hand, in other fragments, he appears as in/iiieros ani- 
 mosqiie sccutiis Archilochi. In the formation of systems or 
 strophes he was very fond of using Glyconic and Pherecratean 
 measures. The spurious collection of poems which goes under 
 his name is of no consequence in the discussion of his metrical 
 forms. At about this same time the lonians Hipponax and 
 Ananius gave to the Iambic Trimeter and Trochaic Tetrameter 
 Catalectic a new form by changing, in a curious way, the last 
 iambus into a spondee, accented on the second syllable like 
 the original iambus. 
 
 These " limping iambics " found great favor among the 
 Alexandrians and Romans. 
 
 6. Doric Lyric Poetry. 
 
 The freest and boldest development of Greek lyric poetry 
 took place among the Dorians. While the /Eolic and Ionic
 
 Doric Lyric Poetry. 19 
 
 lyrics were intended primarily to be rendered by soloists, the 
 Dorians wrote their songs principally to be sung by the chorus 
 on sacred or holiday occasions, so that one of these poets, 
 Stesichorus, is said to have derived his name from this 
 custom. 
 
 The Doric (like the dramatic) lyrical poetry preferred in 
 general strophes of five or more verses, seldom over twenty, 
 skilfully combined out of metres differing both in compass 
 and component parts. In this way it happened that two or 
 three shorter verses, or parts of verses, were united into one 
 verse (Periods). 
 
 The antistrophe . corresponds exactly to the strophe. After 
 the time of Stesichorus the antistrophe was often followed 
 by the epode, to which the following epodes must exactly 
 correspond. 
 
 In contrast with the vivacity of the lonians and the passion 
 of the Cohans, the lyric poetry of the Dorians is character- 
 ized by earnestness, dignity, and calm in a degree appropriate 
 to its object. 
 
 Alcman is considered the father of Doric lyric poetry 
 (about 612 B.C.). He was followed by Stesichorus, Arion, 
 Ibycus, Simonides, Bacchylides, and finally Pindar about 480 
 P..C., the only Doric lyric poet from whom complete poems 
 have come down to us. 
 
 The last offshoot of Greek lyric poetry is the Dithyramb, 
 originating in the worship of Bacchus, and characterized by 
 the boldness and variety of its metres. It was introduced 
 into literature by Arion (about 600 n.c). 
 
 Although this metre also was at first antistrophic, after the 
 year 400 b.c. the antistrophe fell away, and in consequence 
 of this the Dithyramb degenerated into such looseness that 
 by the irregularity of its structure it seems to have passed
 
 20 Dcvelopvicnt of Classical Versification. 
 
 from the highest point of thought and metre into mere 
 
 prose. 
 
 7. The Attic Drama. 
 
 The drama that has developed out of the songs and 
 dances of country iestivals is a combination of epic and 
 lyric poetry of such a kind that the dialogue parts repre- 
 senting the action of the play form the epic element ; and 
 the songs of the chorus or of individual members of the 
 chorus or of the actors, the lyrical. 
 
 In a corresponding way, as regards the metres, the dia- 
 logue is usually written in iambic or trochaic verses (especially 
 iambic trimeter, more rarely trochaic tetrameter catalectic, 
 in comedy frequently in iambic tetrameter catalectic) ; and 
 the songs in lyric metres, partly those used by the earlier 
 lyric poets, and partly new and free inventions of the 
 dramatists appropriate to the situation. 
 
 Excellent evidence to show how the dramatists regarded 
 the harmony and adaptation of the metre is found in their 
 Siichomythy i that is, the frequent cases where question and 
 answer correspond with each other exactly in compass 
 (usually one verse, sometimes two or more). 
 
 The songs of the whole chorus which enters when the 
 action has reached a point of rest or change, and also 
 marks the end of an act, are characterized by the calmness 
 and dignity of the rhythm. The songs of the individual 
 members of the chorus and the actors show more liveliness, 
 excitement, and variation, especially in Euripides, and fre- 
 quently lack the antistrophic form (aTroAcAD/AcVa). 
 
 The metrical style of ^Eschylus is strict and regular, some- 
 times even harsh and rigid. The versification of Euripides is 
 free and graceful, though often careless and arbitrary, or show- 
 ing a striving after effect. These faults are particularly con-
 
 TJic Attic Drama. 1i 
 
 spicuous in the lyrical parts, however much they may please 
 us by the change of measures and variety of rhythm. 
 
 Sophocles stands, in respect to metre, midway between 
 the two ; but, in general with the year 424 B.C., the metre 
 of the tragedians becomes freer and les^s exact, as is 
 shown by the Philoctetes of Sophocles and the later dramas 
 of Euripides. 
 
 Euripides, who differs strikingly in the metres of the dia- 
 logue, and still more in the lyrical parts, from his prede- 
 cessors, has had the greatest influence upon the later writers 
 of Attic tragedy. 
 
 Among the comedians, Aristophanes far surpassed all others 
 in wealth of expression, skill, and tasteful employment of met- 
 rical forms, and was for this reason ranked by antiquity with 
 Archilochus. It goes without saying that the versification of 
 tragedy is distinguished from that of comedy by its greater 
 force and dignity, while the rhythm of comedy is gayer and 
 freer. This fact is apparent not only in the choruses but 
 also in the dialogue, and particularly noticeable in the ready 
 admission of anapossts in all the feet of the comic trimeter 
 except the last. 
 
 The epode rarely occurs in tragedy, and not always, as 
 in the lyrical poetry of the 1 )orians, after one pair of 
 strophes, but even after two qr three, though without epodic 
 correspondence taking place, as in Pindar. 
 
 The dramatists were very fond of the anapgestic metre, 
 which had been used in marching-songs by the Spartans 
 from earliest times. It was often employed, parUy in cho- 
 ruses, especially where there was some movement of the 
 chorus or announcement by tlie leader of the chorus, and 
 partly in tlie songs of the actors, particularly where these 
 expressed sorrow or complaint.
 
 22 Development of Classical Versification. 
 
 The anapaestic system consists of dimeters interspersed 
 with monometers, and forms usually the close of the cata- 
 lectic dimeter inversus parooniaciis). — The catalectic tetra- 
 meter was used only in comedy. 
 
 The lyrical portions, especially the strophes and anti- 
 strophes, show the same variety of metres as the Doric lyric 
 poetry. 
 
 Along with iambic and trochaic rhythms, which are used 
 with peculiar licenses in the resolution of the thesis, neglect 
 of the caesura, suppression of the arsis, etc., which are not 
 allowed in the dialogues, we find also dactylic measures, 
 often with anacrusis, epitrites, cretics, paeons, ionici a minopi 
 (not a »iaJon'), dactylo-trochees, seldom dactylo-epitrites, 
 bacchics, and other verses. 
 
 The dochmiac rhythm deserves special mention, of which 
 the original forms, as well as the most usual, are these : — 
 
   — ■' y / 
 
 From this were developed numerous other forms by reso- 
 lution of the theses and by replacing the short syllable 
 preceding the last by a long. 
 
 The dochmiac measure is sometimes combined with the 
 cretic and trochaic, seldom with the bacchic and logaLtdic, 
 and most frequently with the iambic. It occurs rarely in 
 the comedians, and still more rarely among the Dorians, 
 but very frequently in tragedy, especially to express com- 
 plaint and pain. 
 
 Among the logaoedic verses, the favorite metres were the 
 Glyconic and Pherecratean, partly in the stricter form of 
 Anacreon, partly in the manner of the Dorians (with great 
 freedom in the structure of the basis), and partly, after
 
 Alexandrian and Later Greek Poets. 23 
 
 Sophocles, with transposition of the dactyl in polyschematic 
 form. Not infrequently, as among the Dorians, shorter 
 verses, or parts of verses, were united to form a longer verse 
 or period. 
 
 In the fourth century the choral lyrical part of tragedy 
 degenerated, in the same way as the dithyramb, into loose- 
 ness and trifling (illustrated by Chccremon) ; the Middle and 
 New Comedy lacked the chorus, although not the lyric 
 measures. 
 
 8. The Alexandrian and the Later Greek Poets. — 
 
 Nonuus. 
 
 The independent development of Greek versification ex- 
 tends to about the age of Alexander the Great (330 k.c). 
 In the following, so-called Alexandrian (330-30 b.c), as 
 well as in the Roman and Byzantine periods, hardly any 
 addition was made to the previous stock of metres. The 
 Sotadeus, indeed. 
 
 was an invention of the Alexandrians ; but in point of fact 
 the old Greeks had created such an astonishing number of 
 different sorts of metres, systems, and strophes, that any 
 increase was scarcely possible. The later Greeks restricted 
 this metrical wealth, since in general they limited themselves 
 to a comparatively small number of metres and short systems 
 or strophes, and usually imitated in a mechanical way the 
 Ionic and yI''olic poets. Poetry written Kara cTTiyov 
 predominated, just as it did among the Romans after 
 Augustus, and in this manner the Greeks as well as the 
 Romans often employed verses which previously had formed 
 parts of a system.
 
 24 Development of Classieal Versification. 
 
 A marked peculiarity of the Alexandrian and later poets 
 was the tendency towards artificial verses, the most remark- 
 able examples of which are found in the poems of Simmias, 
 Dosiadas, and Besantinus (Anthologia Lyrica, ed. Bergk 
 p. 511 sqq.). Another peculiarity was a great, and often 
 pedantic and affected, carefulness in the structure of their 
 favorite metres, in which attempt they were powerfully 
 assisted by the grammarians, and, in fact, often led astray 
 by their false theories. 
 
 Almost at the end of Greek literature in the fifth century 
 A.D., probably under the influence of Roman versification, 
 Nonnus wrote hexameter verse of remarkable strictness and 
 consistency, though not always with success, inasmuch as he 
 had a decided preference for the dactyl (always in the fifth 
 foot), made the ro\xJr] Kara TpLTov Tpoxalov the ruling pause, 
 avoided a word-end in the fourth trochee, introduced once 
 more the strict rules of position, greatly limited cases of 
 elision, and still more cases of hiatus, and gave up length- 
 ening short final syllables in the thesis. Though the next 
 succeeding poets followed his example, Nonnus lived too 
 late to effect any thorough reform in Greek versification. 
 While music and dancing or rhythmical movements of the 
 body were inseparable from the lyrical poetry of the old 
 Greeks, after the Alexandrian period versification and 
 music were sharply distinguished. The later Greek poets, 
 even the dramatists, intended their works principally for 
 reading or recitation without musical accompaniment. 
 
 II. Roman. 
 9. The Earliest Period until Ennius. 
 
 The oldest metre of the Romans was the Saturnian,
 
 The Earliest Period until Ennins. 25 
 
 whose original form was made up by combining an iambic 
 and trochaic series : — 
 
 malum dabunt Metelli Naevio poetae. 
 
 This verse, which in course of time became confused on 
 account of neglect of the caesura and frequent suppression 
 of the third, and especially the sixth, arsis, could not, after 
 the Punic wars, satisfy the artistic feeling of the Romans. 
 It vanished with Naivius (died about 200 B.C.), although it 
 was occasionally employed by later poets, lilce Accius, 
 Varro, and Terentianus Maurus, in learned imitation of 
 their predecessors. 
 
 Since on occasions of public and private festivity, the 
 liidi sccV7uci were presented as well as the litdi circcnses, 
 Livius Andronicus (after 240 B.C.), and his successors in 
 tragedy and comedy, — Naevius, Plautus, Terence, Ennius, 
 Pacuvius, Accius, and others, — appropriated the scenic 
 metres of the Greek drama, iambic, trochaic, anapaestic, 
 cretic, bacchic, occasionally also dactylic, yet with the 
 greatest licenses ; as e.g. the short arsis in iambics and 
 trochees could always be lengthened, except in the last 
 iambic foot. In the resolution of the thesis and arsis, in 
 the use of elision and synizesis, and in other respects, they 
 often transgressed the rules. Further, in prosody they made 
 use of many irregularities, occasioned by the archaic or 
 plebeian pronunciation of the Latin language which had 
 been so long neglected. 
 
 In the dialogue parts the Romans, like the Greeks, used 
 principally the iambic trimeter and trochaic tetrameter 
 catalectic, and in comedy the iambic tetrameter catalectic 
 also. These metres attained a comparatively high degree
 
 26 Development of Classical Versification. 
 
 of finish, while the other iambic and trochaic metres, still 
 more the cretic and bacchic, and the anap?estic most of 
 all, remained in a very rude stage of development. 
 
 There was no chorus in Latin drama until the time of 
 Augustus, but there were lyrical passages {cantica) sung by 
 the actors, which were usually written in auapsestic, cretic, 
 and bacchic metres. In general the old dramatists, up to 
 the end of the Republic, and particularly after the year 
 150 B.C., influenced by the contemporaneous dactylic poets, 
 show a continual effort after a greater development of their 
 art. On the other hand, the number of their metres 
 steadily decreased, as is shown by a comparison of Plautus 
 and Terence, so that finally the principal metres employed 
 in the drama were the iambic trimeter and the trochaic 
 tetrameter catalectic, which were universally popular. 
 
 The last representative of the old iambic versification 
 was the fabulist Pha^drus (about 50 a.d.), who, in view of 
 the Proverbs of Syrus, then much in vogue, admitted in his 
 own fables, which were also devoted to ethical purposes, 
 the spondee in the even feet of the iambic trimeter, but 
 elsewhere showed almost nothing of the peculiarities of 
 Plautus and Terence. Iambic and trochaic poetry, with 
 the same free use of the spondee, appears occasionally in 
 the antiijuarian period of the Frontonians (about 150 a.d.) 
 and at the end of Roman literature, in consequence of the 
 increasing decadence of culture. 
 
 10. Emiius. Luciliiis. Accius. 
 
 As there was some reason to fear that the Latin language 
 would afjain degenerate into the rudeness of the Saturnian 
 verse through the irregularity of the dramatic metres and 
 prosody, great credit is due to Ennius (239-169 B.C.) for
 
 Enniiis. LiicHiiis. Ace? us. 27 
 
 his intiodiictioii of the dactyhc hexameter, imitated with 
 care and general good taste from Homer. Resolution of 
 the thesis was excluded from Ennius' hexameter, and he 
 adhered as closely as possible to the model of the Greeks 
 in matters of prosody, as the prosody of Latin had been 
 originally homogeneous with that of Greek, and had 
 degenerated only after the lapse of time. Ennius still 
 retained the original long quantity of the final syllables in 
 at, et, it (2d pers. as, es, 'is), as well as the contem- 
 poraneous dramatists. Further, Ennius employed in his 
 Satires the elegiac distich, besides the most common metres 
 of the dramatists, the iambic trimeter and the trochaic 
 tetrameter catalectic, and finally, in imitation of the Alexan- 
 drians, the wonderful Sotadic measure, though with various 
 licenses. 
 
 Although there are cases of harshness, the great majority 
 of Ennius' hexameters are of remarkable beauty, because of 
 the evident genius of the poet, who <^,<,'-. has far fewer cases 
 of elision in the Annalcs than any poet of the Republic, 
 and also because of his rich and comprehensive knowledge 
 of Latin, (ireek, and Oscan ; but above all, because of 
 the fact that Roman versification already before Ennius 
 had followed the (quantitative principle exclusively, so that 
 he had only to remove the abuses of prosody which had 
 crept in during the rude state of the language between 350 
 and 250 r,.c. It is ([uite evident that Ennius attempted to 
 direct the attention of Roman poets to the strictness and 
 consistency of the metrical art. He exercised an immense 
 influence over all later writers, the more so because, until 
 Vergil's time, his much-praised epic, the Annaks, was the 
 most popular glorification of the heroic deeds of the old 
 Romans.
 
 28 Development of Classical Versification. 
 
 Next followed the satirist Lucilius, who employed the 
 hexameter principally, besides the other metres used by 
 Ennius, in his Satires, though he disdained the Sotadic. 
 He marks an advance in metrical art, although there is no 
 lack of harsh usages which are partially excused by the 
 light colloquial tone of satire. Accius, in his non-dramatic 
 poetry, followed the metrical principles of Ennius. 
 
 H. Contemporaries of Cicero. 
 
 Until the time of Cicero the dactylic poets contented 
 themselves with the metres employed by Ennius ; but they 
 zealously furthered their artistic development. The most 
 perfect example of this effort is the poem of Lucretiu's, 
 composed correctly and according to rule, but without 
 grace and variety of rhythm. 
 
 The younger contemporaries of Cicero, led by Lcevius, 
 hke Catullus, Calvus, and others, disdained that simplicity, 
 and introduced into Roman poetry a great number of 
 metres, mostly or wholly borrowed from the Alexandrians, 
 and which they used in general with great skill. Of course 
 the hexameter was not neglected (cf. Varro Atacinus), but 
 the distich remained undeveloped. Catullus was the first 
 among the Romans to show attention to the .Eolic poetry, 
 by imitating Sappho. It is possible also that he was 
 somewhat influenced by Anacreon. The Horatian epode 
 was not unknown to this period; on the contrary, in imita- 
 tion of the Alexandrians, poets already formed iambic and 
 trochaic metres of various kinds according to the strict 
 metrical laws of the Greeks ; indeed, pure iambic lines with 
 complete exclusion of every other foot. 
 
 Hipponactean Iambics and Hendecasyllabic Phalceceans 
 were particular favorites.
 
 Augustan Age. 29 
 
 Between Lucretius and Catullus, beside Laevius, stood 
 Varro Reatinus, who shows in his Satires, in imitation of 
 the Alexandrians, great skill together with great variety 
 of versification. He also employed iambic, trochaic, and 
 Sotadic verses with the freedom of Ennius and Lucilius. 
 
 The verse-systems found in Catullus are the two-lined 
 Asclepiadean, the four-lined Sapphic, and two Glyconics 
 closing with a Pherecratean, one four-lined, and one con- 
 sisting of three and two lines. 
 
 Catullus' versification had imitators and friends even in 
 the Augustan Age, and in the first century a.d. His great 
 variety of metres did not meet with general approval. The 
 Phalaecean continued to be popular until the end of Roman 
 literature, and the Hipponactean until the time of Trajan, 
 though both were more strictly treated. 
 
 12. Augustan Age. 
 
 The age of Augustus (40 is.c. to 14 a.d.) brought the 
 development of Roman metre to a close. 
 
 The hexameter was brought to its highest perfection by 
 Vergil and Ovid. Vergil, indeed, believed that he could 
 not wholly eliminate the Ucenses which tlie older Romans, 
 especially Ennius, had allowed in respect to the rhythmical 
 laws of the hexameter, harsh elision and hiatus, synizesis, 
 lengthening of a final syllable in the thesis, etc. Ijut he 
 used them rarely and moderately, usually only to paint the 
 situation by the rhythm of the verse, — an art in which he 
 is a master. Unfortunately, his versification is sometimes 
 disfigured by a pctlantic imitation of the Alexandrian 
 philologists where they had misunderstood Homeric verses. 
 
 Ovid diminished still more the number of licenses which 
 Vergil had allowed, so that, although the hexameters of the
 
 30 Development of Classieal Versifieation. 
 
 Metamorphoses are somewhat freer than the elegiacs, his 
 verses, considered singly, are the most beautiful models of 
 harmony and metrical skill. 
 
 If read for any length of time, however, they grow tire- 
 some because of their too great similarity, especially as 
 Ovid comparatively seldom makes use of that rhythmical 
 portraiture in which Vergil was so skilful. 
 
 In the hexameter of Satire, which was distinguished from 
 the prose language cjf the educated only by its metre, 
 Horace preserved the licenses of Lucihus, but lessened his 
 harshness. The verse of the Epistles, especially of the 
 second book, is considerably more polished than that of 
 the Satires. As would be expected in this sort of poetry, 
 the licenses borrowed from the Greeks by poets of tlfe 
 higher style, e.g. hiatus and a sjjondee in the fifth foot, are 
 almost wholly avoided. 
 
 The distich also, which in Catullus still appeared rude, 
 was perfected by Tibullus, by Propertius in his later works, 
 and particularly by Ovid, although the latter, in the works 
 written during his exile, dropped something of the metrical 
 strictness shown in his erotic poetry. 
 
 Horace introduced the epodic versification of Archilochus 
 into Roman poetry, and also the lyrical measures of the 
 zEolic poets Alca^us and Sappho. Archilochus and Ana- 
 creon also may have had some influence on the versification 
 of the Odes. It is not certain whether the two metres 
 not met with in Greek (Od. I. 8; Ep. 13), were invented 
 by Horace or not ; also whether the division of the strophes 
 into four lines, which is the universal rule in the Odes, — for 
 Od. IV. 8 is interpolated, — was borrowed from Alcaeus or 
 not. 
 
 In general, we are in the dark about the origin of many
 
 Aligns tan Age. 31 
 
 of the strophes of Horace, on account of the loss of his 
 Greek models and the contradictory statements of the Latin 
 grammarians. 
 
 On the other hand, Horace paid full regard to the spirit 
 of the Latin language, partly by setting aside, and partly 
 by reducing to a very small number, those licenses which 
 the ^olic poets had allowed themselves in respect to 
 caesura and syllaba anccps, and consequently his metres 
 became more equal and dignified. Thus, for instance, the 
 Asclepiads and the eleven-syllabled Alcaics and Sapphics 
 received a firmly fixed caesura. In the Asclepiads the basis 
 was always a spondee, and the Alcaic and Sapphic verses 
 had a spondee always before the caesura. Still Horace has 
 more cases of elision in his lyrical measures than the 
 tragedian Seneca, and further allows monosyllabic conjunc- 
 tions and prepositions at the caesura and at the end of the 
 verse, contrary to Seneca's' usage. No advance was ever 
 made by the Romans beyond the lyrical versification of 
 Horace. The attempt, condemned by Horace, but made 
 by his contemporaries, to imitate the Odes of Pindar, found 
 no sympathy. 
 
 In the time of Augustus, tragedy, which had been zeal- 
 ously cultivated in opposition to comedy, was emancipated 
 from the metrical traditions of the Republic, and its iambics 
 and trochaics were constructed according to the model of 
 the Greek tragedians, and the rule adopted from the Alex- 
 andrians, that tlie foot preceding the last iambus in the 
 iambic trimeter and trochaic tetrameter catalectic must 
 necessarily be a spondee or an anapaest. In general, much 
 greater freedom was allowed the anapaest in iambic trimeter 
 in Latin tragedy than in Greek. At the same time the 
 chorus was introduced, — that is, in the manner of Euripides, —
 
 32 Dcvclopuicnt of Classical Versification. 
 
 with a loose relation to the action of the drama from which 
 was taken only a point of departure for general descrip- 
 tions and reflections. In these choruses anapjEstic mono- 
 meters were employed in a most unartistic manner, besides 
 the dactylic and logaoedic metres illustrated in Seneca. 
 
 13. The First Centuries after Christ. 
 
 The first century after Christ, until Hadrian (117), 
 represents the Alexandrian period of Roman versification. 
 No further enrichment was made, and there were but few 
 attempts to employ a great number of metres. Poets con- 
 tented themseh'es with a consistent and tasteful though not 
 infrequently pedantic improvement of the metres already in 
 use. y 
 
 The grammarians were responsible for many false expla- 
 nations of the metrical peculiarities of Vergil, who was 
 studied with equal zeal, and enjoyed the same authority 
 among the Romans as Homer among the Greeks, although 
 their erroneous theories exercised little influence over the 
 better poets even to the end of the fourth century, as is 
 shown by Claudianus and Rutilius Namatianus in the time 
 of Honorius. Moreover, it was unfortunate that the taste 
 for the four-lined strophes of Horace, with the exception 
 of the ever-popular Sapphic, disappeared so quickly, as the 
 choruses of Seneca prove. Instead, poets began to use the 
 Ivrical measures of Horace in poetry written Kara cni^v. 
 In a very peculiar manner Seneca, in two tragedies 
 (Agamemnon and G<^dipus), combined freer chorus songs 
 out of the short verses or portions of verses of Horace. 
 In the last centuries this custom of making new verses out 
 of parts of Horatian metres became more common. In 
 other respects the metrical art was very carefully developed
 
 Antiqitarian Tendency in Versification. 33 
 
 in the most minute details, so that hcenses which were 
 common in the Augustan poets became rare, and those 
 which were infrequent in the Augustan poetry were almost 
 unknown. 
 
 The post-Augustan poets generally took Vergil and Ovid 
 for their models, but gave the most weight to Ovid's 
 example. For as his distich was the pattern for most of 
 them, so his influence was very considerable on the form 
 of the other dactylic and logacedic metres. In other lines 
 Horace's lyrical and satirical metres were models. Remark 
 has already been made about Catullus' influence (cf. 11). 
 
 Here, too, must be mentioned that tendency, noticed 
 already among the later Greeks, towards playing with words 
 and affectations, which became stronger and stronger, the 
 more the want of real substance in poetry was felt. The 
 most remarkable example, unique in its kind, of this 
 tendency is the poetry of Porfyrius Optatianus (about 
 
 14. Antiquarian Tendency in Versification. 
 
 After the time of Hadrian and Fronto an antiquarian 
 tendency made its influence felt in Roman poetry, to- 
 gether with the steady imitation of Augustan versification, 
 so that occasionally not only iambic and trochaic verses 
 were written with the metrical (not prosodical) licenses of 
 Plautus, but also poets returned to the metrical variety 
 of Laevius and Catullus, as is seen in Septimius Serenus and 
 Terentianus Maurus, poets of the third century. Their 
 example was followed by the Christian writers after the 
 fourth century, although they also employed the verses of 
 Horace in their poetry.
 
 34 Development of Classical Versification. 
 
 III. 
 15. Final State of Greek and Roman Versification. 
 
 The state of both Greek and Roman versification after 
 the third century is so nearly the same that the subject 
 can be treated as one. 
 
 In both languages the most common metres continued 
 to be the dactylic hexameter and pentameter, as well as 
 various iambic and trochaic verses ; in Greek the iambic 
 trimeter, in Latin the trochaic tetrameter catalectic and the 
 iambic dimeter, partly Kara (jriypv, partly in strophes. 
 
 As is shown by the latest ancient and by the mediaeval 
 Greek and Roman poets, all appreciation of the strophe 
 which, in ancient times, had represented the harmonious 
 combination of different verses into one artistic whole, 
 had vanished. All combinations of like or unlike verses, 
 repeated in the same succession of verses and in the 
 same numbers, if only they had a decided stop at the 
 end, were considered as strophes. Such "strophes" were 
 especially used for religious purposes (Christian hymns). 
 In the structure of these hymns, among the Greeks after 
 Gregorius Nazianzenus (360), the most serviceable metre 
 was the iambic trimeter, and sometimes the catalectic 
 iambic dimeter and Anacreontics : among the Romans the 
 most used metres were the iambic dimeter and the 
 catalectic trochaic tetrameter. 
 
 At this same time men began to lose the exact appreci- 
 ation of the peculiar appropriateness of each metre. From 
 the third century Roman poets (Alfius Avitus, and later 
 Festus Avienus) had employed iambic metres in epic 
 representation, and the same thing was done in Greek in 
 the seventh century by Georgius Pisides. On the other
 
 Final State of Greek and Rotnan Versification. 35 
 
 hand, after the same date, Latin tragedies (Medea and 
 Orestes) had been written in hexameter. So, too, in lyric 
 poetry httle taste was shown in the choice of metres, as 
 we see in the case of Ausonius and Prudentius about 
 400 A.D. 
 
 When after the third century thorough culture and, 
 together with that, the appreciation of language decHned, 
 the erroneous theories of the grammarians exercised con- 
 tinually greater influence on the ancient versification, and, 
 further, many Roman poets, especially the Christians, began 
 to neglect quantities, at first in proper names, particularly 
 Greek, or in long words which were unsuited to the verse. 
 This became much worse in the Middle Ages, though here, 
 too, the poets were very different according to their time 
 and training. Among the Greeks the laws of prosody were 
 more strictly enforced. But the Byzantine poets, even the 
 best of them after Georgius Pisides, allowed themselves to 
 make the letters a, t, r, which have not different forms for 
 long and short, either long or short, except when length- 
 ened by a strong position ; y] and w and the diphthongs 
 were always long, e and o short, when a single consonant 
 followed, and frequently when in weak position. Proper 
 names and technical terms were treated very freely. The 
 extinction of the old spirit of the language was shown in this 
 fact that many Christian poets of Rome allowed no resolu- 
 tion of the thesis in iambic and trochaic feet, and in 
 general avoided trisyllabic feet, as was the rule among the 
 Greeks after Georgius Pisides. 
 
 Nevertheless, the quantitative poetry of the Middle Ages 
 was stricdy separated in the consciousness of the poets 
 from the rhythmical. A remarkable proof of this lies in the 
 fact that the Byzantine poets who observed the principle
 
 36 DevelopDunt of Classical Versification. 
 
 of quantity, when writing in iambics, always had a paroxy- 
 tone word at the end of the trimeter. They wished thus 
 to show that their verse took no account of grammatical 
 accent, but rather that in the manner of the ancients the 
 rhythmical accent differed as much as possible from the 
 grammatical. 
 
 16. Rhythmical Poetry. 
 
 On account of the extinction of the appreciation of long 
 and short syllables, there was developed among the Romans 
 after the third century, and among the Greeks in the first 
 half of the Middle Ages when not earlier, the so-called 
 rhythmical poetry, which, being steadily emancipated from 
 the rules of prosody, though at first observing rhythmical 
 laws, led naturally to the observance of the grammatical 
 accent in certain cases, and restored, in Latin, beside the 
 dactylic hexameter, generally popular iambic and trochaic 
 metres, especially the iambic dimeter and catalectic trochaic 
 tetrameter ; in Greek it restored the iambiciis septenarius, 
 which had always been popular, the so-called versus 
 poHticus, which always consisted of fifteen syllables. This 
 metre was very common among the educated also after the 
 twelfth century. In this, as in the rhythmical verses of the 
 Romans, the ccesura was generally retained {e.g. in the versus 
 politicus after the fourth thesis), just as was usual in metrical 
 poetry. The same remarks apply to the strophes of rhyth- 
 mical poetry that were made above, about the metrical 
 strophes of the end of antiquity and of the Middle Ages. 
 
 The last thing to be noticed is, that in most of the 
 Greek and Latin poets of the Middle Ages all apprecia- 
 tion of the difference between poetic and prosaic produc- 
 tions was lost. Everything possible was written in verse, 
 even such themes as lie farthest distant from real poetry.
 
 FIRST SECTION. 
 
 -oO>©:;CK>- 
 
 GENERAL INTRODUCTION. 
 1. Rhythm and Metre. 
 
 The euphony of the classical languages depends, in prose, 
 upon the Rhythm {nutnerns) ; that is, upon a grouping of 
 the words of the sentence, especially at its beginning and 
 end, which is harmonious and pleasing to the ear. In 
 poetry it depends upon the Metre ; that is, the artistic 
 combination of long and short syllables in verse, system, 
 or strophe. 
 
 In olden times, in the structure of verse, the principle 
 of quantity was the only one considered ; that is, the 
 words making up the verse were measured according to 
 the length of their vowels. The prose accent, as well as the 
 logical importance of the words, or parts of words, was not 
 taken into consideration. 
 
 The combination of a raising and a lowering of the voice 
 forms the verse-foot. 
 
 The poets, too, followed the rules of rhythm partly by 
 causing the accentuation of the verse-feet to vary as much 
 as possible from the prose pronunciation of the words, 
 partly by harmoniously joining the parts of the verse so 
 that the compass of the individual words might differ as 
 much as possible from that of the single verse-feet, and 
 
 418 53S
 
 ^S General Introduction. 
 
 partly by using at the end of each metrical series — as 
 well of the divisions of the verse caused by the caesura 
 as at the end of every artificial combination of verse-feet 
 — the kind of verse-foot peculiar to this end. 
 
 2. Long and Short Syllables. Arsis and Thesis. 
 
 Every vowel of a Greek or Latin word has an exact 
 time (xpoVos, tempus), and is either long or short, — with 
 the exception of those cases, rare in prose, but more common 
 in poetry, where the same vowel can be at the same time 
 either long or short. 
 
 A vowel is either long by nature or it is regarded as 
 long on account of its position before two or more con- 
 sonants. In versification, one long is equivalent to two short 
 syllables. 
 
 A complete verse-foot has at least one raising {Thesis) and 
 one lowering (Arsis) of tone, produced by a greater or 
 less stress of the voice. The Thesis is always long, except 
 when it is resolved into two short syllables. The Arsis is 
 sometimes short, sometimes long. 
 
 To scan is to read the verse according to the Thesis 
 and Arsis. 
 
 3. Verse-feet. Basis. Anacrusis. 
 
 The most common combinations of syllables, or verse- 
 feet, are the following : — 
 
 w w Pyrrhic /egi t l^ i^ 
 
 Trochee teg^^ i i> 
 
 Iambus duces i . I 
 
 •y I 
 
 Spondee I'eg'i .... i \
 
 v^ >^ w 
 v.^ v> 
 
 KJ KJ ^^ K^ 
 
 y^ \J \y 
 
 V^ w 
 
 \y ^ ^ 
 
 WW 
 
 Versc-fcet. 
 
 . . . Tribrach legl Ce 
 
 . . . Dactyl legi miis 
 
 . . . Anapaest lege rent 
 
 ... Amphibrachus|('"^/^"t'"L""-H .. 
 •• egone i 
 
 (Bacchius "| 
 
 Antibacchius 
 Palimbacchius J 
 
 {Antibacchius 1 
 PaUmbacchius , 
 Bacchius J 
 
 f Cretic ) - - ,_ 
 
 . < . ,   \ . . consule 
 
 \ Amphimacer j 
 
 39 
 
 amic OS 
 
 legistis 
 
 Molossus legeriint 
 
 . Proceleusmaticus . . an'imula . . 
 
 . Txon pn'w Its consul} hits 
 
 . Poeon secundus .... 11: ge ntibus 
 
 Paeon (eriius legitote . 
 
 . Pa:on qiiartns ttineri 
 
 . Ionic rt w/«(';'/. .. . retulissent . 
 . Ionic a majoyi . . . . confec'erat . . 
 . Choriambus cont tiler ant 
 
 . Antispast legebaris ... 
 
 f Ditrochee "I - ... - ^ - 
 
 -( T^• u y collirnntur. 
 
 I Dichoreus J '^ 
 
 . Diiambus ^<V" "' ' " ^ 
 
 First Epitrite relegerunt . 
 
 Second Epitrite. . . . eltgebant . . 
 
 Third Epitrite .... selegerint . 
 
 Fourth Epitrite. .. . collegtstis . 
 
 Dispondee sel'tgeru n t . 
 
 P * * 
 
 LU 
 
 
 
 I LJ 
 
 
 U I 
 
 * p * 
 
 
 
 
 \ I U 
 
 
 
 1 ^1 
 
 
 
 
 
 
 MM 
 
 O 
 
 ^1 u 
 
 
 
 U I '^ 
 
 
 Lu I 
 
 
 U { 1 
 
 rrtj- 
 
 r0 
 U I 
 
 Ul I I' 
 
 
 
 1 ^ I 1^ 
 
 rr 
 
 tf • 
 
 w 
 
 Dochmius amtcos tenes ... 
 
 
 
 
 
 
 
 I I I u 
 
 
 
 iiTjr
 
 40 General Introduction. 
 
 The combination of two feet is called a Dipody. Iambic, 
 Trochaic, and Anapaestic verse, but not the Dactylic, is 
 measured by Dipodies. The Iambic verse, containing six 
 feet, is therefore called a Trimeter. The greatest stress of 
 the metrical Ictus is laid upon the first part of the Dipody, 
 consequently on the odd feet. 
 
 Basis is the term applied by later writers on metre to the 
 dissyllabic prelude at the beginning of Phalgecean, Phere- 
 cratean, Asclepiadean, and other verses, which in Greek 
 poetry consists of a Trochee, Spondee, Iambus, or Pyrrhic. 
 
 Anacrusis is the name given to the monosyllabic prelude 
 
 before the first thesis of the Alcaic Hendecasyllabic and 
 
 Enneasyllabic, and other verses. This syllable is either long 
 
 or short. 
 
 4. Verse. 
 
 A Verse is a metrical series consisting of like or different 
 feet, which, however, are not combined arbitrarily or mechani- 
 cally, but according to the law of Symmetry and Euphony, 
 as it presents itself to the artistic sense of the poet. 
 
 A Verse must not number over thirty niorce (a mora is 
 the unit of measure, the time of one short syllable, a j* in 
 music) ; although in lyric poetry and in the choruses of the 
 Greek drama there are found longer combinations, so-called 
 Periods, of which no account is taken in the following 
 pages. 
 
 The metrical correctness of a verse — that is, the correct 
 quantity of the syllables, the strict exclusion of all illegitimate 
 feet — was a matter of course in the old poets until the 
 degeneration of Latin and Greek literature ; but metrical 
 correctness by no means makes a verse a work of art, and 
 it becomes such only by an exact keeping of the laws of 
 rhythm, especially at the close of a metrical series. More-
 
 Verse. 41 
 
 over, a longer verse needs at least one definite division 
 (Caesura), which allows the voice to rest, and divides the 
 verse symmetrically and harmoniously. 
 
 Again, it is a law of euphony that the same letter must 
 not recur too frequently in the same verse, as in Ennius : — 
 
 O Tite tute tibi tanta tyranne tulisti. 
 
 Just as much to be avoided are words of the same number 
 of syllables, or those which are too long, as in Ennius and 
 Namatianus : — 
 
 sparsis hastis longis campus splendet et horret. 
 Bellerophonteis sollicitudiniljus. 
 
 On the contrary, there must be a proper mingling of 
 longer and shorter words. 
 
 A verse of which the last foot is incomplete is called 
 Catalectic. If this incomplete foot contains one syllable, it 
 is called Catalecticus in syllabam; if two, Catalecticus in 
 disyllabtim. 
 
 A verse can be called Hypercatalectic if it has one or 
 two arses after the last complete foot. 
 
 A verse is called Simple if it consists of like feet ; 
 Compound, if it consists of unlike feet. 
 
 A verse in which Dactyls and Trochees are used together 
 is called Logaoedic. 
 
 Verses made up of two different metrical series are called 
 Asynartete. At the end of the first series hiatus and syllaba 
 anceps are generally allowed. 
 
 In view of the great artistic taste of the Greeks and 
 Romans, and the highly finished state of their languages, 
 it can be assumed that their most usual verse-feet were 
 also the most perfect. As such, the Dactylic Hexameter
 
 42 General Introduction. 
 
 and Pentameter, the Iambic Trimeter, the Trochaic Tetra- 
 meter Catalectic, the Anapsestic Dimeter and Tetrameter 
 Catalectic, the Glyconic and Asclepiadean measures, and the 
 Alcaic and Sapphic Strophes, are especially remarkable. 
 
 5. Caesura. 
 
 In order to give the voice a rest, verses of more than ten 
 syllables have usually one or two divisions {Ccesura, rofi-i]); but 
 this rest or pause is shorter than at the end of the verse. 
 
 Sometimes there is besides the Primary Caesura a Secondary 
 Caesura, as e.g. in the Dactylic Hexameter we find the Trithe- 
 miineres (caesura after the third half- foot), together with the 
 Heph the mime res (caesura after the seventh half-foot). 
 
 The division of the parts of the verse by the caesura is 
 not accidental or mechanical, but depends upon the law^f 
 Symmetry and Euphony. Hence it often happens that the 
 parts of the verse made by the caesura (also called "Metrical 
 Series") elsewhere appear as independent verses. This is 
 also further explained by the fact that the caesura usually 
 occurs about in the middle of the verse. 
 
 The original metre is frequently cut in two by the caesura. 
 So e.g. the Penthemimeral and Hephthemimeral caesuras in 
 the Dactylic Hexameter fall upon an Anapaest ; in the Iambic 
 Trimeter, upon a Trochee. The greatest stress of the met- 
 rical ictus rests on the first half of the verse as far as the 
 
 caesura. 
 
 6. Final Syllables. 
 
 The Final Syllable of the verse is common; that is, it 
 can be either long or short. Compare 10. 
 
 For this syllable the law of Hiatus is not binding. Elision 
 and the Apostrophe are not allowed at the end of the verse, 
 except when several verses are united by Synapheia. In this
 
 System, Strophe, and Epode. 43 
 
 case neither Syllaba Anceps nor Hiatus is admitted ; but 
 Elision, and sometimes the division of a word at the end 
 of the verse, occur. Compare 10. 
 
 For the so-called Hexainetri hypermetri, compare 36. 
 
 7. System. Strophe. Epode. 
 
 A System is the combination of two or more verses so 
 as to form an artistic rhythmical unity. Such Systems can 
 be repeated as often as desired. 
 
 It is not necessary, and indeed quite seldom happens, 
 that the parts of a System of verses should be united to 
 each other by Synapheia. 
 
 If such a System is repeated one or more times, it is 
 called a Strophe. 
 
 In Doric and dramatic lyrical poetry the even (second, 
 fourth, etc.) Strophes are called Antistrophes. 
 
 Epode {epodus, eVwSo's, fem. gen.) is the name applied 
 to the Verse-System which, in Doric lyric poetry and in 
 the choruses of tragedy, follows the Antistrophe and 
 closes this and the Strophe. 
 
 Epode {epodus, cttwSos, masc. gen.) is also the name 
 applied to a shorter or even a longer asynartete (Hor. 
 Epod. II, 13) verse, which is combined with the preced- 
 ing into a System. Again, the combination of two such 
 verses, with the exception of the Elegiac Distich, is called 
 an Epode (Horace). 
 
 The repetition of Dactylic Distiches or Epodes is not 
 considered as a Strophic structure. 
 
 8. Punctuation in Verse. 
 
 Since the Verse, the parts of the Verse produced by the 
 caesura, and the Verse-Systems, are merely a result of the
 
 44 General Introduction. 
 
 euphony and rhythm of language, they have originally 
 nothing to do with the logical structure of the sentence, 
 as it is represented to the eye by punctuation. Hence it 
 may happen in verses united by Synapheia, that one word 
 belongs to two different verses. Therefore it is by no 
 means necessary that the verse, the caesura, or even the 
 strophe or antistrophe and epode, should close at a point 
 of punctuation, as e.g. Pindar and Horace show. Further, 
 it introduces no difficulty that a word should stand at the 
 end of a verse, or in the caesura, which is to be closely 
 connected with what follows. Thus we often find, e.g. in 
 Horace at the end of lyric verses (except at the end of 
 the strophe) and of the satiric Hexameter et, ant, vet, etc., 
 or a monosyllabic preposition (as in the strophes of the 
 Greeks) . 
 
 Even a full stop before the last syllable is not avoided, as 
 in Catullus : — 
 
 quid? non est homo bellus? inquies? est. 
 
 and in Horace : — 
 
 scitari libet ex ipso, quodcumque refers, die, 
 ad cenam veniat. 
 
 Still, the desire to make the metrical endings, marked 
 out by the caesura and close of the verse, coincide with the 
 divisions of the sentence, as well as to avoid harsh discords, 
 is so deeply implanted in our poetic instinct, that, at least 
 in verses which are not united in system or strophe, the 
 poets early (as Homer shows) directed their attention to 
 this point. Therefore it seldom happens outside of the 
 strophe, especially among the Greeks, that the last syllable 
 or the last foot of a verse belongs to the following sentence. 
 In the same way a full stop after the first syllable or the 
 first foot of a verse is avoided.
 
 Punctuation in Verse. 45 
 
 Therefore such a case as II. I. 51, 52: — 
 
 avTap cTreir avTolai ySeAos €^£7revK€s ec^tets 
 ySaAA * at£6 ok irvpal vckvwv kuiovto Oafxexat ; 
 
 is not to be commended, although here the harshness is 
 diminished by the logical importance of the first word and 
 by the elision. 
 
 Vergil sometimes for the sake of effect ends a rather 
 long speech with an initial dactyl. ^'En. IV. 570: — 
 
 femina.' sic fatus nocti se inmiscuit atrae. 
 
 The placing of a full stop at the distance of a half-foot 
 before or after the principal csesura is also avoided, and 
 hence such a verse as the following is faulty (^n. I. 17): — 
 
 hie currus /le//; hoc | regnum dea gentibus esse. 
 
 Still more this (JEn. VII. 635) : — 
 
 pulverulentus eques | furit. omnes arma requirunt. 
 
 So too these verses of ^schylus and Sophocles (Seven 
 against Thebes, 1030; El. 1038) : — 
 
 dAA' ov TToAi? crruyei, uv \ rt/xryo-ets Td<f)(t) ; 
 oTav yap €V <j)povrjs, To6 yjyrjo-u av vwv. 
 
 In Homer (Od. V. 234), on the other hand, we must 
 write : — 
 
 SwKcv ol TTcAcKW, fxijav, apfJievov iv TraXdfxycTiv ; 
 
 not : — '\ ' 
 
 TTcAcKuv /xeyav. 
 
 Still, the tragedians, especially after Sophocles, allow them- 
 selves much freedom -with respect to the disagreement of 
 metre and punctuation, in order to increase the effect of 
 passages in the dialogue which are expressive of lofty 
 passion. Much more freedom does Horace allow himself
 
 4^ General Ijitrodiictio7i. 
 
 in his satiric Hexameter. Moreover, many other poets 
 (among the Romans) are accustomed to place monosyllabic 
 Conjunctions or Prepositions in the caesura of the Hex- 
 ameter, especially when elision occurs, as Vergil : — 
 
 si genus humawz^;/; et mortalia temnitis arma. 
 
 The Greeks are much stricter in this respect, and a verse 
 like the following (II. I. 53) occurs but seldom: — 
 
 ii'vrjfA.ap [xev dva aTparbv co^^ero KrjXa Oeolo. 
 
 In the tragedians two verses hke the following from 
 Sophocles are rare : — 
 
 KaKov 8k Kav iv \ rj/Jiepci. yvot7/s fita. 
 
 Therefore words like /aeV, Se, ydp, ovv, enim, autem, vero, 
 which do not stand at the beginning of the sentence, aye 
 not usually found at the beginning of the verse. 
 
 Since languages, with increasing age, always pay more 
 regard to logic, in the poets since the Christian era it is 
 the rule that the end of a Strophe like the Elegiac Distich 
 shall coincide with a full stop ; and these poets also, even 
 the Romans, — at least at the end of the verse, — avoid 
 harsh discords between the end of a metrical series and 
 the punctuation or logical connection. 
 
 9. Rhyme. Alliteration. 
 
 In order to emphasize words connected or related to each 
 other by a similar sound, and because this sounded well to 
 them, the Greeks and Romans often placed in the csesura 
 and at the end of the verse words having a similar ending 
 (rhyme, hoinoeoteleutoii), especially a noun and adjective, or 
 appositive, so that usually the last one, but sometimes the
 
 Rhyme. Alliteration. 47 
 
 last two, syllables of each word had the same sound. 
 Thus Homer : — 
 
 co-TTcre vvv ixot Movcrat, 'OXvfXTnxi Sw/tar €;j(ovorai. 
 
 Ovid : — 
 
 quot caelum Stellas, tot habet tua Roma puellas. 
 
 Especially frequent is this assonance in the caesura and 
 at the end of the verse of the dactylic Pentameter and the 
 asclepiadeus minor, e.g. : — 
 
 et teneat cul/z jugera multa so/j 
 terrarum diOWAnos evehit ad defj. 
 
 The laws of Rhyme in later poetry have been developed 
 from this usage. 
 
 The same purpose is served by Alliteration ; i.e. a similar 
 beginning, consisting of one or even two letters, of two or 
 more words which follow each other. 
 
 Among the Romans Alliteration appears frequently from 
 the earUest time until Lucretius. Later, under the influence 
 of the Greeks, who, with the exception of the comedians, 
 were not fond of it, few instances occur, and these for the 
 most part in single formulas, as pater patriae, more viodocpie, 
 etc., or in very artificial verses; in Vergil occasionally, in 
 imitation of Ennius.^ 
 
 ^ The relative merits and present relation of Greek and Roman 
 versification are discussed in the Introduction.
 
 SECOND SECTION. 
 
 -ooJ^Oo- 
 
 ON PECULIARITIES OF THE FOOT. 
 10. Synapheia. 
 
 The last syllable of every verse can be, as has been 
 already said, either long or short. Yet the old poets were 
 fond of ending verses, especially those closing with a thesis 
 or a trochee, with a long syllable or one ending with 
 a consonant. 
 
 When several verses are joined together into one system 
 by Synapheia, — that is, if tlie metre runs without break 
 through the pauses of the verses to the end, — elision and 
 the division of a word can occur at the end of every verse 
 except the last. Hiatus and Syllaba Anceps, however, are 
 not allowed. Especially frequent is Synapheia in lonici a 
 minori, Glyconics, and Anapaests. Horace has Synapheia, 
 but without ehsion and division of a word, in Od. HI. 12; 
 sometimes also in his other lyrical measures the same thing 
 is found. 
 
 11. Resolution of Thesis and Arsis. 
 
 The Thesis of the dactylic verse, and of the logaoedic 
 verse also in Ionic, ^-Eolic, and Roman poetry, cannot be 
 resolved. The Arsis in verses which have more than three 
 feet can be replaced at pleasure by a long syllable. 
 
 The Arsis of the anapaest can always be replaced by 
 a long syllable. 
 48
 
 Resolution of Thesis and Arsis. 49 
 
 In anapaestic, iambic, and trochaic metres the Thesis, 
 except the last, can always be replaced by a pyrrhic. 
 
 From this must be excepted iambic and trochaic verses 
 ending with a spondee, as well as logaoedic metres ; but 
 this, too, apart from the Doric lyric poetry and the lyric 
 passages of the drama. Further, the shorter an iambic 
 and trochaic measure is, the less often does Resolution 
 occur. 
 
 Wherever in iambic metre the spondee is allowed, it 
 is occasionally replaced by an anapaest. This license is, 
 however, found but seldom in the best poets, as the 
 Greek tragedians and Horace. The Greeks allow the 
 dactyl for the spondee in trochaic verse only in proper 
 names. 
 
 Since the end of the verse, which in the Hexameter 
 consists of two feet, in the longer iambic and trochaic 
 measures, as well as in the Paroemiac, of one and a half 
 feet, must be preserved in its original purity, we seldom 
 find a spondee in the next to the last foot of the Hexa- 
 meter or Paroemiac, or a Resolution of the next to the 
 last Thesis in iambic or catalectic trochaic verse.
 
 THIRD SECTION. 
 
 o-oj43):J<>o 
 
 DESCRIPTION OF THE MOST IMPORTANT METRES, 
 STROPHES, AND SYSTEMS. 
 
 12. Dactylic Metres. 
 
 (i) The Dactylic Hexameter [Homer and Hesiod]^ : — 
 
 — WW _ OO _ II wO _ CXJ _ WW — *->' 
 
 I r Lj' I r u' I r I u' I r u' i r u* i r p ^^ i 
 
 The Hexameter was, in the Golden Age of antiquity, 
 the only metre used for Epic poetry {versus heroia/s), and 
 for the responses of the oracles {versus Pythius). It was 
 also the usual metre for Didactic and Bucolic poetry, and 
 for Satire from Horace on ; but it was seldom used in 
 Epigrams other than those of folk-poetry. The tragedians 
 employ it in some places, in especially solemn passages, 
 and it occurs often in the strophes of Archilochus and 
 Horace. 
 
 By the variety of its rhythms and coesuras it is equally 
 adapted to the most different kinds of poetry. The best 
 Hexameters are those which are made up of three dactyls, 
 and the spondees so arranged that the dactyls fall in the 
 first, second, and fifth feet, as in Vergil : — 
 
 arma virumque cano, Troiaeque primus ab oris. 
 
 1 The name added to each metre in brackets indicates the poet by 
 whom it was first used or through whom it first became known. 
 
 50
 
 Dactylic Metres. 5 i 
 
 For too many dactyls make the rhythm too variable 
 and restless ; too many spondees make it stiff and clumsy. 
 Hexameters made up of spondees only do not occur in 
 the poets considered in this book, Homer not excepted 
 (the readings in Od. XXI. 15; XXH. 175 are corrupt). 
 
 As would be expected from the flexibility of the lan- 
 guage, among the Greeks the dactyl preponderates, so that 
 its frequency of occurrence relatively to the spondee is as 
 four to two. 
 
 Homer, and oftener Vergil, employ the different feet for 
 verse-painting, according as the meaning demands more 
 rapidity or slowness of rhythm. Thus in the well-known 
 lines (Od. XI. 598; JEn. VIII. 596; Georg. IV. 174) : — 
 
 avrts eTTttru Treooi'Oe KvXivotTo Aaas di/atSi^s. 
 quadrupedante putrem sonitu quatit ungula campum. 
 illi inter sese magna vi bracchia tollunt. 
 
 Compare also the description of the storm (^-En. I. 81- 
 91 ; 102-123). 
 
 In the fifth foot, especially in Latin, we do not often 
 find the spondee {vcrsi/s spondiazon), particularly in the 
 Distich. In the Satires and Epistles of Horace it occurs 
 only once (Ep. II. 3, 467). 
 
 If a spondee occurs in the fifth foot, a dactyl usually 
 stands in the fourth. 
 
 The best and usual caesura is the Penthemimeral, after 
 the third Thesis : — 
 
 fj.rjvLv aeiSe, 6m, \ JlrjXrjLaSeo) 'A^iA^os. 
 arma virumque cano | Troiaeque primus ah oris. 
 
 Not less frefiuent in Homer and most of the Greek 
 poetSj much more frequent in Nonnus, is the to/mt/ Kara
 
 52 Metres, Strophes, and Systems. 
 
 TpiTov Tpoxoiiov {i.e. after the second syllable of the third 
 dactyl) : — 
 
 avopa fxoi evvCTre, MoiJcra, | TroAurpoTrov, os fxaXa iroXXd. 
 
 On the other hand, this occurs but seldom among the 
 Romans. With them the caesura which occurs most fre- 
 quently after the Penthemimeral is the Hephthemimeral, 
 which is rare in Greek (/En. I. 251) : — 
 
 navibus infandum ! amissis | unius ob iram. 
 
 In Latin the Penthemimeral Caesura is by far the most 
 frequent, especially so in the Distich, and in general in the 
 most polished poetry. 
 
 Every Hexameter verse which has not one of these 
 caesuras is faulty, and such do not occur in the poets con- 
 sidered in this book (the very few examples to the contrary 
 are wholly corrupt readings). ■• 
 
 In addition to the Penthemimeral Caesura and the to/x,-^ 
 Kara rpiTov rpo)(a1ov, the so-called BiicoUc Caesura {i.e. the 
 caesura especially used in Bucolic poetry) after the fourth 
 dactyl occurs, but only in Greek (II. XXIII. 549; Od. 
 I. I):- 
 
 £(7Tt TOt ev KXicrir] | )(^pvcr6<; ttoXv';, | Ictti oe ^aXKOs. 
 avSpa /xoL evveire, Wovaa, | TroXvTpoTrov, | os /xdXa TroXXd. 
 
 In cases where such verses occur in Latin, as Bucol. 
 10, II, it is to be assumed that there is no caesura at all. 
 
 As the simple Hephthemimeral Caesura divides the verse 
 into two unequal parts, there is frequently found both in 
 Greek and Latin, as a sort of support, the Trithemimeral 
 Caesura (after the second Thesis). So II. I. 145 : — • 
 
 t] Attts I rj 'lSo/A€vews | y Sios OSuo-crew.
 
 Dactylic Metres. 53 
 
 When the end of a word falls both in the second syllable 
 of the third foot and in the fourth Thesis, in the Greek 
 Hexameter the to/at/ Kara. rpiTov Tpo-^alov is employed, and 
 in Latin the Hepthemimeral Caesura, without regard to the 
 punctuation (Od. IV. 126 ; ALn. IV. 582) : — 
 
 'AXKavSprj TloXvjSoio | Sa/xap, 05 ci^at' ivl 0?//3>j9. 
 
 litora deseruere. latet | sub classibus aequor. 
 
 The Hephthemimeral Caesura in Latin is most commonly 
 found when at the same time the Trithemimeral occurs, 
 and there is the end of a word in the third trochee : — 
 
 infandum | rej^uia jubes | renovare dolorem. 
 
 In this place, as at the end of the verse, too long words 
 were avoided. 
 
 Finally, wherever in Latin Hexameter a word ends with 
 the third Thesis, and after the fourth a decided stop occurs, 
 the Hephthemimeral may be regarded as the proper caesura: — 
 
 oscula libavit natae ; | dehinc talia fatur. 
 
 When a word ends in the third trochee, a word ending 
 also in the second or fourth trochee is not good, although 
 in one case Horace has allowed himself both in one line : — 
 
 dignum niente domoque | \<igcntis honesta Neronis. 
 
 In a case like this when the Trithemimeral Caesura does 
 not occur, the best Latin poets are accustomed to unite 
 the second and third feet in one word : — 
 
 Mnesthea Scrgestunujue vocat | fortemque Serestum. 
 
 The Greeks in general very seldom place a trochaic 
 word in the fourth foot, doubtless because they employ 
 a trochee in the third foot so frequently.
 
 54 Metres, Strophes, and Systems. 
 
 (3) The Pentameter (Callinus and Archilochus) is made 
 up by doubling the first half of the Hexameter, which 
 has the Penthemimeral Caesura, though the spondee is 
 allowed only in the first half: — 
 
 \ o 1 \ * \ \ \ 
 
 11 ill '^11 L^ 1 1 i 
 
 The most pleasing effect is produced when a spondee 
 occurs only in the second foot : — 
 
 et tenert/ culi\ jugera multa soli. 
 
 The caesura occurs always after the third Thesis. 
 
 The Distich is made up by the combination of the 
 
 Pentameter (which almost never occurs alone) and the 
 
 Hexameter. 
 
 / / / I / — / ( — ) / 
 
 (3) The Tetrameter (Archilochus): — 
 
 I I 1 
 
 r0 \ 
 
 This metre is employed by Horace and Archilochus in 
 asynartete verse with a following trochaic Tripody, The 
 last syllable is always short in Horace, but not in Archilo- 
 chus. This verse also occurs independently in Greek and 
 Roman lyrical poetry, in which case the dactyl greatly pre- 
 ponderates. The caesura is the Penthemimeral. 
 
 (4) Tetrameter Catalectic /// dissyllalnim (Archilochus) : — 
 
 _ ^.J\^ _ v_/vy -— \^ \J — \J 
 
 r
 
 AnapcEstic JlTetrcs. 55 
 
 In Horace this is a part of the epodic and lyrical 
 strophes. Only once (I. 28, 2) is a spondee found in the 
 third foot, and then in a proper name. 
 
 (5) Trimeter Catalectic /// syllabani (Archilochus) : — 
 
 / / / 
 
 r u \ LI r 
 
 In Horace combined with the heroic Hexameter in the 
 Odes. Elsewhere in the Epodes it is combined with the 
 Iambic Dimeter into asynartete verse. 
 
 (6) Dimeter Catalectic in dissyllabutn, versus adonius 
 (Sappho) : — 
 
 \ LI \ ' 
 
 This forms the close of the Sapphic Strophe. 
 
 13. Anapeestic Metres. 
 
 These do not occur in the Latin authors considered here, 
 but frequently in the Greek dramatists. 
 
 (7) Anapaestic Dimeter: — 
 
 J J J J i i i 
 
 LJ lJ lJ U I Lj' Lj LI r 
 
 In place of an anapaest the spondee, dactyl, and 
 proceleusmatic are allowed. 
 
 The poets usually avoid the proceleusmatic, as well for 
 one foot as for the Thesis and Arsis of two successive
 
 56 Metres, Strophes, and Systems. 
 
 feet, although cases are found where three proceleusmatics 
 
 follow each other : — 
 
 / / ^ / 
 
 \j \j \j \j \j \j \j \j \j \j \j \j w\^_ 
 
 I I 
 
 ; u r 
 
 The following form is usual : — 
 
 \J \J . _ \J ^u 
 
 In solemn and mournful poetry, as in the marching hymns 
 of the Spartans, the spondee was employed by preference. 
 The caesura after the first Dipody is not always observed. 
 
 (8) Anapaestic Monometer : — 
 
 / c / \ 
 
 u r 
 
 This measure is often inserted in Anapaestic Systems. 
 
 Since the Anapasstic verse is employed in systems with 
 Synapheia, Hiatus and Syllaba Anceps are only allowed at 
 the end where there is a change of person, and more 
 rarely at a stop. Elision and Resolution of the last Thesis 
 are allowed. Notwithstanding the Synapheia, however, Dim- 
 eters and Monometers generally end with a complete word. 
 
 (9) The close of the Anapaestic System is usually formed 
 by an Anapaestic Dimeter Catalectic inversus paroemiacus) : — 
 
 
 
 '^1 LJ I 11 \f 
 
 The rule is that this measure should have the form of 
 the second half of the Hexameter after the Penthemimeral 
 Caesura. Hence it is rare that the Theses (especially the 
 last) are resolved, or a spondee placed in the next to the 
 last foot.
 
 Iambic Metres. 57 
 
 14. Iambic Metres. 
 
 (10) Iambic Trimeter versus senarius (Archilochus) : — 
 
 \J-^\^.L— \J1— ^J-^\U.L— \^ 
 
 P 
 
 \^ \ \> \ \^ \ \> \ \^ \ ^^ \ 
 
 without Resolution or spondees. In Horace used with the 
 dactyUc Hexameter. 
 
 (v-yw) (ww) (ww) 
 
 P'or Phsedrus' Iambics, see below. 
 
 This is the most frequently used verse in the dialogue of 
 Tragedy and Comedy and in the Epodes of Horace, and 
 occasionally employed in other poetry, especially satirical 
 and sarcastic. It is the favorite and most beautiful metre 
 after the Dactylic Hexameter. 
 
 Spondees are particularly numerous in the Trimeter of 
 Tragedy, especially in ^schylus. In general, here and in 
 what follows, we are concerned only with the Iambic and 
 Trochaic verses of the dialogue, not with those of the lyrical 
 passages, which have their own peculiar licenses in Resolu- 
 tion, Caesura, etc. Resolution of the Thesis occurs rarely in 
 Archilochus, and is confined to the first syllable of words. 
 
 In ^schylus and the older plays of Sophocles, Resolu- 
 tion of the Thesis in the second, third, fourth, or fifth 
 foot occurred more rarely, and then generally in the begin- 
 ning of words of three or more syllables ; less frequently 
 in the case of dissyllabic words except prepositions, or 
 monosyllables which belong closely to the following word 
 beginning with a short syllable, e.g. the article. Euripides, 
 in whose poetry Resolution is oftenest found, allows himself 
 to unite a short monosyllable with a following short syllable 
 when both words do not belong together.
 
 $8 Metres, Strophes, and Systems. 
 
 In words of four or more syllables in the second and 
 fifth feet inclusive, Resolution through the two last sylla- 
 bles occurs, but in Euripides only through the middle 
 syllables. In the first foot it is naturally the second and 
 third syllables of a word of three or more syllables which 
 are employed in a Resolution of the Thesis. Yet Sophocles 
 and chiefly Euripides (even ^schylus when the first foot 
 is a dactyl) allow the verse to begin with a monosyllable. 
 A Resolution is not allowed when its first syllable forms 
 the close of a polysyllabic word. 
 
 Resolution occurs most frequently in the third Thesis 
 (after the caesura) and in the first, most rarely in the fifth. 
 A dactyl instead of a spondee occurs only in the first 
 and third feet. 
 
 ^schylus has seldom two cases of Resolution in one 
 verse ; Euripides has not seldom even three. 
 
 Horace never divides a Resolution into two words, and 
 out of the first foot, he uses it only twice in a dissyllable 
 (Ep. 2, 23; 5, 87). A word consisting of three short 
 syllables in Latin is used in the place of a trochee, never 
 in place of an iambus, as is customary in Greek. Therefore 
 in Latin verse it is possible to scan genera., never genera. 
 
 In the same way the two last short syllables in words 
 of more than three syllables, e.g. materJ^, are not used 
 in a Resolution. Horace does not even use in this way 
 the last syllables of a dactylic word like xbbord. He never 
 resolves the fifth Thesis ; the tribrach occurs most fre- 
 quently in the second foot, and he rarely has two Resolu- 
 tions in one verse, almost never three (cf. Ep. 17, 12). 
 
 The anapaest can stand instead of the spondee in the 
 first foot in tragedy, and for this a word of three or more 
 syllables is generally used. Elsewhere the anapaest is
 
 Iambic Metres. 59 
 
 allowed only in proper names, but (in such names) in 
 all feet except the last. 
 
 Horace has an anapsest twice in the first foot, three 
 times in the fifth (Ep. 2, 35; 65; 3, 35; 5, 79; 11, 23), 
 always in a word of at least three syllables. 
 
 The Arsis of an anapsest can never be formed by the 
 two last syllables of a word of more than two syllables, or 
 by the final of a polysyllable and the first syllable of the 
 following word. The proceleusmatic is not allowed. 
 
 As in the Dactylic Hexameter, the most usual caesura 
 is the Penthemimeral after the third Arsis : — 
 
 \-/ \J \J I \J \J v,^ 
 
 ibis Liburnis | inter alta navium ; 
 Next the Hephthemimeral after the fourth Arsis : — 
 
 \J \J \J ^u I \J \J 
 
 nam qualis aut Molossus | aut fulvus Laco. 
 
 The Penthemimeral Caesura preponderates in Euripides 
 more than in ^schylus and Sophocles, and most of all in 
 Horace. 
 
 Since the Hephthemimeral Caesura divides the verse too 
 unequally, it is especially used when the second foot ends 
 with the end of a word : — 
 
 nam quahV aut Molossus aut fulvus Laco ; 
 
 or when the second and third trochees of the verse are 
 formed by one word : — 
 
 quae sidera excantata voce Thessala. 
 
 Finally, when the end of a word comes after the second 
 trochee, and a stop after the third, the Hephthemimeral 
 Caesura is to be assumed : — 
 
 quid dixit aut quid tacuit? | o rebus meis.
 
 6o Metres, StropJies, and Systems. 
 
 The rules of the Hephthemimeral Caesura in the dactyUc 
 Hexameter are quite similar (cf. 12). 
 
 A Trimeter without either the Penthemimeral or Hephthe- 
 mimeral Caesura is faulty. Such verses do occur in the 
 tragedies, but not in Horace. In such cases a word usually 
 ends with the third foot. 
 
 Finally, it is to be noticed that when the fifth foot is 
 a spondee, the tragedians are not accustomed to use the 
 final of a polysyllabic word as its Arsis, except when an 
 enclitic follows, or a monosyllable closely connected with 
 the preceding word, as e.g. : — 
 
 aTreuSw/Aev, iyKovwixev' rjyov fxoi, yepov. 
 olov re fJiOL racrS' iart. 6vy]T0L<i yap yepa. 
 
 Otherwise the last cretic would be too forcibly separated 
 from the rest of the verse. Horace has observed this rule 
 only in Ep. 1 7. 
 
 Iambics of Ph^drus. 
 
 Phgedrus allows the spondee or the anapaest in every 
 foot except the last. 
 
 In the fifth foot the spondee or anapaest predominate ; 
 the iambus is admitted principally when a four-syllabled 
 word stands at the end of the line : — 
 
 ranae vagantes liberis pahuiibus. 
 
 In no case does a verse close with two iambic words. 
 
 The tribrach occurs only in the second, third, and fourth 
 feet ; the dactyl generally in the first, third, and fourth. 
 The Resolution of the Thesis follows in other respects the 
 same rules as in Horace, except that pyrrhic words are 
 more frequently employed, very rarely monosyllables, which
 
 Iambic Metres. 6 1 
 
 are usually closely connected with the following short 
 
 syllable : — 
 
 calumniator ab ove cum peteret canis. 
 
 The verse in app. lo, lo, is an isolated case and probably 
 to be emended : — 
 
 non ut labores facio, sed ut istum domes. 
 
 If the fifth Thesis is resolved, a four- syllabled word at 
 least must stand at the end of the verse. The verses V. 7, 
 22 ; app. 9, 6, hardly constitute an exception. 
 
 The anapaest stands everywhere in place of the spondee ; 
 the two never follow each other immediately. Except in 
 the first foot and (very seldom) in the fifth (III. 10, 4; 
 14, 11; app. 19, 3; 30, 10), Ph?edrus follows exactly the 
 same rules as Horace for the anapisst. 
 
 In the first foot the Arsis is almost always formed by a 
 dissyllable or the beginning of a polysyllable. The proce- 
 leusmatic occurs only in the first foot ; in such wise that the 
 Arsis always forms a word for itself, and the Thesis does 
 the same, or at least forms the pyrrhic beginning of a 
 word, e.g. : — 
 
 i/a capttt ad nostrum furor illorum pertinet. 
 itaque hodie nee lucernam de flamma deum. 
 
 The caesura is exclusively the Penthemimeral or (more 
 rarely) the Hephthemimeral. 
 
 (11) Catalectic Trimeter (Archilochus) : — 
 
 KJ /- \J — \D \ — \J ^ \J ^ \J 
 
 » \ 
 
 This is used by Horace in the Odes. Resolution of the 
 Thesis and the anapaest are not allowed ; for the reading
 
 62 Metres, StropJics, and Systems. 
 
 in II. i8, 34, is certainly corrupt. Tlie caesura is always 
 the Penthemimeral, 
 
 (12) Iambic Dimeter Hypercatalectic (Alcgeus) : — 
 
 (w) / / 
 
 . — _ v^ -r_ . 
 
 r r ' r r r * r c 
 
 This is used by Alcseus and Horace as the third verse 
 of the Alcaic and logaoedic strophe, and therefore the 
 Thesis is never resolved. This verse occurs only in Horace 
 in the form given above. In Alcseus the first and fifth 
 syllables can be either long or short. The Anacrusis is 
 generally long in the first three books of Horace's Odes, 
 and always long in the fourth. 
 
 (13) Iambic Dimeter (Archilochus) : — 
 
 Used by Archilochus and Horace as the second verse 
 of the Epode, and in asynartete verses ; by the other 
 poets also as a separate measure. 
 
 An iambus seldom occurs in the third foot. Resolution 
 occurs in Horace only twice, once in the first foot, and 
 once in order to produce a rhythmical picture (Ep. 15, 24; 
 2, 62). 
 
 15. Trochaic Metres. 
 
 (14) Trochaic Tetrameter Catalectic (Archilochus): —
 
 Trochaic Metres. 63 
 
 This metre is common in the Greek and Latin drama, 
 and resembles closely the Iambic Trimeter, except that 
 it is much more strictly formed in the Greek tragedy. 
 Resolution of the Thesis takes place most frequently in 
 the uneven feet and in the first half of the verse — very 
 seldom in the seventh Thesis. 
 
 In resolving the Arsis, Euripides first dared to make use 
 of dissyllabic words, or the penultimate and final syllables of 
 trisyllables. Very rarely do two monosyllables occur in the 
 Resolution, if they are not closely connected with a follow- 
 ing short syllable. The dactyl in place of the spondee is 
 not allowed. The caesura falls in tragedy without exception 
 after the second dipody (^sch. Pers. 164; Soph. Philoct. 
 1402 are corrupt readings). 
 
 If the sixth foot is a spondee, it cannot be a dissyllable 
 or the end of a polysyllable. 
 
 (15) Trochaic Dimeter Catalectic : — 
 
 r ' r ' r ' r 
 
 This form occurs in Horace, Od. II. 18 ; also frequently 
 in the lyrical parts of tragedy, and usually just as here 
 given. 
 
 (16) Trochaic Tripody, versus ithyphallicus (Archilo- 
 chus) : — 
 
 r 
 
 Used by Archilochus and Horace in asynartete verses 
 with a preceding Dactylic Tetrameter.
 
 64 Metres, Strophes, and Systems. 
 
 16. lonici a Minori. 
 (17) Decameter (Alcaeus) : — 
 
 \j \j —— — \y w — — w \J — - _ \y \j — _ I 
 
 This is used by Horace, in imitation of Alcaeus, in the 
 Odes. The caesura falls after the fourth and eighth feet. 
 
 17. Logaoedic Metres. 
 
 (18) Glyconic (Sappho, Anacreon) ; in Horace in the 
 following form : — 
 
 r ' I ' * b ' I r p I r * (°m' •? )i 
 
 The Glyconic metre was originally a logaoedic series 
 
 with a two-syllable Basis of any convenient quantity. 
 
 This license belonged also to the Pherecratean and 
 Asclepiadean verses, which were derived from the Glyconic. 
 Catullus, the predecessor of Horace, used as the Basis of 
 the Glyconic and Pherecratean, generally the trochee, more 
 rarely the spondee and iambus. Horace, however, uses 
 the spondee exclusively as the Basis of all these verses, as 
 Catullus does in his Asclepiadean verses. 
 
 The tragedians allow in the Basis of the Glyconic verse 
 beside the trochee, spondee, iambus, and pyrrhic, also the 
 tribrach, and Euripides even admits the anapaest. 
 
 Sophocles, and much more frequently Euripides, created 
 very varied forms of the Glyconic {glyconei polyschematisti) 
 by means of a displacement of the dactyl and other 
 licenses,
 
 Logacedic Meters. 65 
 
 (19) Pherecratean (Sappho, Anacreon) : — 
 
 r f I ' b ' I r • I r •? i 
 
 This metre is not used alone, but only in combination 
 with Glyconic and Asclepiadean verses. 
 
 (20) Lesser Asclepiadean (Alcaeus) : — 
 
 j1 ^ >.^ ji w w ^ 
 
 r0 I •: a \ » ' 10' o \ \ ' I 
 
 This verse is formed by the insertion of one choriambus 
 after the Basis, as the following verse is formed by the 
 insertion of two. 
 
 The caesura after the sixth foot is always observed by 
 Horace, though sometimes neglected by Alcseus. 
 
 (21) Greater Asclepiadean, dodecasyllabic (Alcaeus) : — 
 
 Cf. No. 20. The caesura after the sixth and tenth sylla- 
 bles is often neglected by Sappho, Alcseus, and Catullus, 
 never by Horace. 
 
 (22) Greater Sapphic (Sappho): — 
 
 ^ \j Z- Z- I \j \j /- \j Z^ \j 
 
 Two trochaic dipodies, divided by a dactyl. Horace 
 employs always a spondee in place of the second trochee. 
 In Sappho the fourth syllable is either long or short, and 
 Catullus has sometimes a short syllable here. The caesura 
 is generally after the third Thesis, occasionally (especially 
 in the fourth book of the Odes and the Carm. Saec.) after
 
 66 Metres, Strophes, mid Systems. 
 
 the trochee of the third foot. This caesura is often neg- 
 lected by Sappho — by Catullus only twice. 
 
 (23) Lesser Sapphic, aristophaneus (Sappho): — 
 
 _^ w \j —^ ~ \j 
 
 (24) The so-called Sapphic verse of fifteen syllables is 
 formed by a combination from the beginning and end of 
 the Greater Sapphic, but this occurs first in Horace : — 
 
 \^\jJL\JL^^ — \J — ^ 
 
 The fifth and eighth syllables are always final. 
 (25) Alcaic Hendecasyllabic (Alcaeus): — 
 
 ^^^^wji_ljiwwZ.^Z. 
 
 Logaoedic Metre with Anacrusis, which is common in 
 Alcseus. In Horace the Anacrusis is generally long in the 
 first three books of Odes and always so in the fourth. 
 Alcseus has often the original trochee instead of the spondee 
 before the csesura. The caesura is sometimes neglected by 
 Alcseus, but only twice by Horace (I. 37, 14; IV. 14, 17, 
 though these verses are probably corrupt). 
 
 (26) Alcaic Decasyllabic (Alcseus) : — 
 This forms the close of the Alcaic Strophe. 
 
 18. Asyuartete Verse (Mixed Measures). 
 (27) Greater Archilochian (Archilochus) : — 
 
 A combination of the Dactylic Tetrameter (No. 3) and 
 the Versus Ithyphallicus (No. 16).
 
 The Elegiac Distich. 67 
 
 (28) Elegiambic (Archilochus) : — 
 
 A combination of the Dactylic Trimeter Catalectic 
 (No. 5) and the Iambic Dimeter (No. 13). Short sylla- 
 bles and Hiatus occur in the third Thesis. 
 
 (29) lambelegiac : — 
 
 A combination like the preceding, but with the order of 
 the measures reversed. This verse, which possesses much 
 force and liveliness because of the transition from the 
 Iambic to the Dactylic measure, is first found in Horace. 
 A short syllable often occurs in the fourth Thesis. 
 
 19. The Elegiac Distich (Callinus, Archilochus). 
 
 / — / — / I — / — / ( — ) / 
 
 This is the oldest and most beautiful verse-system of the 
 Greeks ; a combination of the Dactylic Hexameter and 
 Pentameter (Nos. i and 2). It was employed mostly in Epi- 
 gram and Elegy. It is somewhat less appropriate for Didactic 
 Poetry, though used by Ovid in the Fasti and by Propertius. 
 
 It was a favorite metre among the Greeks, and still more 
 so among the Romans, who developed it with marvellous 
 artistic skill. 
 
 20. The Lyric Strophes of Horace.' 
 
 These Eyrie Strophes of Horace are all composed of four 
 lines, perhaps in imitation of Alcgeus, but not of Archilochus. 
 
 1 Here are given, just as in the case of the Distich, the most usual 
 schemes of the individual verses. For everything else, see the discussion 
 in the preceding section.
 
 68 Metres, StropJics, and Systems. 
 
 At the close of single verses Syllaba Anceps (sometimes with 
 a final consonant) often occurs, more rarely Hiatus. Some- 
 times at the ends of the first three verses we find a word 
 divided or Elision. In both these cases the final syllable is 
 always long. 
 
 (i) Alcaic Strophe (Alc?eus) : — 
 
 (-^ ^ w -^ _ 
 
 (-) ^ w ^ _ 
 
 This is made up of the Alcaic Hendecasyllabic (No. 25), 
 the Enneasyllabic (12), and the Decasyllabic (26), and is 
 characterized by force and energy. Therefore it is Horace's 
 favorite metre, and employed by him especially in Odes of 
 political and moral content, though also in those treating 
 of erotic and convivial subjects. It is the metre of 37 
 Odes: I. 9. 16. 17. 26. 27. 29. 31. 34. 35. 37; II. I. 3. 5. 
 7. 9. II. 13-15. 17. 19. 20; III. 1-6. 17. 21. 23. 26. 29; 
 IV. 4. 9. 14. 15. 
 
 Elision at the end of the verse occurs in II. 3. 27; III. 
 29- 35- 
 
 (2) Sapphic Strophe (Sappho): — 
 
 / 
 
 \J 
 
 / 
 
 
 
 / 
 
 \ w 
 
 KJ ^ 
 
 W 
 
 / 
 
 \J 
 
 /_ 
 
 w 
 
 / 
 
 
 
 / 
 
 1 ^ 
 
 W I- 
 
 w 
 
 / 
 
 \u 
 
 / 
 
 \J 
 
 / 
 
 
 
 / 
 
 I w 
 
 W ^ 
 
 Vw/ 
 
 / 
 
 \J 
 
 / 
 
 w 
 
 KJ 
 
 / 
 
 KJ 
 
 
 
 
 
 
 Made up of the Sapphic Hendecasyllabic (22) and the 
 Adonic (6). This strophe has more grace and tenderness 
 than force and energy. Hence it would, perhaps, have 
 been better if Horace had not used it so often in Odes 
 written in lofty style. It occurs in 26 Odes: I. 2. 10. 12.
 
 Lyric StropJus of Horace. 69 
 
 20. 22. 25. 30. 32. T^^; II. 2. 4. 6. 8. 10. 16; III. 8. II. 
 14. 18. 20. 22. 27; IV. 2. 6. II ; Carm. Saec. 
 
 Elision at tlie end of a verse occurs in II. 2, 18; 16, 34; 
 IV. 2, 22 ; 23 ; Carm. Saec. 47. The division of a word, 
 in each case at the end of the third verse, occurs four 
 times: I. 2, 19; 25, 11; II. 16, 7; III. 27, 59. 
 
 (3) Second Sapphic Strophe: — 
 
 Made up of the Lesser Sapphic (23) and the so-called 
 fifteen-syllabled Sapphic (24). i, 8. 
 
 (4) First Asclepiadean Strophe : — 
 
 Made up of the Lesser Asclepiadean repeated (20). I. i ; 
 III. 30; IV. 8. 
 
 (5) Second Asclepiadean Strophe: — 
 
 / 
 
 / 
 
 ^ 
 
 KJ 
 
 / 
 
 1 ^W 
 
 W 
 
 / 
 
 \J 
 
 / 
 
 /_ 
 
 / 
 
 w 
 
 \y 
 
 / 
 
 1 ^W 
 
 \J 
 
 / 
 
 \J 
 
 / 
 
 / 
 
 / 
 
 w 
 
 K^ 
 
 / 
 
 i ^w 
 
 vy 
 
 / 
 
 \J 
 
 ^ 
 
 / 
 
 / 
 
 v^ 
 
 \J 
 
 / 
 
 \J 
 
 
 
 
 
 Made up of the Lesser Asclepiadean (20) and the 
 Glyconic (18). The metre of nine Odes: I. 6. 15. 24. 
 II ; II. 12 ; III. 10. 16; IV. 5. 12. 
 
 (6) Tliird Asclepiadean Strophe : — 
 
 Made up of the Lesser Asclepiadean (20), the Phere-
 
 70 Metres, Strophes, and Systems. 
 
 cratean (19) and the Glyconic (18). Occurs in seven 
 Odes: I. 5. 14. 21. 23; III. 7. 13; IV. 13. 
 
 (7) Fourth Asclepiadean Strophe: — 
 
 / 
 
 / 
 
 ^u 
 
 W 
 
 / 
 
 W ^ 
 
 
 
 
 
 / 
 
 / 
 
 \J 
 
 W 
 
 / 
 
 1 ^ 
 
 \y 
 
 \y 
 
 ^ w 
 
 / 
 
 z. ._ 
 
 / 
 
 KJ 
 
 ^ 
 
 / 
 
 ^ z. 
 
 
 
 
 
 / 
 
 / 
 
 \J 
 
 W 
 
 / 
 
 1 ^ 
 
 \y 
 
 V-/ 
 
 ^ w 
 
 / 
 
 Made up of the Glyconic (18) and the Lesser Asclepia- 
 dean (20). This is the metre of 12 Odes: I. 3. 13. 19. 36; 
 
 III. 9. 15. 19. 24. 25. 28; IV. I. 3. Elision at the end 
 of a verse occurs in IV. i, 35. 
 
 (8) Fifth Asclepiadean Strophe (used by Sappho and 
 Catullus, but in couplets) : — 
 
 Made up of Greater Asclepiadean verses (21). I. 11. 18; 
 
 IV. 10. • 
 
 (9) First Archilochian Strophe : — 
 
 j^ CX3 j£. cx:; j1 | 00 j^ 00 ^ \j \j ^ \j 
 -^ ^ \j /- ^u \j Z- 
 
 Z. Cv:> /- CyO /^ I kKj — kKj ^ \J KJ — kj 
 
 — \^ KJ — \^ KJ jI. 
 
 Made up of the Dactylic Hexameter (i) and Catalectic 
 Trimeter (5). IV. 7. 
 
 (10) Second Archilochian Strophe: — 
 
 _ 00 _ C^ _ I C70 _ CXy _ Kj'^ — KJ 
 
 ^ 00 jL cv^ ^ I cx^ -^ ca:> J^ kjkj — kj 
 Made up of the Dactylic Hexameter (i) and the Cata-
 
 Epodic Systems. 71 
 
 lectic Tetrameter (4). i, 7, 28, Also used in Epode 12, 
 but in couplets. 
 
 (11) Third Archilochian Strophe (used by Archilochus 
 in couplets) : — 
 
 /_ wO — WW /- I WW /^ \J \J 
 
 Made up of the Greater Archilochian (27) and the 
 Iambic Trimeter Catalectic (n). I. 4. 
 
 (12) Hipponactean Strophe: — 
 
 A combination of the Trochaic Dimeter Catalectic (15) 
 and the Iambic Trimeter Catalectic (n). II. 18. 
 
 (13) Strophe of lonici a minori : — 
 
 Kj \j — — \j \^ — — \^ '^ — — \y \j — — I Kj \j S- S- \j \j _ _ 
 
 Made up of the lonicus a minori Decameter (17) 
 (repeated). III. 12. 
 
 Syllaba Anceps and Hiatus are not allowed except at the 
 close of the Strophe. 
 
 21. Epodic Systems. 
 
 The Epodes of Horace are written in couplets, except 
 the last, which is made up of Iambic Trimeters Kura
 
 72 Metres, Strophes, and Systems. 
 
 (i) Iambic System (Archilochus) : — 
 
 /_ 
 
 Made up of the Iambic Trimeter (iO(^) and the Iambic 
 Dimeter (13). Ep. i-io. 
 
 (2) First Archilochian System: — 
 
 Tj (ww) ^^c: |Z.^Z. "cr ^ \j A. 
 
 / 
 
 \J^ 
 
 — \J \J — I \^J— \jZ-\-/ — \J — 
 
 Made up of the Iambic Trimeter (10/^) and the Elegi- 
 ambic (28). Ep, 11. 
 
 (3) Second Archilochian System : — 
 
 Kj _ ^j _ O _ v^ _ I __ \u \y — \^ w — 
 
 Made up of the Dactylic Hexameter ( i ) and the lam- 
 belegiac (29). Ep. 13. 
 
 (4) Third Archilochian System; — 
 
 _ 00 — 00 -^ I 00 _ CaIP — \J \J — \J 
 
 Made up of the Dactylic Hexameter (i) and the Dac- 
 tylic Tetrameter Catalectic (4). Ep. 12. Cf. also Strophe 
 10 of the Lyrical Metres, 
 
 (5) First Pythiambic System: — 
 
 _ 00 — CO — I CO — 00 — \J \J ^ KJ 
 
 Made up of the Dactylic Hexameter (i) and the Iambic 
 Dimeter (13). Ep. 14. 15.
 
 Epodic Systems. 73 
 
 (6) Second Pythiambic System : — 
 
 Made up of the Dactylic Hexameter (i) and the pure 
 Iambic Trimeter (loa). Ep. 16.
 
 FOURTH SECTION. 
 
 -O-oJSt^CKJ- 
 
 ON METRICAL LICENSES. 
 22. Preface. 
 
 The structure of the verse is determined by fixed laws, 
 which, however, are sometimes neglected or evaded. 
 
 Still such exceptions are almost never arbitrary in the 
 classical writers; rather do they, too, fall under definite 
 rules, which are of only less wide application than the 
 general laws of the verse. They are, so to speak, disso- 
 nances resolved into a higher consonance. 
 
 23. Metrical Licenses. 
 
 The metrical licenses and peculiarities of the poets can 
 be reduced to eight cases : — 
 
 (i) The beginning of every metrical series has greater 
 freedom than the end of the series formed either by the 
 caesura or the close of the verse. 
 
 (2) Long verses enjoy greater freedom than short; so 
 long poems offer greater opportunity for metrical licenses 
 than shorter ones which must be especially characterized 
 by elegance and beauty of form. 
 
 (3) Variation of subject-matter often introduces varia- 
 tion of metrical laws, especially among the Greeks. Thus 
 epic poetry, as well as didactic, satirical, and elegiac, has 
 certain peculiar forms of the Hexameter. In the same 
 
 74
 
 Metrical Licenses. 75 
 
 way, the lyric versification has its pecuUarities, and tliat 
 of comedy often differs widely from that of tragedy. 
 
 (4) Further, it is clear that the inventor of a metre, or 
 one who introduces innovations therein, treats it more 
 freely than later writers who follow the path which he has 
 opened up. The Hexameter of Ennius differs widely from 
 that of Vergil ; the lyric measures of Horace from those of 
 Seneca. In general, it is true that the more frequently any 
 metre is used, the more exact is its structure. 
 
 (5) The later works of an author are usually more 
 polished than the earlier. For instance, the Hexameters 
 of Horace's Epistles are more carefully formed than those 
 of the Satires. 
 
 (6) As the metrical art of the famous poets of antiquity 
 continued to be the standard for those who came later, or at 
 least exercised great influence on them, it is very important 
 to observe what model of versification each poet has fol- 
 lowed. The Hexameters of the post-Augustan poets vary 
 according as they follow Vergil or Ovid. 
 
 (7) Frequently the metrical licenses of a verse can be 
 explained by the occurrence in the verse of proper names, 
 or (in I>atin) of Greek words, especially immediately before 
 such words ; for proper names cannot be arbitrarily 
 rewritten or altered. It is to be remembered, too, that the 
 ancient poets employed these names much more frequently 
 than those of modern times. Moreover, Greek words 
 appear in Latin verse to justify of themselves, at the same 
 time, the metrical licenses borrowed from the Greeks. In 
 didactic poetry the so-called termini tecJinici, in the Chris- 
 tian writers sacred words as spiritus, eccksia, usually give 
 rise to metrical licenses, for these words approach nearly 
 to proper names. In the same class, too, fall sometimes
 
 76 Meti'ical Licenses. 
 
 words of four or more syllables, as well as the most com- 
 mon pronouns, adverbs, prepositions and conjunctions, and 
 certain fixed expressions like "^ ow," "ergo age," " quare 
 
 age." 
 
 (8) Finally metrical licenses are not seldom occasioned 
 by an impassioned movement of the language which shows 
 itself in rhetorical figures, especially in the repetition of the 
 same word (anaphora) or in antithesis. 
 
 It is often true that several of these reasons combine to 
 explain a single case of metrical license. 
 
 In Greek poetry metrical licenses are much more fre- 
 quent than in Latin, although there are cases where several 
 of them occur in a single Latin Hexameter when some of 
 the above-mentioned excuses are present. Thus the verse of 
 Vergil : — 
 
 Nereidum matii et Neptuno Aegaeo, 
 
 contains two cases of hiatus — a spondee in the fifth foot, 
 and a violation of the rhythmical laws of this same foot.
 
 FIFTH SECTION, 
 
 -»o>0<oo- 
 
 ON THE RHYTHMICAL STRUCTURE OF THE VERSE. 
 24. General Remarks. 
 
 (i) As poetry, at least in its higher forms, seeks to differen- 
 tiate itself as widely as possible from the ordinary methods 
 of expression in prose, the poets avoid allowing the rhythm 
 of the verse to coincide with the prose accent of the words, 
 as far as it is at all practicable. It is likewise considered 
 improper to introduce a verse into prose. 
 
 The first law, then, is that there shall be the greatest 
 possible difference between the metrical rhythm and the 
 prose accent. This law naturally has the least force in those 
 parts of the verse which admit the greatest freedom of 
 structure, as at the beginning of a metrical series, at the 
 beginning of the verse, or after the caesura, as e.g. in Vergil : — 
 
 litora, multum ille et terris jactatus et alto, 
 ipse hostis Teucros | iitsigiii laude ferebat. 
 
 This law is further modified by the following rules. 
 
 (2) As the verse becomes in the first instance a work 
 of art by the close and harmonious combination of the 
 feet, it is not usual, at least in the longer verses, to allow 
 single feet to consist of single words. Consequently the 
 following Hexameter from I-^nnius is bad : — 
 
 sparsis hostis longis campus splendet et horret. 
 
 77
 
 ^8 RJiytliniical StJ'uctnre of the Verse. 
 
 In order therefore that the verse may not fall apart, the 
 individual feet must run into each other as much as possi- 
 ble, and this result is obtained most effectually by the 
 greatest possible difference between the compass of the indi- 
 vidual words and of the single feet. 
 
 (3) Finally, the end of every metrical series, be it at 
 the caesura or at the end of the verse, must faithfully 
 represent the rhythm of the foot with which it ends, e.g. 
 the anapaestic rhythm at the Penthemimeral Caesura of the 
 Dactylic Hexameter, and the trochaic at the end of the 
 whole verse. 
 
 Since it is the first law of ancient versification that the 
 metrical rhythm shall differ as far as possible from the 
 prose accent, it is considered less objectionable when, at 
 the end of a metrical series, the original rhythm is vio- 
 lated in such a way that the poetical rhythm does not 
 harmonize with the prose accent, than when the opposite 
 is true. This can be observed in the rhythmical laws of 
 the Hexameter. 
 
 At the caesural pause, the end of the metrical series, the 
 rhythm of which must be kept pure, consists of one foot; 
 at the close of longer verses of one and a half or two 
 feet, and at the close of shorter verses of one. 
 
 It is, however, clear that this rule is principally applica- 
 ble to poetry written Karh. cttlxov, and to such systems as 
 are not united by Synapheia, as e.g. the Dactylic Distich 
 and the Iambic Epode. Where Synapheia can or must 
 occur, and the single verses forni properly only sections of 
 a single metrically harmonious whole, violations of the rule 
 in (juestion are more frequent and more easily pardoned. 
 
 Above all, the occurrence of a monosyllable at the end 
 of a metrical series is to be avoided, except where another
 
 Structure of Hexameter and Petitaiiietcr. 79 
 
 monosyllable precedes, and this rule is so much the more 
 to be observed, the longer the preceding word is and the 
 greater the number of morcc. that it has. Therefore the 
 least unpleasant effect is produced where a pyrrhic word 
 precedes the monosyllable. Hence the verses from Vergil 
 are bad : — 
 
 et cum frigida mors anima secluxerit artus. 
 
 dat latus. insec|aitur cumulo praeruptus aquae mons. 
 
 The reason for this is that a monosyllable, on account 
 of its small compass, has not the force necessary to draw 
 towards itself a polysyllable, and the consequence is that 
 in verses like those just quoted the end of the metrical 
 series is separated. A 
 
 25. Rhythmical Structure of the Hexameter and 
 
 Pentameter. 
 
 We wish to explain more at length the above-mentioned 
 laws in the case of some of the most common metres, 
 since these are at the same time the most carefully formed, 
 and to do this by illustration from the Latin poets — for 
 they developed the single metres especially as regards 
 rhythm more carefully than the Greeks, — remembering that 
 nearly the same rhythmical laws are valid also for Greek 
 poetry. \ 
 
 We consider first the Dactylic Hexameter; 
 
 First of all, it is plain that no wor4 ending with a dactyl 
 can stand in the second and third foot of the Hexameter, 
 — still less- a word ending in a spondee, since the spondee 
 is not the original measure of this verse. 
 
 So we find only twice in Horace (Epist. I. 18, 52 ; H. 
 3, 41) a word ending with a dactyl in the third foot, and
 
 8o RhytJiniical Structure of the Verse. 
 
 not at all in the other poets considered here. Furt'^'^'- 
 Vergil and Propertius (not Ovid and Tibullus) h?- 
 seldom in the second foot a word ending with a dactyl, 
 and never one ending with a spondee. In Horace in the 
 Episdes and Satires this occurrence is somewhat more 
 frequent. Hence the following lines cannot be recom- 
 mended in this respect : — 
 
 et cum frigiiia mors anima seduxerit artus. 
 per coniibia nostra, per inceyijs hymenaeos. 
 
 Still worse is this from Ennius : — 
 
 ore Cf/hcgiis Mnn^r' Tuditano coUega. 
 
 Less frequently vlo "e find before the Penther "meral 
 Cresura a pyrrhic word followed by a monosyllable, as : — 
 
 ille autem : neqiie tc \ Phcebi cortina fcfellit. 
 
 it- to find two or three monosyllables : — 
 
 /jff) quoties et quae \ nobis Galatea locutast. 
 1' simplicior qitis ct est? \ qualem me saepe libenter. 
 
 It is usual at -t>^ ^^ ^rithemimeral Caesura to find a word 
 ending with an r ^.^z., spondee, or iambus. 
 
 What has just been said holds also for the Hephthe- 
 mimeral Caesura.' On the other hand, a dactylic word not 
 seldom forms ue arsis of the Trithemimeral Caesura, since 
 this lies in the' first foot. 
 
 At the end of the sixth foot a monosyllable is allowed 
 only when the the.:'*; is also a monosyllable : — ,/. 
 
 at Boreae de part "^ cum fulminat et cum. 
 
 The thesis of the fifth fdJl"^"*' 'onosyi'able if a 
 
 pyrrhic word or two monosyllables tollow ■. ,f 
 
 nam ncque Parnasi vobis juga, nam iieqiie Pindi. 
 cederet aut quarta socialiter. hie et in Acci.
 
 Structure of Hexameter and Pentameter. 8i 
 
 But a monosyllable is disagreeable in the fifth thesis if a 
 ^ more than two syllables follows. In the same way 
 me titiii rhesis cannot be the end of a polysyllable, for in 
 such a case this (thesis) becomes anapaestic through its close 
 connection with the arsis of the fourth foot, while it is 
 separated from the fifth arsis which belongs to it, by the 
 end of the word. Consequently verses like the following 
 seldom occur : — 
 
 fixerit aeripedem cervam licet aut Erymanthi. 
 
 tres Antenoridas Cererique sacrum Polyphoeten. 
 
 Finally, it is not usual to make the fifth and sixth feet 
 
 out of one word of five '^yllaVles, ... 'n Horace : — 
 
 .1 . 
 di visit medium fortissima Tyndaridarum. 
 
 The reason for this is that, in Latin, the two last sylla- 
 bles of such long words are usually inflectional rv derivative 
 endings, and the beginning a prepositic'- cnt^ ?- 
 
 fore have more sound than meaning, they see feebi t 
 the end of the verse which should close forcibly, i. is foi 
 this cause and not for metrical .-^ '^n'' that they are 
 avoided. In Greek, on the other hi , . e-syllabled words 
 are usually compounds of noun and verb, and not so empiy 
 of meaning as most Latin words of corresponding length. 
 Hence there is far less reason for avoiding vhem. 
 
 It is usual to find in the fifth foot a word with dactylic 
 
 ending, and in the sixth a dissyllable ; or ii; the first case a 
 
 'ord with trochaic ending, and in the ser'nd a trisyllable : — 
 
 in nova fert animus ' tS dicere formas. 
 
 insit"" «)t ad/>t' labor es. 
 
 But the secoi. arsis of the fifth foot can be a mono- 
 syllable : — 
 
 arma virumque cano, Troiae c|ui primus ab oris.
 
 82 RhytJimical Structure of the Verse. 
 
 A monosyllable in the sixth foot preceded by a polysyl- 
 lable occurs more than forty times in Vergil ; in Ovid only 
 eleven times. Vergil violates the rhythmical laws of the 
 fifth foot about one hundred and twenty times; Ovid only 
 eighty times. Many of the later writers are still more 
 strict. Horace, on the other hand, is less strict in the 
 Hexameters of the Satires and Epistles. The exceptions 
 found in Vergil and Ovid are almost always occasioned by 
 one of the reasons noted in Section 4. 
 
 Moreover, Vergil makes admirable use of discordant 
 verse-endings in order to express by the rhythm itself what 
 is dreadful, weird, or unexpected : — 
 
 vertitur interea caelum et ruit Oceano nox. 
 et nunc ille Paris cum semiviro comitatu. 
 
 With comic effect in the Georgics, I. 181: — 
 
 turn variae illudant pestes: saepe exiguus mus. 
 
 To excuse the license of a spondee in the fifth foot, the 
 poets used most frequently a word of four syllables, or 
 sometimes a trisyllable ; but in this latter case another 
 word of at least three syllables ordinarily preceded this 
 
 trisyllable : — 
 
 armatumque auro circumspicit Oriona. 
 
 perqiie hiemes aestusque et inaequales autumnos. 
 
 Vergil has twice (^n. HI. 12; VHI. 679) an ending 
 borrowed from Enpius of this sort : — 
 
 et magois dis. 
 
 It is never the case in Latin that a spondaic verse has 
 a dissyllable in the fifth or sixth place. 
 
 All these laws, with the exception of the use of five- 
 syllabled words at the end of the verse, are equally valid
 
 Structure of the Remaining Metres. ^t, 
 
 for Greek poetry, although here exceptions are more 
 numerous, especially as regards the verse- ending, and also 
 the second foot where, on account of the frequency of the 
 Top.-q Kara rpirov Tpo)(ixlov, dactylic words occur not rarely, 
 and spondaic words much more rarely, as in the first verse 
 of the Odyssey : — 
 
 avopa fxoL Ivvcttc, Movaa, TroXvTpoirov, os paXu. TroAXa. 
 
 A spondaic verse with a dissyllable in the fifth or sixth 
 foot never occurs in Homer. 
 
 The rhythmical laws of the Pentameter are still more 
 strict than those of the Hexameter. 
 
 In the Latin writers considered here, a dactylic or spon- 
 daic word never occurs in the second foot, and only once 
 (Ovid, Pont. I. 6, 26) at the end of the verse do we find 
 a monosyllabic enclitic word with preceding pyrrhic. Here, 
 too, the Greeks are less strict than the Romans, but much 
 stricter than in the Hexameter. 
 
 The Greeks and many Romans are wont to close the 
 Pentameter with any polysyllable, but the most careful 
 Romans (Tibullus, Propertius in Bks. IV. and V., and Ovid 
 in most of his post-exilian poetry and always in his pre- 
 exiliafi) close this verse with a dissyllable preceded by 
 a word with trochaic ending — evidently in order in this 
 way to make its termination like that of a Hexameter 
 ending with a trisyllable. /' 
 
 26. Rhythmical Structure of the Remaining Metres. 
 
 The rhythmical laws of the Iambic Trimeter are precisely 
 the same as those of the Hexameter. Consequently those 
 verses cannot be approved which have at the Penthemime-
 
 84 RJiytJiniical Striictjire of tJic Verse. 
 
 ral Caesura a monosyllable with a preceding polysyllable, as 
 in Horace : — 
 
 diris again vos, dira detestatio. 
 
 It is much worse when the third iambus closes with a 
 word which has an iambic or spondaic (anapaestic) ending, 
 as in the following examples : — 
 
 regnante te vides ut imperium cadat. 
 
 sed simplici carmen per omne evectus est. 
 
 For in these cases the verse is divided into two equal 
 halves, and the Hephthemimeral Caesura loses almost all its 
 force. Such verses never occur in Horace and Phsedrus, 
 very seldom in the Greek tragedians, but more often in 
 the Greek comedians. 
 
 Phsedrus, in order not to dim the original iambic char- 
 acter of his verse, never places a word which itself has a 
 spondaic or anapaestic ending at the end of the second, 
 third, or fourth foot. 
 
 A monosyllable with preceding polysyllable at the end of 
 a verse is bad, unless this monosyllable is enclitic, as in 
 Sophocles : — 
 
 ovh av 8i/<at'ws es kukov TreaoLjxi tl, 
 or in Phsedrus : — 
 
 timore mortis ille turn confessus est. 
 
 Horace avoids even this, but the Greek tragedians occa- 
 sionally admit monosyllables at the end of the verse when 
 they are not at all enclitic. .^^ 
 
 As regards systems and strophes in which Synapheia is 
 allowed, — that is, in the ^olic, Doric, and dramatic, — it is 
 clear that the rhythmical laws for ccesura and verse-endings 
 must be less strictly observed at the time when the appre- 
 ciation of the original metrical unity of the individual parts
 
 Structure of the Remaining Metres. 85 
 
 of system and strophe is still lively. Hence even Horace 
 has sometimes a monosyllable with preceding polysyllable 
 at the end of a logaoedic verse, as : — 
 
 alme Sol, curru nitido diem qui. 
 audivcre, Lyce, di mea vola, di. 
 
 In the Alcaic hendecasyllabic an iambic word rarely 
 occurs before the caesura, as HI. i, 9 : — 
 
 est ut viro rir latius ordinet. 
 
 In too close imitation of the Alcaic hendecasyllabic, 
 Horace avoids similar words in the same places in the 
 Alcaic enneasyllabic, except at I. 26, 11 in a proper name. 
 In Horace, too, a spondaic word-ending never occurs as 
 the second foot of the Sapphic hendecasyllabic, but, on 
 the other hand, he has sometimes a dactylic word before the 
 cgesural pauses of the Asclepiadean verses, as I. 6, 17: — 
 
 nos convivia, nos | proelia virginum. 
 
 In general, wherever the strophes were only mechanically 
 formed, or the verses written Kara (ttCxov, the rules were 
 much more strictly observed, as is shown in the tragedies of 
 Seneca.
 
 SIXTH SECTION. 
 
 -oo'i^ixy- 
 
 ENCLISIS AND TMESIS. 
 
 27. Enclisis. 
 
 In order to correctly employ the rules already given, it 
 is necessary to observe the rules of Enclisis and Tmesis. 
 By Enclisis the enclitic word loses its own individuality and 
 is merged into the preceding, so that no monosyllable really 
 concludes the verse, as in the following example : — 
 
 aXX oio ws erdpovi ippvaaro, te/ACi/os irep. 
 
 pectora quorum inter fluctus arrecta ]nba.egt{e. f 
 
 It is natural that monosyllables which always stand in 
 the second place, as /xev, 8e, ydp, should be quite closely 
 connected with the preceding word. 
 
 The cases of Enclisis in Greek are well known. In 
 Latin, besides (^i/t\ ve, ne, the monosyllabic forms of the 
 indefinite pronoun qiiis, in combination with si, ne, nmn, 
 cum, and in the formula ucscioqiiis, are used enclitically ; 
 also the pronouns me, te, se, nos, vos in combination with 
 prepositions ending in tra and kr, as e.g. inter, intra. 
 Finally the monosyllabic forms of the Indicative and Sub- 
 junctive of sum were sometimes used as enclitics, when a 
 pyrrhic, tribrachic, or trochaic word preceded, but only on 
 condition that the syllable before the auxiliary remained short. 
 
 It is to be noticed that titm cum usually becomes one 
 word, as satisesse, quopacto, and often priusquam, iamiam. 
 86
 
 Tmesis. 87 
 
 28. Tmesis. 
 
 I. By Tmesis, in Homer and the tragedians, the prep- 
 osition of a compound verb is frequently separated from 
 its verb, generally by the interposition of one, or two or 
 three short words, e.g. : — 
 
 The preposition stands only rarely after the verb, evi- 
 dently because it might then easily seem superfluous. 
 
 In Latin poetry, cumque is often separated from the 
 relative pronouns and adverbs to which it belongs, and 
 sometimes in prose. Thus in Horace : — 
 
 qiictii sors dierum ctiDiqtie dabit. 
 
 Mihi ciimqiic (Od. I. 32, 15) is a corrupt reading. 
 
 Note the passage in Vergil, Georg. IH. 381, septem 
 siihjecta trioiii, which was imitated by Ovid, Met. I. 64, 
 Scythiam septemqiie trione?ti. 
 
 Elsewhere in the poets of the Augustan age. Tmesis 
 occurs only in the case of jDrepositions which are also used 
 independently as adverbs, as in Horace: — 
 
 circum 
 spectemus vacuam Romanis vatibus aedem. 
 
 Vergil, however, following the example of the older poets, 
 has (Eel. VHI. 17; ^E^n. IX. 288; X. 794) prae, and the 
 negative in (always with a following qtic^ thus separated ; 
 and Ovid imitates him in Met. XII. 497. 
 
 II. Another kind of Tmesis, not expressed in the text, 
 occurs when the monosyllabic preposition or negative in of 
 a compound word is for metrical reasons attracted to the 
 preceding word.
 
 88 Eiiclisis a?id Tmesis. 
 
 This occurs on account of the caesura in the following 
 Horatian verses : — 
 
 dum flagrantia dc \ torquet ad oscula, 
 ut adsidens iii \ plumihus pullis avis, 
 parentibusque ah \ ominatus Hannibal ; 
 
 and in this from Vergil : —   
 
 magnanimi lovis in \ gratum adscendere cubile, 
 
 to release the first syllable from the second part of the 
 compound. This kind of Tmesis occurs (very rarely) in 
 Vergil in the ^neid, in Horace and Propertius, in Phcedrus 
 III. 15, 6 ; V. 7, 19, but not in Ovid and Tibullus.
 
 SEVENTH SECTION. 
 
 -oo;*;oo- 
 
 ON THE TREATMENT OF SUCCESSIVE VOWEL 
 
 SOUNDS. 
 
 29. Synizesis, Diaeresis, Crasis, Elision, Hiatus. 
 
 Introductory Remarks. — When two vowels come to- 
 gether in the middle of a word or at the end of one word 
 and beginning of another (in Latin also, when the first 
 syllable or word ends with in, or the second syllable or 
 word begins with h), there is a certain stopping of the 
 voice, and a so-called Hiatus occurs. An attempt is made 
 to avoid this in various ways, principally through a com- 
 bination of the vowels by means of Contraction, Synizesis, 
 and Elision. Among the Greeks those least sensitive to 
 this Hiatus were the lonians, as is shown by the Homeric 
 poems. The Romans were on the whole much more sensi- 
 tive in this matter than the Greeks. 
 
 Hiatus at the end of a word is the most unpleasant ; less 
 so in the middle of a compound, and still less so in a 
 simple word. Hiatus in Homer is often only apparent, as 
 the digamma frequently is to be considered as removing it ; 
 
 e.i^. feKTf^oXo'i, /reVo?, /toikos, ^oivos, OeopeLKeXo?, Aftbrj';, 
 af€K(ji)v, 'Arpe/rtST^?. In course of time the (Greeks and 
 Romans became gradually less sensitive to hiatus, and there- 
 fore cases of harsh synizesis and elision became rare. How- 
 ever, greater freedom in the allowance of hiatus at the end 
 
 89
 
 90 Treatment of Successive Vowel Sounds. 
 
 of a word is not found in the most artistic poets, but in 
 folk-poetry. 
 
 Hiatus appeared especially harsh when in the body of a 
 word the first of the two vowels was long. This was quite 
 frequent in Greek, but rare in Latin {z'oca/is ante vocalem 
 brevis). Hence as early as Homer, the first vowel in olos, 
 ijpwos (Od. VI. 303) was sometimes shortened, and perhaps 
 in Zy']io<i also. The Attic writers did the same in the case 
 of TToieo) (for which in colloquial language iroiw was com- 
 mon), TotoiiTos, etc., but generally only in the case of diph- 
 thongs whose second letter was i. Thus for eto?, eta, etov, 
 occur frequently eos, los, in Latin ens, Ins. 
 
 \x\ Latin the first long vowel in combinations is regularly 
 shortened ; e.g. dxhisco, proinde, praeaeutus. 
 
 Li Latin of the Augustan age the penultimate vowel is 
 long in the endings ai, ei, ais, eis from nominatives in a, 
 es, aius, -a, -inn, eiiis, -a, -urn, as aiilai, Gai, diet, Mais, 
 Circeis, except ;r/, spei, fidei ; the / is long in f'lo and its 
 derivatives except where followed by er (Ovid : omnia iain 
 /lent, fieri quae posse negabani), and common (always long 
 in Phaedrus) in genitives in ins except alius (contracted 
 from aliins). The first vowel in Diana and o/ie is also 
 common. 
 
 Greek words when transferred to Latin retain their orig- 
 inal quantity. For those ending in etos, etc., see above. 
 Afaeotis Ov. Trist. HL 12, 2, is a corrupt reading. 
 
 Like vowels, especially when they have the same quantity, 
 are usually contracted into one by the Attic writers. 
 
 In Latin of the Augustan age the combinations aa, ce, 
 00, are not found, except in the compound coorior (but 
 coperio) ; somewhat oftener ii, in/, although until the time 
 of Propertius, the genitive of substantives in ins, ium was
 
 Introductory Remarks. 91 
 
 always contracted, e.g. fi/i, coiisili. Hiatus was felt less 
 when an // intervened, as incoho, picliciuio, niJiil, mihi, 
 although even in these cases the contracted forms p7-endo, 
 nil, and (rare)}') ;/// occur; so always vcmcns for vchemens. 
 
 Compounds of iacio, in forms derived from the ])resent, 
 were written with one /, as ohicio, although a weak / was 
 sounded before the /. In biiugiis, qiiadriiiigiis, 7 only was 
 sounded. 
 
 Until the time of Augustus ito, vo were written in place 
 of ////, vii, not only at the beginning and in the middle of 
 a word, but also at the end, e.g. assidiios, servom. 
 
 Other combinations of a and with a vowel did not 
 occur in Latin, except in compound and proper names, as 
 coalesco, Gai, Grais, etc., while in Greek they were not 
 uncommon. Of more frequent occurrence were ///, ///, 
 ut, vt. 
 
 Ea, ei, eo, en, ia, ie, io, ua, ue, no are admissible, no 
 matter whether the initial / and // are vowels or conso- 
 nants. In Greek, on the contrary, simple e before a follow- 
 ing vowel produced an unpleasant effect, as is shown by the 
 frequent synizesis. 
 
 To avoid hiatus in the middle of a word, the Attic 
 writers employ contraction very often. Homer much less 
 frequently. 
 
 This contraction, however, is not rare in Latin, especially 
 in many forms of perfects in -vi ; usually in declension and 
 conjugation ; e.g. in the dative of the fourth declension u 
 for ///, in the genitive and dative of the fifth e for c/, but 
 also elsewhere : thus always desse dcrrarc, usually di, dis, 
 idem, isdem for dei, deis, eidein, eisdein.
 
 92 Treatment of Successive Vowel Sounds, 
 
 30. Synizesis in Greek. 
 
 Synizesis hardly occurs in Greek except when the first 
 vowel is e, since this vowel, being the weakest, most easily 
 combines with the following. 
 
 Synizesis is most acceptable in the arsis, e.g. : — 
 
 ^/3Dcrea) ova. crKijirrpo}. 
 )(pv(r£ov (TKrjTTTpov e)(ovTa. 
 
 But it often appears in the thesis : — 
 
 firjvtv aeiSe, Otd, H-qXrj'idSew 'A^LXrjo<;. 
 
 Cases where the synizesis unites two short vowels to 
 form the thesis, are also allowable, — as in Be'Aea, IlT^Aeos. 
 
 Homer and Archilochus employ synizesis especially in 
 cases where contraction occurs in the later writers. The 
 tragedians went still farther, following Pindar, not only in 
 proper names, but also in other words ; thus synizesis 
 occurs in IlrjXea, ©t/o-ccos, 'Pc'd, Kpewv, frequently in ©eo's and 
 
 its cases, ttoAcws, ao-Tew?, iKTTvicjv, ta, etc. 
 
 Synizesis appears especially in the thesis of the first and 
 in the third foot of the Trimeter. As the Greeks did not 
 know the consonantal t, synizesis with iota seldom occurs. 
 Thus in Homer we find AiyuTTTcus, 'la-Tiaiav, also 'EwaAiw 
 (but compare 34, beginning) and ttotvo. for irorvia, written 
 out. This is rare and doubtful in the tragedians, particularly 
 outside of the lyrical passages. 
 
 The cases of synizesis in 'Epivuo)v (Eur. Iphig. Taur. 931, 
 970, 1456), Suoiv (Soph. Oed. Rex, 640) rest on corrupt 
 readings, and also in 'HX^KTpvuiv-q (Hes. Scut Here. t6).
 
 Synizcsis in Latin. 93 
 
 31. Synizesis in Latin. 
 
 I. .-Xs regards the Latin poets now under consideration, 
 tlie first thing to be noticed is the change of / to j and 
 (more rarely) // to v. This usually occurs in words which 
 otherwise would not fit the metre. It never occurs in iambic 
 and trochaic verse, and very seldom in logaoedic (Hor. Odes 
 III. 4. 41 ; 6. 6, consiljum, principjum). It does not occur 
 when two like consonants precede. Hence in Verg. .^n. 
 VI. 653, we write curniin. It is not allowed in the middle 
 of compounds, and we must always read and write semesus, 
 semanimus, setnhomo, not semiesiis, etc. To avoid this synizesis 
 the poets often shortened the gen. plural ''ii/in'' to "/////," e.g. 
 moderanttim, sapicntinii ; sometimes they used in the fourth 
 conjugation -ibain for -icbam, e.g. loiibant. The Greek words 
 Aiax, Gniius, Main, Troia, had the consonantal / from earliest 
 times. In other Greek words / and 7/ did not become con- 
 sonants, and hence we read : loat, 'iambus, Agaiw, eiiaiigcliiim, 
 and always hiliis ; for Hor. IV. 2, 2, is a corrupt reading. 
 
 This synizesis, which had already appeared in Ennius, 
 occurs twenty-two times in Vergil, generally in words which 
 would not otherwise fit into Hexameter verse, as in deriva- 
 tives of aries, and paries, though sometimes in others, as 
 flitvjoniin omnja. U is changed to v only in tenvis, tcnvia, 
 genva. In Mx\. I. 2, Lavinaque is to be read. Horace 
 has in his hexameters Nasidjeni, vindevijator, vjctis, Serviljo, 
 and pitvita twice. For his logaoedic verse, see above. 
 
 Synizesis" occurs only twice in Ovid: Metam. VII. 151; 
 XV. 7i<S, arjetis ; Antjiim. XV. 709 does not belong in this 
 category, as the antepenultimate in promunturiiim is short. 
 In Propertius there are three cases of synizesis (in abjcgnac, 
 abjegni), but none in Tibullus.
 
 94 Treatment of Siieeessive Wnvel Sounds. 
 
 Fortuitiis and coiiubiuiii never suffered synizesis, as / in 
 the first word and a in the second were common. Etlain 
 and quo Ilia III have always vocahc /. 
 
 II. The second kind of Synizesis consists in the forma- 
 tion of a diphthong from two vowels. In the Augustan 
 poets this occurs always (except in proper names) in dis- 
 syllabic or compound words. The diphthongs thus produced 
 do not occur separately elsewhere in Latin. In prose of the 
 best period dein, delude, deiiieeps, neuter, a/i, /luic, have 
 always the diphthong ; on the other hand, the Romans pro- 
 nounced uutiquani and autire. 
 
 Dehiuc, however, is generally dissyllabic, monosyllabic 
 rarely in Vergil, and once only in Ovid and Propertius. 
 In Hor. Sat. II. 3, 91 quoad is a monosyllable, and 
 perhaps prout in II. 6, 67. Reieere is trisyllabic in Verg. 
 Buc. 3, 96 (as euius is monosyllabic in the Elegia ad Mer- 
 sallam, 35), and dcicere in Hor. Sat. I. 6, 39. Finally, the 
 vocatives Poinpci (Hor. Od. II. 7, 5) and Vultei (Ep. I. 7, 
 31) are dissyllabic, and ei is probably to be pronounced 
 as »?. 
 
 Compounds of eiiruin never suffer synizesis, but lose the 
 iu, e.g. eireuco, eireuago, or are divided by Tmesis, e.g. 
 eiiruui errant, ALi\. II. 599. 
 
 In (ireek words the diphthong m remains, as in Harpyia ; 
 also £11, e.g. Orpheus. Only in the Culex 117, 269, and in 
 Phtedrus V. i, i do we find Orpheus, Phalareus. Ei in 
 the genitive and (Eclog. 4, 57 ; ^'n. V. 184) in the dative 
 is diphthongal, and is so pronounced even in prose. 
 
 This synizesis (proper names excepted) is much more 
 frequent in Plautus, who changes thereby words of three 
 and more syllables when they are derived from dissyllables, 
 as suoruiii, piiella, eamus, duellatores.
 
 Dicercsis. 95 
 
 III. The third kmd of Synizesis, borrowed from the 
 Greeks, occurs in the two last syllables of Greek proper 
 names in eu's, the substantives alveus and balteus, and 
 adjectives in eus, ea, eu/n, denoting material. It appears 
 first in CatuUus, and is employed only in dactylic metres, 
 especially in the first and sixth foot of the Hexameter, and 
 almost always in such a way that the last syllable is long. 
 Vergil is fond of employing this form of synizesis as well 
 as the others to express what is hard or dreadful, as ^-En. 
 VI. 280: — 
 
 fer;v/que Eumenidum thalami et Discordia demens. 
 
 Vergil has it twenty-one times, in proper names and 
 elsewhere ; Ovid has it fourteen times ; Horace, Sat. I. 8, 
 43; II. 2, 21, cerca, ostrea ; Propertius, Euipeo, Nerco, 
 Prometheo ; TibuUus, alveo. 
 
 The genitive of words in ei's has been discussed above. 
 In the Georgics, IV. 34, alvea?'ia occurs with four syllables, 
 where some write alvaria. Dissyllabic Penei, Georg. IV. 
 355) points to the nom. riTyved?, as 'AA<^eos beside 'AA</)«os. 
 Wherever adjectives have a shorter form this is employed 
 to avoid synizesis, as aheniis, chitrnus, ilignus. 
 
 32. Diceresis. 
 
 Diaeresis is the division of a diphthong into two syllables. 
 In most of the cases in Homer where diaeresis was 
 formerly assumed it is now plain that the older form 
 of the word appears. Thus patronymics in -€l8i]<;, like 
 Tlr]XeL8r]^, in which in Homer and Hesiod e was always 
 separated from t, come from -e^tSr/? ; so I7 from ear], eu from 
 i(TV, AvKoopyos from AvK6popyo<;, etSuia from fefiBxKa. In 
 Latin we find siluac with change of the //, in Hor. Od. I.
 
 96 Treatment of Successive Vowel Sounds. 
 
 23, 4 ; Ep. 13, 2 ; in the Satires cind in Phgedrus suetus, 
 suesco ; in Ovid and Tibullus soluo and voliio, but only in 
 compounds ; in Propertius Veins. 
 
 33. Elision. 
 
 General Remarks. — The meeting of two vowels at the 
 end and beginning of two words appeared to the classic 
 poets much less admissible than hiatus in the middle of the 
 word. Latin words ending in ;;/ were considered as ending 
 in a vowel, and those beginning with h were regarded as 
 vocalic also. 
 
 Hiatus was considered especially harsh in the case of 
 short final syllables or those ending in vi, as these were 
 spoken more rapidly than the long. Among the short finals, 
 e and o in Greek, and e and / in Latin, had a particularly 
 weak sound. 
 
 The meeting of two like vowels was especially dis- 
 agreeable. 
 
 To avoid hiatus, Elision of the final vowel was employed. 
 This is a false name, for the final vowel except in que, ve, 
 lie, and the monosyllabic particles in €, as re, ye, 8e, was 
 never coinpletely dropped, but when a long syllable followed 
 or was demanded by the metre, the final was combined 
 with the succeeding vowel into a kind of diphthong. If 
 the following vowel was short, the final was so weakened by 
 rapid utterance, that it was not considered from a metrical 
 point of view. This last kind of elision was regarded as 
 the harsher. Li Greek, the later writers usually drop those 
 final vowels which are not counted metrically, and mark 
 their omission by the apostrophe, e.g. ovXo/xcvtjv, rj yivpC 
 'AxMoU. This is faulty in so far as it makes no distinction 
 JDetween the two different kinds of elision just described.
 
 Elision in Greek. g/ 
 
 In the discussion of Elision and Hiatus, Greek and 
 Latin usages must be sharply distinguished. 
 
 34. Elision in Greek. Crasis. Aphaeresis. 
 
 In Greek the combination of a long vowel or one that is 
 not otherwise elided, with a following long vowel, is of 
 comparatively rare occurrence, but when it does occur it 
 
 is usually when the words Kat', 8r/, /xt/, tS, rj, rj, cyw, ixoi, a-oi, 
 
 TOL, irpo, o, a, come first, or apa, av, ov, follow, as e.g. in 
 Homer at the beginning of the verse : — 
 
 ■» > 3.1. 
 
 7} ovK atets- 
 In Homer, also, the following cases are to be noted : — 
 
 Od. I. 226, dXaivivq rjk ydfxo^ ; 
 
 II. XVIII. 458, 'I'tEt e/Aw wKVfxopo) ; 
 
 perhaps, too, II. 651, 'EvuaAtw avSpeic^ovrr^ ; 
 and I. 277, fJiT^T€ crv, HrjXelSrj, eOeX'. 
 
 This union of two vowels was expressed, frequently by 
 the Attic writers, rarely by Homer, in the writing itself by 
 means of Crasis, principally when the article or one of the 
 short words w, o, a, Kat, -n-po, p.01, (Tot, TOL, came first, as e.g. 
 ovv€Ka, TavSpi, TuAAa, etc. It can also be counted a case 
 of crasis when c or (seldom) a at the beginning of a word 
 is elided (aphseresis) if a long vowel precedes, e.g. e^eAai 
 'k t^s oiKtas. This happens principally with the augment, 
 single prepositions, adverbs, conjunctions, or other very 
 common words, like eyw, ia-riv, earoi, lo-rat, as well as dvd 
 and aTTo, especially in the Attic poets, who employ crasis 
 and aphffiresis usually in dialogue. 
 
 a, e, o, are cut out of the verse without hesitation, 
 although the optative ending ete, the imperative 18/, and
 
 98 Trcatuioit of Successive ]'\nvel Sounds. 
 
 the genitive 010, are not thus ehded. Further, i was ehded 
 
 in the verb-endings o-t, fxi, 6l, tl ; in Homer the t of the 
 
 dative plural, and rarely that of the singular, as (II. V. 5 ; 
 
 X. 277):- 
 
 dcrrep oirMpLvw ei/aXtyKtos, 
 
 p^at/3€ ok Tw opviO' 'OSvcrews. 
 
 In the same way Ol in o6t, aXXo$i, avToOi, TrjXoOi. It is 
 doubtful whether 6Vt is elided in Homer, for in such cases 
 Bekker writes ore. Short v is never elided. 
 
 Again ai (which very early appears to have suffered a 
 weakening of the diphthongal sound in many forms, and 
 sounded to the Greeks of the latest period always as short 
 e) was elided in the forms of the Passive and Infinitive, as 
 fSoyXo/xai, Sovnu. Once ai in the nominative plural is elided, 
 II. XI. 272, o^et' oSwat. The elision of ot in fxoL, aoi, toL as 
 II. I. 170, ovhi 0-' otco ; VI. 165, o? /u, W^Xtv, is very rare and 
 questionable. 
 
 35. Elision in Latin. 
 
 Compare first the general remarks on Elision, 33. The 
 means which the Romans employed to avoid hiatus were 
 exactly the same as those of the Greeks. The harsher 
 character of the Latin language was shown by this fact, 
 that in general no account was taken in the metre of a 
 final vowel occurring in the hiatus, whether long or short. 
 Compare for Elision in Latin, Cic. Orator, 44, 150; 45, 152; 
 Quint. IX. 4, 33. 
 
 As would be expected, the harshest form of elision was 
 that of long vowels, particularly of the diphthong ac ; less 
 harsh was the elision of syllables ending in ;//, which were 
 pronounced short, but in which the closing consonant still 
 kept some sound ; easiest of all was the elision of short
 
 Elision in Latin. 99 
 
 syllables. Elision was harsher in the cases where the second 
 vowel was short than where it was long, as has been already 
 remarked. 
 
 In the ^■Eneid I. 1-80 there are twenty-one cases of the 
 elision of short vowels (among them fourteen cases of que, 
 ne), twelve cases of the elision of syllables ending in iii, 
 and eight of long vowels, one of the diphthong ae. 
 
 The elision of a vowel preceded by a vowel without a 
 consonant {7'ocalis piira) seldom occurs if the first vowel 
 is long, as ^-En. X. 179, Alpiieae ab origine. 
 
 The best poets seldom elide Greek endings, as in Ino, 
 Penelope (Ovid never). 
 
 As regards the compass of the word, elision seems most 
 harsh in monosyllables which are long or end in m, and 
 these were never elided if they belonged to a regular 
 declension or conjugation, except qi/i (nom. sing.), ;ne, te, 
 dc, til, mi (^milii), sum. 
 
 Further, since by the laws of Latin accent, the final 
 vowel was less emphatically pronounced in words with 
 spondaic, trochaic, or tribrachic endings, than in those with 
 cretic, dactylic, iambic, and pyrrhic endings, the former 
 class of words suffer elision more frequently. 
 
 Again, elision is considered harsher when the second 
 syllable has the acute or circumflex accent, than when it 
 has the grave or none at all. 
 
 Monosyllabic pronouns (except the interrogative quis, 
 qui), prepositions, conjunctions, atque when the e is elided, 
 and the most common monosyllabic adverbs, like itii, ut, 
 haud, have the grave accent. 
 
 From what has been said, it is not surprising that the 
 best poets never elide iambic words before acute, circum- 
 flex, or short syllables (except twice in Phaedrus, III. 7, 15 ;
 
 lOO Treatment of Sueeessive Voivel Sounds. 
 
 V. 9, 4, veni ergo; tace inqitii), and cretic words almost 
 never. Of the poets considered here, Horace alone in the 
 Satires allows the elision of cretic words before followinsf 
 short syllables. 
 
 Dactylic or pyrrhic words ending in w, a, o, are treated 
 in the same way, though somewhat less strictly. 
 
 In the oldest Roman poets, Plautus, Terence, and some- 
 times in Lucilius, there are many, and often harsh cases of 
 elision, remarkably few in the Annals of Ennius ; in Vergil 
 a considerable number, but rarely of a harsh character. In 
 the Satires Horace has more and harsher cases of elision 
 than in the Epistles ; still fewer in the Odes and Epodes. 
 In Propertius the cases of elision are more numerous and 
 harsher than in Tibullus. The most polished of the poets 
 in this respect was Ovid, and the later writers for the most 
 part followed his example. To illustrate this, in the first 
 book of the Metamorphoses, which as epic poetry contains 
 an especially large number of cases of elision, long sylla- 
 bles are elided only 8 times ; syllables ending in ;;/, 23 times ; 
 short syllables, 129 times. On the other hand, the corre- 
 sponding numbers in the first book of the JEntid are 85, 
 
 97> 173- 
 
 In general, elision is most frequent in the epic hexam- 
 eter, and less frequent in the didactic, bucolic, and elegiac 
 pentameter, and in the logaoedic and iambic verse of 
 Horace. 
 
 Vergil frequently employs numerous or harsh elisions in 
 the painting of difficult or fearful situations (^n. III. 658; 
 IX. 427):- 
 
 ■monsXriwi horrend^w inform*? ingens. 
 
 non vie ! adsum qui fee/; in me convertite ferrum.
 
 D iff c voices ill Elision. lOi 
 
 Those cases do not properly belong to elision where est 
 (sometimes also cs^ stands after a vowel or ;;/. In this 
 case est (^es) loses its vowel, and we read tnagmnnst, illast, 
 ille's. 
 
 36. Differences in Elision in Greek and Latin Verse. 
 
 For Greeks and Romans alike this rule holds good, that 
 elision very seldom occurs before the first or after the last, 
 or the next to the last syllable of the verse. 
 
 EHsion after the last syllable occurs properly only in 
 verses which are united by Synapheia, as in anapaestic 
 systems. 
 
 In such verses in Greek elision is tolerably frequent, and 
 even a few times in the Odes of Horace (in Horace the 
 preceding syllable is always long, and the elision never 
 occurs at the end of the strophe). 
 
 The Alexandrian poets, and imitating them the Romans, 
 falsely explaining some Homeric verses, occasionally (very 
 seldom except in Vergil) allow ehsion at the end of a 
 hexameter. The Romans do this only when another hexam- 
 eter follows. Such exceptional verses are wrongly called 
 hypermetrical {versus hypermetri). The preceding syllable 
 is always long. 
 
 With the exception of two verses in Vergil ending in m, 
 e is the only vowel thus elided, especially in que. Vergil 
 sometimes employs such a verse-ending in rhythmical paint- 
 ing, as /En. VI. 602, 3 (cf also IV. 629, 30) : — 
 
 quos super ntra silcx iam iam lapsura cadenti./?/^ 
 imminet adsimilis. 
 
 In Sophocles, but not in ^schylus and Euripides, elision 
 sometimes occurs at the end of the trimeter, generally if
 
 102 Treatment of Successive Vowel Sounds. 
 
 Bi or re stands at the end ; always with a preceding long 
 syllable, e.g. Oed. Rex 29 ff. : — 
 
 v(j> ov KtvovTai owfjia KaS/xeiov, jxiXa'i S' 
 "AiSt/s o-revay/Aots kol ydois TrXouTt'^erat. 
 
 Elision before the last syllable of the verse occurs in 
 Horace's Satires and Epistles, as well as in the Odes the 
 verses of which can be united by Synapheia, but not at the 
 end of the strophe. 
 
 Of the other poets considered here Vergil alone twice 
 elides at(ji/e after the sixth thesis (^n. IX. 57; 440). 
 
 The Greeks not infrequently placed monosyllables with a 
 following ov at the beginning of the verse, as Od. I. 29S : — 
 
 T] ovK diets ; 
 
 Elision occurs at the beginning of a hexameter several 
 times in the Satires of Horace, once (3, 48) in the Eclogues 
 of Vergil, but never in the other poets. Of course no account 
 is made here of those cases where a so-called hypermetric 
 verse precedes. 
 
 On the other hand, elision is never prevented by a 
 following or preceding caesura or by punctuation or even 
 by a change of the persons speaking, e.g. {JEn. I. 96 ; II. 
 I. 2 ; Hippol. 612 ; ^En. III. 98 ; Soph. El. 1502) : — 
 
 contigit oppete;-(? ! o Danaum fortissime gentis ! 
 ovXofXiVTjv, ■^ /xvpL 'AT^atots aXye' eOrjKev. 
 7} yX(h(T(T o/xfo/xoK : aXX o vov<i avw/Aaros. 
 et nati nator/tw et qui nascentur alj illis. 
 (3dXX' alei 8e in^pai veKvo)v kcuovto Oa/xtiaL 
 Orest. dXX' €p4>'- /Egisth. v(f>r)yov. Or. a-ol /SaStoreov irdpo^.
 
 Hiatus. 103 
 
 The better poets, however, Hke Tibullus and Ovid, avoid 
 eliding, in general, a long syllable in the third arsis or 
 where there is a decided stop. They also elide very rarely 
 a long syllable at the caesura. 
 
 The greatest freedom and frequency of elision in the 
 Latin Hexameter is noticed in the arsis of the first, 
 the thesis of the second, and also in the whole of the 
 fourth foot, except when the Hephthemimeral Caesura 
 occurs. In the other places elision is far less frequent, 
 especially in the arsis of the second and the thesis of the 
 sixth. 
 
 In the Pentameter the elision of syllables which are either 
 long or end in w, is in the best poets mostly restricted to 
 the first arsis or the second thesis. Elision at the csesural 
 pause does not occur in Propertius, Tibullus, and Ovid ; 
 never after the caesura in Ovid, and almost never in Pro- 
 pertius and Tibullus. In the second half of the Pentameter 
 Ovid never elides long syllables or those ending in m. 
 Horace allows elision not infrequently before and after the 
 caesura of the lyric measures, except in the Sapphic Hendeca- 
 syllabic, where it almost never occurs. 
 
 37. Hiatus. 
 
 General Remarks. — When a vowel at the end of a word 
 remains unchanged before a following vowel. Hiatus is 
 occasioned. The harshest case of this is when vowels 
 having the same sound come together, as Od. XI. 596, 
 Aaav avco w^ecrKe. On the Other hand, hiatus is most 
 allowable where the elision of the final syllable would be 
 harsh.
 
 104 Treatment of Successive Vowel Sounds. 
 
 38. Hiatus in Greek. 
 
 When a long final syllable occurs in the thesis of a 
 dactylic measure, hiatus with the following vowel is 
 allowed, e.g. : — 
 
 fjirjvLv aetSe 6td, HrjXrjidS c <a 'A^lXtjos. 
 
 So, too, in the arsis when the long vowel is shortened : — 
 
 avSpa jJiOL tweire, M.ovaa, TvoXvTpoivov. 
 
 Hiatus also occurs with short vowels, especially such as 
 are seldom or never elided, as with v and i in the dative 
 singular, in the genitive oio, in rt, Trept, 6', dvd = dvaa-T-qOL. 
 
 It is very rare in the thesis, and only admissible in the 
 regular Penthemimeral and Hephthemimeral Caesura, as II. 
 11. 781; V. 576; VIII. 556; XXIV. 285. 
 
 An unpleasant effect is produced if the long vowel of the 
 arsis remains long, as sometimes is the case in the first and 
 fourth foot of the hexameter, and especially with mono- 
 syllables, as e.g. : — 
 
 Kovprj iKaptoLo, TrepicfipMv TlrjveXoTreLa. 
 
 ^Xaivav T rjoe ^j^trcova ra t ai8a> a.fJi(}}LKaXvTrT€t. 
 
 rj )^L6vi \jjv^p(2 rj e^ vSaros KpvardXXi^. 
 
 Hiatus of short syllables is most harsh in the case of a 
 final £ in monosyllabic and trochaic words, and least harsh 
 in dactylic words. In these last, hiatus occurs most fre- 
 quently in the first and fourth foot (bucolic caesura). Again, 
 hiatus is sometimes allowed on account of the to/a^ Kara 
 TptTov rpo-)(alov, as e.g. : — 
 
 dXA' aKeoDua KaOrjcro, ifiiS o iTrnreLoeo p.vd(a. 
 
 Punctuation also relieves the harshness of hiatus in a 
 similar way. Of course those cases in Homer where the
 
 Hiatus in Greek. 105 
 
 following word has the digamma, or in earliest times began 
 with a consonant, have nothing to do with hiatus, as e.g.: — 
 
 aT€fji.fj.aT exwv iv X^P^' fi.KTjl3oXov AttoAAwvos. 
 
 So e.g. aAts = saf/s, eSos = se^/es, l^ = sex, cVo/xat = sequor, 
 Q^ = S////S. Cf. e.g. 0.1. XVII. 303 : SL-vryo-aro olo avaKTOs. 
 
 Cases of hiatus in Homer and Hesiod, which are not 
 comprehended under the rules already mentioned, rest 
 either on a corrupt text or are to be explained by old 
 forms of the words now unknown, but which removed the 
 supposed hiatus. 
 
 Homer was the model of the later epic poets, who 
 through misunderstanding introduced hiatus in places where 
 in Homer's time none existed because of the digamma. 
 But hiatus occurred in this epic poetry less frequently 
 than in Homer, and still less often in the bucolic and 
 didactic poets. 
 
 In the fifth century after Christ Nonnus and his imitators 
 restricted hiatus to a few cases with long final syllable. 
 
 Hiatus occurs more rarely in the elegiac hexameter, and 
 still more so in the pentameter, where it appears usually 
 in the first foot and the first dactyl after the caesura. At 
 the caesura it is doubtful. 
 
 Cases of the hiatus of long syllables with shortening in 
 the arsis of anapaestic metres are not uncommon. 
 
 In the iambic trimeter and trochaic tetrameter of tragedy 
 hiatus very seldom occurs in the thesis, and then before 
 a stop, or where the same word is repeated, as /Esch. 
 Agam. 1 2 1 6 : — 
 
 OTOTOL, AvKU "AttoWov, 01 cyw, iyui. 
 
 Such examples stand mostly by themselves, outside of 
 the verse. Hiatus is not allowed in the arsis.
 
 io6 Treatment of Successive Vozvel Sounds. 
 
 39. Hiatus in Latin. 
 
 Compare the general remarks in 37. 
 
 Hiatus occurs extremely seldom in the Roman dactylic 
 poets, especially in the arsis of the foot. In Vergil, the 
 hiatus of a short final syllable occurs only twice. Eel. II. 53 ; 
 ^n. I. 405 : — 
 
 addam cerea pru«r7. honos erit huic quoque pomo. 
 et vera incessu patuit de^?. ille ubi matrem, — 
 
 in each case before a decided stop. 
 
 Long monosyllabic words or those ending in w allow 
 hiatus in the arsis if a short syllable follows; as once in 
 the Satires of Horace : — 
 
 si me amas inquit; cocto num adest honor idem? 
 
 This kind of hiatus is very common in Plautus and 
 Terence, but generally in cases where these small words 
 form the first syllables of resolved iambic or trochaic 
 theses. The first kind of hiatus occurs sometimes in 
 Vergil, e.g. Mn. VI. 507: — 
 
 te, amice, nequivi. 
 
 Elsewhere iambic or cretic words are sometimes short- 
 ened in the arsis (as Ennius had already done), e.g. in 
 Vergil (Eel. 3, 79; M\\. III. 211) : — 
 
 et longum formose vale val^ inquit lolla. 
 insul«d" loiiio in magno. 
 
 This hiatus occurs sometimes in Vergil ; in Propertius 
 once, IV. II, 17; in Ovid three times, Am. II. 13, 21; Met. 
 
 I. 155 ; III. 501 ; the conjectural reading in Horace Ep. 
 
 II. 3, 65, din aptaque reinis, is incorrect. 
 
 Spondaic words at the hiatus occur only in Horace Ep. 5. 
 100 (probably a corrupt reading) and Vergil, Georg. I. 437 : —
 
 Hiatus in Latin. 107 
 
 et Esquilin^fi? alites. 
 
 Glauco et Panopea^ et Inoo Melicertae. 
 
 In this verse in imitation of Euphorion, the spondaic word 
 of the first foot is preserved unshortened with the hiatus. 
 
 In the poets considered here hiatus in the arsis, except 
 in Hor. Ep. 5, 100, and hiatus in the thesis, except in the 
 dactyhc tetrameter ossibus et capiti inhumato, Hor. Odes 
 I. 28, occur only in hexameter verse. The final syllable is 
 always long, except in three cases, Tibullus I. 5, 2)Z \ Prop. 
 III. 15, i; T^2, 45, where the syllable ends in in; and this 
 syllable is always the final of a polysyllabic word except in 
 Verg. xEn. IV. 235 : — 
 
 quid struit aut qua spe inimica in gente moratur. 
 
 Most poets allow hiatus only in the regular cresuras of 
 the hexameter or before Greek words, e.g. : — 
 
 Nereidum matr/ et Neptu«o Aegaeo. 
 
 Vergil, however, in imitation of Ennius, sometimes allows 
 hiatus at other points in the verse, at the end of words 
 having an anapaestic ending or before a stop: — 
 
 evolat infelix et feminet; ululatu. 
 
 si perec, hominum manibus periisse juvabit. 
 
 He allows this hiatus in the thesis perhaps forty times ; 
 Ovid has it only twenty-six times; Horace once (Ep. 13, 3) 
 besides the example quoted above. 
 
 Finally, it is to be noticed that the interjections and a, 
 both in thesis and arsis, before long and short vowels, can 
 be kept long by the poets, e.g. : — 
 
 et de Latia, o ct de gente Sabina ; o ego laeviis ; 
 
 but not heu, which never occurs in a hiatus {eheu is always 
 to be read instead of /leu heit).
 
 EIGHTH SECTION. 
 
 -0-0>»iOO- 
 
 LENGTHENING BY POSITION. 
 40. General Remarks. 
 
 A SHORT vowel followed by two or more consonants or by 
 ^, ^, ip, was usually considered long, although in most cases 
 it was pronounced as short. 
 
 A short vowel was always regarded as long, when it was 
 followed in the sa/ne wo#:l by two or more consonants, 
 except a mute and a liquid ; c.i;^. rifxvw, Eupucr^evs, e/^Ae'yw, 
 omnis, aspicio, adliio. 
 
 41. Greek. 
 
 Generally in Homer a short vowel is lengthened before a 
 mute and a liquid, whether these occur in the same word 
 or at the beginning of the following. Only if the second 
 vowel is A. or p (except |SA, yA, 8A), a final vowel sometimes, 
 and a medial vowel less often, remains short. 
 
 Hence the reading in II. XVI. 857, Xiirovcr avSpoTrJTa kuI 
 yj^rjv, is wholly corrupt and long since rejected. 
 
 Homer's example was followed by the older epic poets, 
 and the iambic writers like Archilochus. 
 
 On the other hand, in the Old Comedy, only ^A, yA 
 usually, and y/x, yv, S/x, Sv always, lengthen a preceding 
 short vowel {correptio Attica). The tragedians conform 
 more to the example of Homer. 
 108
 
 Latin. 109 
 
 Consonants, not mutes and licjuids, always render the 
 preceding vowel long, although Homer shortens final syllables 
 
 preceding the words a-Keirapvov, SKa/^avSpos, ZolkwOo?, ZcAeta, 
 
 as otherwise these would not fit the metre. 
 
 42. Latin. 
 
 Plautus and Terence did not recognize the lengthening of 
 a preceding short vowel by a following mute and liquid 
 except gm and g/i, and usually in other cases, especially in 
 dissyllables, e.g. Vie, ute, vnmo, esse, ferentarius, disregarded 
 the rules of position. 
 
 On the other hand, the dactylic poets followed the Greek 
 system in the middle of a word at least, although a mute 
 and liquid exercised a lengthening force on the preceding 
 short less frequently than in Homer. Hence many com- 
 binations of consonants, which appear frequently in (Ireek, 
 are quite foreign to the Latin language. 
 
 Cycmis is always written with a short y ; in latrare the a is 
 long by nature. 
 
 A final short vowel remains short when followed by a 
 mute and liquid except gn. 
 
 Hence after a preceding short final vowel we must write 
 Cnosi/s, Cnidiis, narits, natits, nariis, instead of Gnosiis, etc. 
 
 Short final syllables before other combinations of con- 
 sonants appear usually only in those cases where syllaba 
 anceps is allowed, as in Horace, Ep. 17, 26: — 
 
 levare tcnt« spiritu praccordia. 
 
 The oldest Roman poets sometimes permit a short vowel 
 to retain its quantity before impure s (,s- followed by a con- 
 sonant), especially after a dactyl in the first and fifth foot of 
 the hexameter. This usage is limited in the Augustan poets
 
 no Lengthening by Position. 
 
 to the words Zdcyniliiis, Scaniaudcr, smdragdus {zmarag- 
 dus), with these few exceptions; in the Satires of Horace; 
 Propertius; Vergil, ^n. XL 309; Ovid, Halieutica, 120; 
 Phsedrus, III. 3, 14; apjD. 9, 12. 
 
 Lengthening of a short final syllable when the next word 
 begins with two or three consonants, does not occur in the 
 poets considered here, except in Tibullus I. 5, 18; 6, 34; 
 and in general it is very rare.
 
 NINTH SECTION. 
 
 -»<»J»io«~ 
 
 HOMERIC PROSODY. 
 43. Peculiarities of Prcsody in Homer. 
 
 In consequence of the mobility of the Ionic dialect, in 
 Homer's time the quantity of many syllables might vary, or 
 at least the lengthening of short syllables would be less 
 offensive to tlie ear, especially in words which otherwise 
 would fit the hexameter metre either with difficulty or not 
 at all, and in words of very frequent occurrence. The 
 vowels, a, L, V, show the most frequent exceptions to the 
 usual quantity. 
 
 Thus the penult is common in Homer in the names of 
 the chief heroes of the poem, 'A;)(tXevs and 'OSvo-eus ; the 
 first syllable is often lengthened in avy'ip, "Aprj?, 'AttoAAwvos, 
 
 etc., draAXco, lAaos, Tvpiv, TrtaiVw, rt'w, pvofxai, voutp, vo) ', 
 always lengthened in d^avaros, dKdfJ.aTO<;, diroviovTo, dyopdaade, 
 Stoyeves, Zecf)vpLr), 7rto/xei/os, lIpLafXLSr]<;, Swap-ivoLo, Ovyaripeaai. 
 Sometimes this lengthening is expressed in the script, as in 
 rivep.6ei<i instead of dv€p6a<;, r/v'i instead of cvs, Aiwvvo-os 
 instead of AtoVvo-o?. The liquids A, p., v, p, also s and occa- 
 sionally 7T and 8, helped on the lengthening of short vowels, 
 because they were easily doubled in pronunciation, espe- 
 cially in compounds and after the augment. Thus for 
 instance /<aTdAoc/)aSeta. This lengthening is frequently ex- 
 pressed by the doubling of the consonant ; thus in lAAa/Je, 
 
 III
 
 112 llouwnc Prosody. 
 
 efxixaOe, irdwaae, oTrTrws, and also ill 'A^^tAAeus and '08v(Ta-ev<i. 
 For ixiXavi (II. XXIV. 79) another reading gives n-uXavi. 
 
 For well-known reasons, metrical licenses not permitted 
 elsewhere occur frequently in the first foot ; thus e.g. Sid 
 (II. III. 357) ; often ineiSr] as a Molossus ( ). 
 
 Many cases of lengthening of short syllables are explained 
 by an older form, as in the first syllable of c(/r)tes, and the 
 
 second in dTro{f)enruiv. Thus the forms /3a0vp^:oo<;, cVvcTre, 
 (Tvve^e^, eSSeio-ev, came from (Scidvapoos, eva-eire, o-i;vo-e;(es, 
 €8(/r)ao-ev. In the same way ea-aeva-a, <^tAo/A/A€t87^s point to 
 an original €.T(/r)eu, </)tAoo-/x£i8r;s. The long vowel in o^is 
 (II. XII. 208) allows the conjecture of an old form ottc^is. 
 The digamma has remained as a vowel in evaSov and avrdp. 
 The later hexameter poets adopted a large number of 
 the Homeric lengthenings already described, partly without 
 knowing the reason for their existence. In some cases the 
 usual Homeric quantity remained the regular one in almost 
 all later poets, as e.g. in d^avaro?, dKa/xaros. 
 
 In Homer the first syllable of kuAo's and i'o-os is always 
 long, but in the Attic poets short. 
 
 In the arsis of the foot a short syllable is sometimes 
 lengthened, especially the vowel t, e.g. in to-rti;, ottw/divw, 
 vTrepoTrXiaiyjaL, etc. On the Other hand, instead of -^vcv and 
 jSXoa-vpwTns we have the other readings yjviv and (3Xo<rvp(x)Tn<; 
 with long finals. 
 
 The long vowel in the arsis of these two verse-beginnings 
 TToAAa Xia-crofxevr] and TroAAa pvard^taKev has not yet been 
 sufficiently explained ; in ttvkvo. pwyaXerjv it is necessitated 
 by the older form ppwyaXeo?. 
 
 The shortening of long syllables, on the other hand, is 
 a very doubtful matter. In cjioivLKoeo-aa the i is not 
 shortened, but oe united by synizesis. The shortening of rj
 
 Peculiarities of Prosody in Homer. 1 1 3 
 
 and oj in the Subjunctive, e.g. (SovXcrat, eiSo/Atv, tofxev, is by 
 no means an arbitrary process, but based upon older 
 formations. 
 
 In general, many old forms remained in the language of 
 Homer, which were employed along with those later forms 
 which were the more usual at the time when the Iliad and 
 Odyssey were composed. This fact explains on the one 
 hand the astonishing fulness and variety of Homeric forms, 
 and on the other their uncertainty and irregularity.
 
 TENTH SECTION. 
 
 <x>;4^o« 
 
 LATIN PROSODY. 
 44. Peculiarities of Latin Prosody. 
 
 In the Roman poets any variation in the quantity of the 
 stem syllable is very rare, except in some proper names. 
 It is to be noticed, however, that in dactylic and logaoedic 
 metres the first syllable is long in irligio, n'ligiosiis, reliquiae 
 (on account of the old form fed in these words, originally 
 written rell) ; in the perfects reperi, repidi, retudi, ?'eiuli, 
 and also elsewhere {fecidd) the first syllable always 
 remained long. In iambic metres we find only frligio, 
 religiosiis, reliquiae. 
 
 On the other hand, the Latin language, in a manner 
 corresponding to its barytone character, allowed frequent 
 shortening of final syllables. 
 
 While, for example, in Plautus not a few endings which 
 afterwards were shortened, of nouns and verbs in ;•, s, i, 
 which had a long vowel in the genitive and in the second 
 person, still appear as long, after Ennius in dactylic poetry 
 these are regularly shortened. The long syllable remained 
 only in /// and petiit (to compensate for the dropped z'), 
 sometimes in sanguis and pulvis, and often in the 2d Pers. 
 Sing, of the Perf. and Fut. Perf. Subj. Thus for example 
 Hor. (IV. 7, 20) : — 
 
 quae dederis animo. 
 114
 
 Peculiarities of Latin Prosody. 115 
 
 The first syllable of -imiis, -ids, of these tenses is com- 
 mon. Greek endings retain their original quantity. Even 
 words ending in a sometimes keep this quantity, e.g, 
 Andromeda, Electra. 
 
 Further, it was customary from an early date to shorten 
 a number of iambic words which were especially common. 
 Plautus and Terence carry this very far, but in the best 
 dactylic poets it is principally the following words which 
 appear shortened : ego, duo, here, bene, male, eito, viodo, ita, 
 quia, nisi, quasi. The final is common in ///////, ///'/, sibi, 
 ibi, ubi. Ulnnani and ubivis have a short vowel ; ulnque, 
 utroblque, ibidem, a long vowel. / is long in utl, sieutl, 
 velut'i, but short in utinam, utujue, nuflquam. 
 
 The poets went still farther with the ending ; for while 
 the older dactylic poets shortened the only in some 
 iambic substantives and verbs, as homo, puto, dabo, and 
 always in nescio quis ; in Vergil, Horace, TibuUus, and 
 Propertius, some cretic feet are also shortened, as Polio, 
 dixero; in Horace's Satires quomodo ; in Propertius yf// ^(5. 
 
 Ovid, except in the Metamorphoses, shortens variously 
 cretic and also spondaic words (always Sulmo, Naso), 
 among them the adverb e?go. Of the poets of the first 
 century after Christ many shorten the generally in sub- 
 stantives (except in Greek words like Piilo), in verbs, in 
 several adverbs, and the numerals ambo, oeto, and finally 
 (though very rarely) in tlie ablative of the Genmd. The 
 o of the interjection id is always long.
 
 ELEVENTH SECTION. 
 
 -00^:^CK>- 
 
 LENGTHENING. 
 
 45. Lengthening by the Thesis at the End of a Word. 
 
 General Remarks. — A short syllable at the end of a 
 word coukl be lengthened more easily than one in the 
 middle of the word, since in the former case a short pause 
 of the voice naturally ensues. 
 
 46. Greek. ' 
 
 Hence in Homer, the lengthening of short vowels occurs 
 especially before those liquids X, /x, v, p, which are easily 
 doubled, and occasionally, too, before 8 and a, though in 
 connection with this it must be noticed that words now 
 beginning with the former letters then began quite fre- 
 quently with two consonants ; e.g. fprjyvfXL, cr/xoipa, avevpr/, 
 
 The letter p often has the effect of lengthening a preced- 
 ing short vowel in the dialogue as well as in the lyric parts 
 of the Attic drama. 
 
 A final syllable ending in a consonant is also unhesi- 
 tatingly lengthened. This happens generally in the caesura 
 or before punctuation, but rarely in the first or sixth 
 thesis. 
 
 Cases where the following word has the digamma, as II. 
 I. 74, jaeATToi/res ptKapepyov, are of course not counted here. 
 ii6
 
 Lengthenmg by the Thesis. 1 1 7 
 
 The later epic poets before Nonnus imitate Homer with 
 more or less frequency ; less often still the didactic and 
 bucolic poets, sometimes without understanding the lin- 
 guistic reasons for their licenses. 
 
 In lyric and dramatic measures the lengthening of final 
 syllables is not produced by the simple force of the thesis, 
 especially in the iambic and trochaic verses. In the dactylic 
 pentameter, also, such lengthening is very rare and usually 
 doubtful, even in the first half of the verse, and at the caesura 
 itself. 
 
 47. Latin. 
 
 In this respect also the Romans followed much stricter 
 rules than the Greeks. 
 
 Monosyllables are never lengthened, nor short syllables 
 ending in a vowel, except that Vergil (/En. III. 464) 
 seems to have permitted gravia. In the same way (Ennius 
 Ann. 149), aquilci. The enclitic que is sometimes length- 
 ened in Vergil, in imitation of Homer, and after him in 
 Ovid's Metamorphoses, but always when preceded by a 
 dactylic or spondaic word, and followed by a second que 
 with the foot w or w w _ w, e.g. (.-En. III. 91) : — 
 
 limaqw^ laurusque dei totus'iue moveri. 
 
 This is allowed only when two consonants or a double 
 consonant, a liquid or s, follow. Elsewhere Vergil, in imita- 
 tion of Ennius, lengthens a final syllable (as in a case of 
 hiatus) not only at the caesura and before Greek words, 
 but also when three short syllables close a word or punct- 
 uation follows it. Thus we find not only (Eel. X. 69 ; 
 Georg. I. 13S): — 
 
 omnia vincit awcr. ct nos cedamus amori. 
 Pleiaa'ai Hyadas claramque Lycaonis Arcton.
 
 1 1 8 LoigtJicning. 
 
 But also (Georg. II. 5; ^'En. XI. 11 1) : — 
 
 muneribus, tibi pampineo gravi(/»j autumno. 
 ora.iis ? equidem et vivis concedere vellem. 
 
 There are about fifty cases in Vergil of the lengthening 
 of final syllables ending in a consonant. 
 
 The other poets allow this lengthening only at the regular 
 caesura, or before following Greek words, and all very 
 rarely. Ovid ten times; Horace (in the logaoedic measures 
 also) eleven times : Sat. I. 4, 82 ; 5, 90 ; II. i, 82 ; 2, 47 ; 
 3, i; 260; Odes I. 3, 36; 13, 6; II. 6, 14; III. 16, 26; 
 24, 5 ; TibuUus four times ; Propertius three times. 
 
 This lengthening does not occur in iambic and trochaic 
 measures.
 
 INDEX. 
 
 Accius, 25, 26. 
 Adonius, 55. 
 ^schylus, 20. 
 Alcseus, 17. 
 
 Alcaic Decasyllabic, 66. 
 Alcaic Hendecasyllabic, 66. 
 Alcaic Strophe, 68. . 
 Alcman, 19. 
 Alexandrians, 23, 24. 
 Alfius Avitus, 34. 
 Alliteration, 46. 
 Anacreon, 18. 
 Anacreontic, 18. 
 Anacrusis, 40. 
 Ananius, 18. 
 
 Anapaestic Metres, 55, 56. 
 Antiquarian tendency in Roman Ver- 
 sification, 33. 
 Antistrophe, 19, 43. 
 Aphaeresis, 97. 
 Archilochian Strophe, 70, 71. 
 Archilochian Systems, 72. 
 Archilochus, 17. 
 Arion, 19. 
 Aristophanes, 21. 
 
 Aristophanic, see Lesser Sapphic, 66. 
 Arsis, 38. 
 
 Resolution of, 48, 49. 
 Asclepiadean Strophes, 69, 70. 
 Asynartete, 41. 
 Ausonius, 35, 66, 67. 
 Avienus, 34. 
 
 B. 
 
 Bacchylides, 19. 
 Basis, 38, 40. 
 Besantinus, 24. 
 
 C. 
 
 Caesura, 5. 
 
 Caesuras of the Hexameter, 51-54. 
 
 Caesuras of the Trimeter, 59, 60. 
 
 Callinus, 16. 
 
 Calvus, 28. 
 
 Cantica, 26. 
 
 Catalectic Verse, 41. 
 
 Catullus, 28, 29. 
 
 Cha^remon, 23. 
 
 Choral Lyric, i8, 19. 
 
 Claudianus, 32. 
 
 Crasis, 97. 
 
 Dactylic Metres, 50-55. 
 Dancing, 24. 
 
 Development of Ancient Versifica- 
 tion, 13, 14. 
 Diaeresis, 32. 
 
 Dialogue in Attic Drama, 20, 21. 
 Dialogue in Roman Drama, 25, 26. 
 Dipody, 40. 
 Distich, 16, 54, 67. 
 Dithyramb, 19. 
 Dochmi.ic Rhythm in Attic Drama, 
 
 22. 
 
 119
 
 120 
 
 Index. 
 
 Dosiadas, 24. 
 Drama, Attic, 20-23. 
 Drama, Roman, 25-27. 
 
 Elegiambic Verse, 67. 
 Elision, 96, 97. 
 
 in Greek, 97, 98. 
 
 in Latin, 98-101. 
 
 variation in, 101-103. 
 Enclisis, 86. 
 Ennius, 25-28. 
 Epode, 43. 
 
 of Horace, 71-73. 
 
 in Doric Lyric, 19. 
 
 in Attic Drama, 21. 
 Euphony, 40, 41, 44. 
 Euripides, 21. 
 
 Fronto, 33. 
 
 Georgius Pisides, 34, 35. 
 Glyconic Verse, 64. 
 Greater Archilochian, 66. 
 Greater Asclepiadean, 65. 
 Greater Sapphic, 65. 
 Gregorius Nazianzenus, 34. 
 
 H. 
 
 Hesiod, 15. 
 Hexameter, 50-55. 
 Hiatus, 103. 
 
 in Greek, 104, 105. 
 
 in Latin, 106, 107. 
 Hipponactean Iambics and Trochees, 
 
 18, 29. 
 Hipponactean Strophe, 71. 
 Hipponax, 17, 18. 
 Homer, 15, 16. 
 Horace, 30, 31. 
 Hymns, Christian, 34, 35. 
 Hypercatalectic Verse, 41. 
 Hypermeter, 101. 
 
 I. 
 
 lambelegiac Verse, 67. 
 Iambic Metres, 57, 62. 
 Iambic System, 72. 
 Ibycus, 19. 
 Ictus, 42. 
 
 Ionic Decameter, 64. 
 Ionic Strophe, 71. 
 Ithyphallic Verse, 63. 
 
 Lrevius, 28. 
 
 Lesser Asclepiadean, 65. 
 
 Lesser Sapphic, 66. 
 
 Logaoedic Metres, 41, 64-66. 
 
 Lucilius, 26-28. 
 
 Lucretius, 28, 29. 
 
 Lyric Poetry of the ^olians, 17, 18. 
 
 of the Dorians, 18, 19. 
 Lyrical Parts in Attic Drama, 20-23. 
 
 M. 
 
 Metre, 37. 
 
 Metrical Licenses, 74-76. 
 
 Music, 24. 
 
 Myurus, 16. 
 
 Nsevius, 25. 
 Nonnus, 23. 
 
 Ovid, 29, 30. 
 
 N. 
 
 O. 
 
 P. 
 
 Pacuvius, 25. 
 
 Peculiarities of Prosody in Homer, 
 111-113. 
 in Latin, 114, 115. 
 Pentameter, 16. 
 Period, 19, 23, 40. 
 PhEEdrus, 26. 
 Phalascean Verse, 18, 28. 
 Phcrecratean Verse, 65.
 
 Index. 
 
 121 
 
 Pindar, 19. 
 Plautus, 25, 26. 
 Porfyrius Optatianus, 33. 
 Position, 108, 109. 
 Propertius, 30. 
 Piudentius, 35. 
 Punctuation, 43-46. 
 Pythiiambic System, 72, 73. 
 
 Quantitative Principle, 13. 
 
 R. 
 
 Rhyme, 46, 47. 
 Rhytiim, 37. 
 Rhythmical Poetry, 36. 
 
 Rhythmical Structure of the Verse, 
 
 77-79- 
 
 of the Hexameter and Pentam- 
 eter, 79-83. 
 
 of the Remaining Metres, 83-85. 
 Rutilius Namatianus, 32. 
 
 Sapphic Strophe, 68, 69. 
 
 Sapphic Verse of Fifteen Syllables, 66. 
 
 Sappho, 17, 18. 
 
 Saturnian Verse, 24, 25. 
 
 Seneca, 32. 
 
 Septimius Serenus, 33. 
 
 Simmias, 24. 
 
 Simonides, 19. 
 
 Sophocles, 21. 
 
 Sotadic Verse, 23. 
 
 Spondi.tzon, 51. 
 
 Stesichorus, 19. 
 
 Stichomythy, 20. 
 
 Strophe, 19, 43. 
 
 Strophes in Later Times, 32, 34. 
 
 Syllable, Last, of the Verse, 42, 43, 48. 
 
 Symmetry, 40. 42. 
 
 Synapheia, 42, 43, 48. 
 
 Synizesis in Greek, 92. 
 
 in Latin, 93-95. 
 Syrus, 26. 
 System, 43. 
 
 Terence, 25, 26. 
 Terentianus Maurus, 33. 
 Thesis, 38. 
 
 Resolution of, 48, 49. 
 Tibulhis, 30. 
 Tmesis, 87, 88. 
 
 Tragedy of the Time of Augustus, 31. 
 Trochaic Metres, 62, 63. 
 
 V, 
 
 Varro Atacinus, 28. 
 Varro Reatinus, 29. 
 Vergil, 29, 30. 
 Verse, 40, 41. 
 Verse-feet, 38-40. 
 
 Versification, Ancient, in its Develop- 
 ment, 13, 14. 
 
 Greek, compared with Roman, 14. 
 
 Final State of, 34. 
 Versus Politicus, 36. 
 Vowels, Successive Sounds, 89-91.
 
 Allyn &- Bacon .... Boston. 
 
 Adopted at the High Schools of 
 
 Bath and Hallovvell, Maine 
 Hartford, Norwich, and Meriden,Ct.; 
 Ogdensburgh, Ilion, Plattsburgli, 
 Adams, and Flushing, New York ; 
 Jersey City, Montclair, and 
 Plainfield, New Jersey ; York 
 Pittsburgh, and Beaver, Pa. ; 
 Cleveland and Findlay, Ohio ; 
 Decatur and Ouincy, 111 
 Saint Paul, Minnesota 
 
 and at 
 
 Phillips Academy 
 Exeter, N. Hampshire; 
 Wesleyan Academy, 
 Wilbraham, Mass.; 
 Thayer Academy, 
 Braintree, Mass. 
 
 
 i2mo. 
 Bound in 
 Leather. 
 With col- 
 ored Maps, 
 Plans, Illus- 
 trations, and 
 Vocabulary. 
 60. 
 
 Colgate Academy, 
 Hamilton, N.Y.; 
 Friends' Central 
 High Schoo 
 Philadelphia; 
 Horner Sch., 
 O -x f o r d, 
 N, C. 
 
 m^ 
 
 Adopted at 
 
 Berkeley School, 
 
 New York City; 
 
 Dr. Pingry's School, 
 
 Hasbrouck Institute, 
 
 Dearborn-Morgr.n Sch., 
 
 Peddie Institute, N.J. ; 
 
 De Lancy School and 
 
 Germantown Academy, Pa. 
 
 In the Prep. Defts of 
 
 Bucknell University, 
 Westminster College, 
 
 De Pauw University, 
 
 Oberlin College, Cornell College, 
 
 University of Neb., Iowa College, 
 
 Colorado College, Hanover College, 
 
 and many other 
 
 schools and colleges.
 
 Kelsey's Anabasis. 
 
 F. S. Morrison, High School, Hartford, Conn. — 
 I have exaiiiincd it with interest and pleasure; and 
 am particularly pleased with the maps, the introduc- 
 tory matter, and the table of idioms, supplying so 
 much that is val"able yet lacking in our other editions. 
 
 E. E. Wentworth, Principal High School, Great 
 Barriugton, Mass. — In my opinion, it leads all other 
 editions by as wide a stride as was made by Kelsey's 
 Caesar, and that is saying a great deal. 
 
 C. S. Halsey, Classical Institute, Schenectady, 
 iV. Y. — The notes are excellent; they furnish just 
 the aid that is needed and no more. Especially 
 worthy of commendation are the vocabulary and 
 pages devoted to Idioms and Phrases. 
 
 John G. Wight, H/gh School, Worcester, Mass. 
 
 — For complete and comprehensive excellence — in 
 respect to text, clear and exhaustive introduction, 
 instructive illustrations, maps, notes, and vocabulary 
 
 — it is unequalled by any text-book with which I am 
 acquainted. 
 
 D. C. Farr, Glens Falls Academy, A^. V. — I am 
 sure every teacher ought to introduce the book first, 
 and then congratulate you on making the best edition 
 of the Anabasis yet published. 
 
 Prof. H. S. Kritz, IVabask College, Crawfords- 
 ville, Ind. — Nothing so beautiful and elegant in 
 typography, or more neat and substantial in binding, 
 has come under my observation for a long time. 
 The editorial work has been done with rare appre- 
 ciation of the wants of the students. 
 
 Dr. William B. Corbyn, Principal High School, 
 Qnincy, III. — It is admirable. We shall adopt it 
 immediately.
 
 Allyn &■ Bacon Boston. 
 
 Keep's Homer's Iliad 
 
 AND 
 
 Thurber's Vocabulary 
 
 Are now made hi the following forvis : 
 
 Iliad, Books I.-III., cloth $0.90 
 
 — bound with Vocab., leather 1.20 
 
 Iliad, Books I.-VI., leather 1.40 
 
 bound with Vocab., leather 1.60 
 
 Vocabulary to Iliad, I.- VI., paper . . . 0.50 
 
 The editions of Books I.-VI. contain a fac- 
 simile of a page of the famous Venetian Manu- 
 script of the Iliad. 
 
 No pains have been spared to make these 
 the most nseftil and practical editions of 
 the Iliad that can be put into the hands of 
 a beginner, and they are almost univers- 
 ally accepted, not oiily as the best school 
 editions of any part of Homer, but also 
 as text-books of altogether exceptional 
 merit. 
 
 Prof. M. L. D'Ooge, U7iiversity of Michigan. — 
 It is unquestionably the most useful and the most 
 attractive school edition of the first six books of the 
 Iliad that has ever been prepared for English-speaking 
 students. The Introduction, the Essay on Scanning, 
 and the Sketch of the Epic Dialect are each admi- 
 rable in its way; while the notes in general seem 
 well suited to the wants of those for whom the edition 
 is designed.
 
 Allyn Sr Bacon .... Boston. 
 
 * 11- •) Greek Prose 
 
 AllinSOn S composition. 
 
 i6mo. Cloth. 212 pages, ^i.oo. 
 
 This book is intended foi- use in writing con- 
 nected Greek jirose. It presupposes a knowl- 
 edge of forms, and contains : 
 
 I. Notes on Idiom and Syntax , explaining 
 the use of the article, jDronouns, participles, 
 moods, and tenses. 
 
 II. Stimmary of Rules for cases and accents. 
 
 III. Exercises, carefully graded, and arranged 
 in three sections. 
 
 IV. A general vocabulary. 
 
 Dr. Morris H. Morgan, Harvard College. — I 
 have carefully examined the book, and believe that 
 it is a most excellent guide. Especially am I pleased 
 with the fact that the exercises are based on real 
 Greek models and expressed in the English language. 
 This book is not disfigured by the vile jargon which 
 was invented by the adversary for particular use in 
 such manuals, a language certainly spoken nowhere 
 upon earth and, I trust, not elsewhere. I hope the 
 book will meet the great success it deserves. 
 
 Prof. Robert Sharp, Tidane University, N'ti.u 
 Orleans. — I need not say that I am pleased with 
 the book ; my action in adopting it for my classes 
 shows as much. But I will say that it supplies our 
 needs here better than any other book I have yet 
 seen.
 
 Allyn &- Bacon .... Boston. 
 
 Kelsey's 
 
 Selections 
 from 
 
 Ovid. 
 
 With Introduction, 
 Notes, Vocabula- 
 ry, and ten full- 
 page Illustrations. 
 
 1 2 mo, half leather. 
 
 $1.25. 
 
 L C. Hull, Lawrenceville School, iV /. — Pro- 
 fessor Kelsey has again shown how admirably he 
 can meet the wants of good teachers. But his edi- 
 tion of Ovid does more than this. We have all 
 read tlie stereotyped testimonial which tells that the 
 author has satisfied a long felt want. But a really 
 good text-book should inspire as well as gratify an 
 appetite The book that merely chronicles the best 
 system in vogue at the time of its pul^lication, adds 
 little to the history of teaching. This edition of 
 Ovid must quicken the poetic sense of even the most 
 prosaic teachers. I cannot help feeling that many a 
 pupil will thank Professor Kelsey for having shown 
 how the same themes that inspired Ovid still live in 
 our own Engli-h classics. Hitherto we have had to 
 send our boys to Bulfinch for such proofs. Now we 
 have had given to us a manual of mythology, a 
 scholarly edition, a Latin classic, and an abundance 
 of poetic inspiration, all at once. Ovid did not wriie 
 to exhibit the grammatical usages of his day, or to 
 help make philologists; and Professor Kelsey has 
 not made the mistake of supposing that his verses 
 are best used when they are made to serve such 
 purposes.
 
 Allyn &- Bacon .... Boston. 
 
 Kelsey's 
 
 Caesar's 
 
 Gallic 
 
 War. 
 
 Books I. to VII. 
 
 W ith Introduction, Notes, 
 Vocabulary, Table of 
 Idioms, a.id twenty full- 
 page Illustrations. 
 
 i^mo, half leather. $1.25. 
 
 Throughout the book every ejfort has 
 been made, by way of illustration ajid 
 coviine7it, to render the study of Caesar 
 attractive and useful, — a means of culture 
 as well as of discipline. That the result 
 has been to produce the best-equipped edi- 
 tion of the Gallic War is generally con- 
 ceded. 
 
 Dr. J. H. Hanson, Classical lusiitttte, Watcrville. 
 Me. — It is the ideal Caesar realized. That it sur- 
 passes all its competitors must, it seems to me, be 
 the universal verdict. 
 
 Richard M. Jones, Head-Master Wm. Perm 
 Charter School, Philadelphia. — Kelsey's Caesar is in 
 my judgment the nearest approach yet made in this 
 country to what a school edition of an ancient classic 
 should be. 
 
 Dr. James M. Milne, Norvtal School, Ouconta, 
 N. V. — I am delighted with it. Its accurate scholar, 
 ship, its broad scope, its illustrations, the clearness 
 of the text, all make it the ideal book.
 
 Allyn S- Bacon .... Boston, 
 
 Pennell's 
 
 Pennell's 
 
 Ancient 
 
 Ancient 
 
 Greece. 
 
 Rome. 
 
 i5mo, 198 pages. 6octs. 
 
 i6mo, 284 pages. Co cts. 
 
 Revised Editions witJi Alaps and Plans. 
 
 Tor these new editions the books have been 
 entirely rewritten, the size being increased by 
 about one half. The distinctive features are: 
 
 I. The concise and readable Form in which 
 tlie leading facts are presented ; minor details 
 and unimportant names are omitted. 
 
 II. The Maps and Flans, drawn and en- 
 graved specially for the books, containing all 
 the data, and only the data, necessary for fol- 
 lowing the story. 
 
 III. The Index, serving also as a key to the 
 pronunciation of proper names. 
 
 IV. Recent Examination Fapers, used at 
 Harvard, Yale, and by the Regents of the Uni- 
 versity of New York. 
 
 One or both hoo/cs are in use in tlie High Schools at 
 Portland, Bath, Waterville, Me. ; Nashua, N. II. ; 
 Salem, Somerville, Taunton, Waltham, Mass. ; 
 Utica, Ithaca, Lockport, Geneva, Warsaw, N. Y. ; 
 Wooster, Ohio ; Ottawa, 111. ; Milwaukee, Wis.; 
 Kansas City, Mo.; Mankato, Minnesota; 
 
 Also at Phillips Exeter Academy, St. Paul's 
 School, Concord; Albany and Colgate Academies, 
 Berkeley School, N. Y. ; Peddie Institute, N. J. ; 
 Smith Acad., St. Louis; University School, Chicago; 
 And the preparatory departments of many colleges.
 
 ^-^^ 
 
 University of California 
 
 SOUTHERN REGIONAL LIBRARY FACILITY 
 
 405 Hilgard Avenue, Los Angeles, CA 90024-1388 
 
 Return this material to the library 
 
 from which it was borrowed. 
 
 UNIVERSITY ot CALIFUKNLfw 
 
 AT 
 LOS ANGELES
 
 3 1158 00314 1800 ^ 
 
 REGIONAL LmRARY f ACIUTY 
 
 ^A 000 343116