‘11 Sudwfli ‘~ MW .~ mm w 1 . kfll‘r 1 ‘K 5 mi’. ‘ ‘ 'flfiqwle - " ' "HQ? ' ‘III 7 h \ a I : ,. l I ‘1 , \ \ y L Q .v *3 I 1 I J l ‘Hull . ; ~QI’IQQ,%\’ 31mm“?! ' "?~‘\l§'"# i1" , u‘ ~D> ' ‘ --:-‘.';-;§\‘ o-qLe-A- pail #5 av . "all ‘a flaw] ' m: MM 1‘ u (""3 Aulflv- 1-" “hum a5’! hush; > if 5: ' W m: =, . ' ' ~ - 1 n no "QIIIIHU "< ‘ ‘ ' ‘4 ' “HM: in’, . H. 1 an. _ _ V _ V ."k'fl'm'lm'l'tjl .r- k , ' ‘ ; ,Mum“ " "law", - . ‘ ' u l.\ ' u l -:l’ 1-?‘ “n --~>~,~~ .- w, §B?%2€ (0/ a HISTORY OF AN GIENT GEOGRAPHY AMONG THE GREEKS AND BOMANS FROM THE EARLIEST AGES TILL THE FALL OF THE ROMAN EMPIRE. BY E. H. EUNBURY, F.R.G.S. WITH TWENTY ILLUSTRATIVE MAPS. IN TWO VOLUMES.—-VOL. I. LONDON: JOHN MURRAY, ALBEMARLE STREET. 1879. The right of Translation is reserved. ‘fdfi'mmwz, 31- 17-l7u’ ‘my: 91 V760». LONDON I PRINTED BY WILLIAM GLOWES AND SONS, STAMFORD s'rmm'r AND OHARING (moss. PREFACE. —O<>O———- THE present work is designed to supply a deficiency, which I have reason to believe has been felt by classical scholars. While there is no Want bf elementary trea- tises on ancient geography, and the requirements of the more advanced student are fully answered by the elaborate articles in Dr. Smith’s Dictionary of Ancient Geography, so far as relates to the details of particular @Iizésknd countries, it is remarkable that there does not exist in the English language any such historical review of the subject as a whole, as is to be found in the introductions to the works of Mannert, Ukert, and Forbiger. But even these treatises, besides being con- fined to students who are familiar with the German language, can hardly be said to meet the demands of either the scholar or the geographer at the present day. Ukert’s introductory volume was published as far back as 1816, and that of Mannert still earlier, while the more recent work of Forbiger (itself published in 1841) is so disfigured by the accumulation of cumbrous and unnecessary lists of names as to be altogether repulsive to the English reader. Since the date of the works in question, not only has there been a great advance in classical scholarship, and the critical study of the an- cient authorities; but still greater progress has been a 2 iv PREFACE. made in the detailed examination of the regions and localities described by ancient geographers, many of which were very imperfectly known in modern times down to a recent period. It is not too much to say that there is scarcely a disputed question in ancient geography upon which some additional light has not been thrown by local researches and investigations within the last fifty years. ' It has been my object in the following pages to pre- sent to the reader the results of these recent inquiries : and while basing my work in all cases upon a careful and critical examination of the ancient authorities, to avail myself to the fullest extent of the assistance to be derived from modern travellers and geographers. Several instances have occurred during the progress of my researches, where additional information of this kind has served to elucidate questions which were still obscure when the chapters in which they are discussed were originally written. At the same time it must be constantly borne in /mind that it is the main purpose of a historical review, such as the one I have attempted to give, of the pro- gress and development of geography in ancient times, to record the ideas formed and the information pos- sessed by the different authors from whom we derive our knowledge; and with a view to this I have en- deavoured in every instance, so far as possible, to arrive at my conclusions from a conscientious and independent study of the ancient authorities themselves, before con- sulting or referring to the comments and discussions of modern writers. In pursuing this plan, I have been materially aided by the valuable editions of Strabo_by,_ PREFACE. V Kramer, and of Pliny by Sillig—both of them subse- quent to the publication of the three German text-books above referred to—as well as by the admirable edition f of the Geographz' Grnecz' fllz'nores edited by Dr. G. ‘Muller, a work which has conferred an inestimable boon upon all students of ancient geography. In referring, as I have done above, principally to the labours of the Germans in the same field with myself, I must not be regarded as ignoring What has been done in this country and in France in connection with the same subject. The first volume of Mr. Oooley’s His- t07y_0f Maritime and Inland Discovery (published in Dr. Lardner’s Gyclopaedia in 1833) contains a good popular sketch of the progress of geographical know- ledge in ancient times; but is certainly not adapted to meet the requirements of the classical scholar. The far superior work of M. Vivien de St. Martin—the £1219- toz're de la Ge’ograp/zz'e et des Décozwertes Ge'ogra- plagues—which was not published until the present work was already far advanced—contains a sufficiently full review of the whole subject for the purposes of the general reader: but the limits within which the author was necessarily confined, in order to include in a single volume the whole history of geographical progress down to our own time, precluded his entering in detail into many questions the discussion of which formed an essential part of my own plan. In one respect I have thought it expedient to depart from the example of my German predecessors, who have deemed it necessary to treat separately of mathematical and physical geography, apart from the descriptive and historical portions of the subject. In a work of which vi PREFACE. the historical form is an essential character, it appeared to me desirable to bring together all portions of the subject under one view; so that the reader might see at once the condition of geographical knowledge at every successive period :—-say for instance in the time qf ‘Eratosthigres, Strabo, Pliny and Ptolemy—instead of having to turn to several different chapters for the information he requires. The unity and completeness thus given to the successive portions of the historical review appear to me greatly to outweigh the advan- tages derived from the more methodical subdivision. The comprehensive character of such a work as the one now submitted to the reader unavoidably exposes it to one disadvantage. It must of necessity comprise many subjects which could not be fully discussed with- out extending it beyond all reasonable limits. The W; that of the Anabasis of Xenopmhgngaimqf~ theflqaimpaignlsfof Aleziaiideriijflthe vol- age; iqf Nearchusmand a‘ii'no—might well require _a volume FREE” of them“ instead of a chapter: while \ \_._...'-’ several special topics, such as the passage of the Alps by Hannibal, the landing of Caesar in Britain, or the defeat of Varus in Germany, have to be dispatched in a few pages, though each of them has formed the sub- ject of numerous tracts and treatises amounting to a little literature of its own. In such cases I have been compelled to bring together in a brief summary the grounds of the judgement at which I have arrived in each case, without being able to enter into the discus- sion as fully as I could wish. If in some cases I may have in consequence appeared to pass rather slightly over the arguments on the other side, I can only beg PREFACE. vii my readers to believe that this arises from the necessity of the case rather than from any overweening confidence in my own conclusions. In some instances, on the other hand, my readers may be disposed to complain that I have left questions unsettled, without pronouncing any opinion, where the evidence does not appear to me such as to afford rea- sonable grounds for a decision. To both classes of objections I can only reply in the words of Cicero— adopted by Ukert as the motto of his work——-“ sequimur probabilia, nec ultra quam id quod verisimile ogpurrerit progredi possumus, et refellere sine pertinacia, 6t refelli sine iracundia parati sumus.” It is only those who have devoted a considerable amount of attention to the study of ancient geography who are fully aware of the difficulties that beset the path of the inquirer at almost every step. But these difiiculties have in many instances been greatly aggra- vated by the unwillingness of modern writers to apply to the statements of ancient authors the same rules of reasonable criticism by which they would be guided in other cases. Not only is geography in its very nature 9/ a progressive science, but the slightest attention to its , history in mediaeval or modern times will show that \ the steps of its progress are often vacillating and uncer- E tain. Vague and fluctuating ideas concerning distant r’e'gions will be found floating as it were in a dim haze of twilight long before their outlines come to be dis- tinctly discerned. Even the most trustworthy travellers are compelled to rely to a certain extent upon hearsay evidence, in regard to the countries or provinces that they have not themselves visited; and in the case of viii PREFACE. less qualified observers it is generally difficult to dis- tinguish what they have really learnt from their own observation and what they have derived from other sources. Without recurring to the case of Marco Polo and other mediaeval travellers, it will be sufficient to turn to the map of Central Africa as it stood before the time of Park and Hornemann, and to trace the progress of discovery in that great continent, and the various theories by which geographers sought to explain or reconcile the statements of successive travellers,ni\n order to see what difficulties surround any such attempt\ in the absence of clear and definiteinformation. But if this is the case even in modern times, where the information of the geographer is derived from the statements of trustworthy observers and scientific tra- vellers, far more does it apply to the position of the ancient geographer. Voyages and travels for the pur- pose of exploration were almost entirely unknown : and he had to pick up his information as best he could from the accounts of merchants and casual travellers. The inaccuracy of these reports, and the consequent vague- ness of the statements derived from them, is frequently pointed out by ancient writers. But they had nothing else to fall back upon, and no means of correcting them by more accurate observations. Yet, notwithstanding this obvious consideration, it has been too much the custom in modern times to treat the ancient writers on this subject as if they possessed an authority to which they are certainly not entitled. Instead of at once drawing the line, as would be done without hesitation in the case of a medizeval writer, between what was accurate and trustworthy and what PREFACE. ix was vague and inaccurate, the most fanciful suggestions have been made and ingenious theories invented to account for what was simply erroneous. Even the supposition of vast physical changes has been intro- duced or adopted, rather than acknowledge that Hero- dotus or Strabo can have made a mistake. It has been my endeavour in the following pages, as far as possible, ‘to exercise a discriminating judgement in sifting truth from falsehood; and while doing full justice to the real merits of successive writers, to discard without scruple those statements where they have been obviously mis- led by imperfect information, or by adherence to a mistaken theory. The historical form of the present work necessarily involves a certain amount of repetition. Nor have I been careful to avoid this. Many persons will, I believe, take up particular chapters of the book who will shrink from the labour of perusing the whole: and I have ‘therefore endeavoured to make each successive portion , ——especially the reviews of the different leading authors l on the subject, such as S_trabo, Pliny, or Ptolemy—as \ complete in itself as possible. No English book, as far as‘ I am aware, contains a similar analysis of these well- known authors, who are too often quoted for detached statements by writers who are wholly unacquainted with their real authority and value. I am aware that I shall incur the censure of many of the more advanced scholars of the present day for having adhered to the old orthography of Greek names. I must confess myself one of those who fail to see the advantage of the changes recently introduced: changes that can hardly yet be said to form a permanent or X PREFACE. established system. But, independent of my own pre- ference for the system to which I have been accustomed from my youth, two considerations would in themselves have deterred me from introducing any such innova- tions in the present work. The one is that as my subject included Roman as well as Greek geography, I should have had to follow two different systems of orthography in different portions of the book, and to write the same names in two different modes, according as I was reviewing a Greek or a Latin author. To this must be added that as the book now presented to the reader is, from its nature, in some degree a supplement to the two valuable Dictionaries of Ancient Biography and Geography edited by Dr. Smith, which are at pre- sent in the hands of all scholars, I should have been unwilling to deviate from the practice which has been there adopted. With regard to the orthography of oriental names, which are necessarily of frequent occurrence in the fol- lowing pages, I have contented myself with writing them as I found them in the authority before me, or in the case of well-known names in the mode commonly received. Having no knowledge myself of any of the oriental languages, it was impossible for me to attempt to follow any uniform system in this respect. The present work, like those of Mannert and Ukert, is confined to the geography of the ancient world as known to us through the Greeks and Romans. In the very brief introductory chapter, I have only touched upon that of other nations with reference to its bearing upon the geography of the Greeks at the earliest period when we have any information concerning it. To in- PREFACE. xi vestigate the details of the geographical knowledge—— limited as it undoubtedly was—possessed by the Egyp- tians or Assyrians, would be wholly foreign to my subject, and is a task for which I feel myself entirely incompetent. It may perhaps be necessary to observe that the maps inserted in these volumes are not designed in any degree to supply the place of an Atlas of Ancient Geography, but solely to illustrate the particular sub- jects discussed, or to bring more distinctly before the eyes of the reader the general outline of the geogra- phical systems formed by successive writers, so far as it is possible to extract these from their writings alone. In the preparation of these maps, I have derived the greatest assistance from the series of those contributed by Dr. C. Muller to the valuable “ Atlas of Ancient Geography edited by Dr. Sniith and .Mr. Grove;” and I take this opportunity of expressing in the strongest manner my sense of the obligations that I owe him. Scarcely less valuable is the aid 1 have received from his admirable edition of the Geoyraplii Grceci Minores (already referred to), without which it would have been scarcely possible for me to have executed in a satis- factory manner the portions of my work relating to V,’ the writers in question. CONTENTS OF von 1. O H A P T E R I. m'rnonucroav. § 1. Geography more or less studied by different nations according to circum- stances—little cultivated by the Ohaldaaans or Egyptians. §2. Or by the Jews. N o influence exercised in this respect upon the Greeks. §3. The Phoenicians; their extensive commerce and long voyages. §4. Geographical ‘knowledge derived from thence: its extent and limits very imperfectly known. § 5. Trade of the Phoenicians in tin : the Oassiterides, or Tin Islands. Gades, the centre of the trade. § 6. Their trade in amber, brought from the northern shores of Europe. § 7. The carrying trade of the lEgean and the Mediterranean in their hands. §8. The ()retans the earliest Greek people devoted to navi- gation. Notices of them in the Odyssey. §9. The Trojan War a proof of an advanced state of navigation .. .. .. Page 1 O H A P T E R I I. VOYAGE OF THE ABGONAUTS. § 1. The voyage of the Argonauts a mere legendary tale—originally quite un- connected with Oolchis or the Phasis. §2. Applied by the Greek colonies to the localities on the Euxine. § 3. Various accounts of their return. § 4. That given by Apollonius Rhodius. § 5. Different accounts of earlier writers. Possible basis of truth in the legend. 19 -- PAGE NOTE A. Argonautica of Orpheus .. .. 28 O H A P T E R I I I. HOMERIO GEOGRAPHY. SECTION 1.—-General Views. ( L,’ § 1. Earliest notions of Greek geography derived from the Homeric Poems— these perverted by later writers and commentators. Necessity of adhering to the original. §2. Ideas of Homer concerning the earth xiv CONTENTS OF VOL. I. and heavens. His mention of Atlas. §3. Ideas concerning the sun and stars. §4. No names for cardinal points. Names of winds. § 5. His views of the sea and the Ocean river. § 6. No knowledge of the three continents Page 31 SECTION 2.—-Geogmphy of the Iliad. § 1. Local geography of Greece and the neighbouring seas well known to Homer: but all beyond vague or unknown. §2. The Catalogue of the ships. §3. Catalogue of the Trojan forces. § 4. Extent of know- ledge proved by these very limited. § 5. Vague indications of more distant countries. §6. Knowledge of Phoenicia and Egypt. §7. Vague and fabulous ideas of Ethiopia .. . . .. 39 SECTION 3.-—Geogmphy of the Odyssey. §1. The outer world in the Odyssey—the scene of poetical fictions and legends. § 2. These worked up by Homer into a poetic whole, but not into a geographical system.- §3. Attachment of legends to definite localities. §4. Voyage of Ulysses: the Cicones, Cape Malea, the Lotophagi. §5. The Cyclopes. §6. The Island of Bolus. § 7. The Laastrygones. §8. The Island of Circeh—visit of Ulysses to Hades. The Cimmerians. § 9. Homeward voyage: the Sirens, Scylla and Charybdis. Thrinakia.- § 10. Island of Calypso: voyage from thence to Phaeacia. § 11. The Phzeacians—a fabulous people, erroneously placed in Corcyra. § 12. Ithaca and the neighbouring islands: difii- culties in regard to their description. § 13. Dulichium. N 0 clear idea given of these western islands. § 14. Voyage of Menelaus. Egypt. § 15. The Ethiopians. § 16. Ignorance of Homer of the eastern nations: and of the west of Europe .. .. .. .. 49 PAGE NOTE A. Composition of the Homeric Poems .. 75 ,, B. Homeric idea of the world .. .. ib. ,, C. The four winds in Homer .. .. 77 ,, D The aestrygones .. .. .. db. ,, E The Island of Circe .. .. .. 78 The Planctae .. .. 79 \l \0 '15‘ I Q ' G Course from Ogygia to Scheria . . .. 81 ,, H. Dulichium .. .. .. .. 'ib. I Homeric Ithaca .. .. .. .. 83 CONTENTS OF VOL. I. XV CHAPTER IV. HOMER T0 HECATZEUS. SECTION 1.-—Poetical Notices subsequent to Homer. § 1. Want of materials for tracing the progress of geographical knowledge after Homer. Hesiod: his poems largely interpolated. Geographical notices contained in them. § 2. The Epic Cycle and the Homeric Hymns. § 3. Other poets before the First Olympiad. Aristeas of ' Proconnesus .. .. .. .. .. Page 85 SECTION 2.--Oolonz'es. §1. Great development of Greek enterprise in the 7th and 8th centuries 13.0. Foundation of colonies in Sicily. § 2. In Southern Italy. § 3. At Corcyra and in the Adriatic: and in Campania. § 4:. Massilia and its colonies. § 5. The extension of Greek settlements in Spain and Africa checked by the Phoenicians. §6. Cyrene and its colonies in Libya. § 7. Commercial intercourse Wit-11 Egypt. Naucratis. § 8. Exploration of the Euxine. Colonies of Miletus and Megara. §9. Settlements on the west and south coasts. § 10. On the northern shores and in the Crimea. § 11. Communications with the interior. Aristeas of Procon- nesus. The Arimaspians and Hyperboreans. §12. Destruction of Miletus, and fall of its power. § 13. The Phocaeans : their voyage to Tartessus. Maritime power and trade of the Samians and Rhodians— Colonies of Rhodes. §1¢L Commercial activity and naval power of Corinth. §15. Of Megara. §16. Of ZEgina. §17. Of Chalcis and Eretria in Euboea. Colonies of Chalcis. Inferior position of Athens before the Persian War. § 18. Colonies and commerce limited to the inland seas—the Mediterranean and the Euxine. § 19. Relations of Greeks with the Phoenicians. §20. And with Egypt. §21. Indi~ vidual visits to Egypt and Babylon. Increasing knowledge of Asia after the rise of the Persian monarchy .. .. .. 91 SECTION 3.-'—PhysicaZ Philosophers. 1. Rise of philosophical speculation in Greece. Thales of Miletus, his astro- nomical views. §2. Anaximander: the inventor of maps. § 3. Anaxi- menes. § 4:. Pythagoras and his followers: their improved system of cosmical science. §5. How far influenced by ideas derived from the East. The spherical form of the earth not yet a recognized truth. 120 PAGE NOTE A. Date of foundation of Cyrene .. .. 127 ,, B. Dates assigned to Greek colonies in th Euxine .. .. 128 ,, C. The “ Thalassocraties ” of Castor .. 130 xvi CONTENTS OF VOL. I. oHAPTER'v HEOATEUS . SECTION 1.— Geography of Heeatoeus. § 1. The work of ' Hecateeus of 'Miletus the first regular geographical treatise. § 1. Writers after Hecataeus. His life and date. Imperfect character of existing remains. § 2. His work mostly a Periplus of the Mediterranean, but contained notices of Asia. § 3. His travels. §4. Arrangement of his Work. § 5. Extent of his information, in regard to the coasts of the Mediterranean. 6. He had no knowledge of Western and northern Europe. Imperfect acquaintance with Scythia. §7. His knowledge of Asia in general. He gives the first notices of India. §8. Described Egypt in detail. His notices of the rest of Africa. § 9. His general notions on geography. His map of the world. § 10. His division of the world into continents. § 11. His want of judgement Page 134 SECTION 2.——Hecata2us t0 Herodotus. Hellanicus and Damastes. § 2. Geographical notices in ZEschylus and Pindar. The wanderings of lo in the ‘ Prome- theus Vinctus.’ § 3. The ‘ Prometheus Solutus ’ and the ‘ Persae.’ §4. Pindar: his geographical notions .. .. .. .. 148 PAGE NOTE A. Character of extant fragments of Heca- teeus .. .. 153 ,, B. Intercourse of the Greeks with Etruria ib. CHAPTER VL / HERODOTUS. / SECTION 1.—— General Views—Europe. § 1. Importance of the work of Herodotus in a geographical point of view. I § 3. His travels: their extent. Its desultory and irregular character. Slight notice of the Oarthaginians and Tyrrhenians. §2. Brief notices of Southern Italy and Sicily. § 4. Outline of his views of the con- figuration of the world. Asia. The projecting Actae. § 5. Explana- tion of his views. § 6. His ignorance of the west and north of Europe. He rejected the notion of the Oassiterides, and of the river Eridanus ; and of a northern ocean. § 7. Imperfect knowledge of central Europe. Thrace and the Danube. The Getee and Sigynnae. §8. His de- scription of the Danube and its tributaries. The Oarpis and Alpis. CONTENTS or VOL. I. xvii § 9. The Keltze'and Kynetes. §10. His comparison of me Nile and the Ister: their mouths opposite to one another .. .. Page 156 SECTION 2.—Scythz'a. § 1. Extensive information obtained by Herodotus concerning Scythia. § 2. Derived in part from personal knowledge. He discards the fables current concerning the Hyperboreans, &c. §3. His account of the Euxine Sea. §4. Erroneous estimate of its length. § 5. Exaggerated notion of the Palus Maeotis. § 6. Erroneous idea of the Tauric Penin- sula. §7. Great rivers of Scythia: the Ister. § 8. Its magnitude and equable flow. § 9. The Tyras. § 10. The Borysthenes. § 11. The Tana'is. §12. Extent of his knowledge of these rivers. §13. Other rivers: the Panticapes, the Hypacyris, and the Gerrhus, cannot be identified. §14. The Oarus, the Lycus, &c., equally obscure. § 15. Imperfect sources of information. § 16. Valuable ethnographical notices. H is Scythians a distinct people. § 17. Form and boundaries of Scythia. §18. Division of the Scythian tribes. §19. Nations surrounding Scythia. The Agathyrsi : the N euri. § 20. The Andro- phagi: the Melanchleeni. §21. The Budini and Geloni. §22. The Sauromatze. § 23. Nations towards the north-east beyond the Budini. 'l‘hyssagetae and Iurcae. The Argippaei. §24:. Fabulous nations to the north. The Issedones. §25. The Arimaspians. Abundance of gold. §26. N o mention of the Volga. §27. The Caspian regarded by Herodotus as surrounded on all sides by land .. .. 172 SECTION 3.-—Empedition of Darius into Scythia. § 1. Geographical details of the expedition accord well with the other state- ments of Herodotus. § 2. Narrative as given by Herodotus. § 3. Its (lifliculties: cannot be received as historically correct; but presents few geographical difliculties .. .. .. .. .. 202 PAGE NOTE A. The Actae of Herodotus . . .. 207 ,, B. The Cimmerians .. .. .. 208 ,, C. The Greek Stadium .. .. .. 209 ,, D. Navigation of the Borysthenes . . 211 ,, E. Rivers of Scythia .. .. .. 212 ,, F. The river Oarus .. .. .. .. 213 ,, G. Limits of Scythia .. .. 214 ,, H. Ethnographical relations of the Scythians of Herodotus .. .. 215 ,, I. Expedition of Darius .. .. .. 217 VOL. I. b xviii CONTENTS or VOL. I. _ CHAPTER v11. GEOGRAPHY OF HEBODOTUS ; ASIA. SECTION 1.— General Views. § 1. Limits of his knowledge of Asia nearly confined to the Persian Empire. §2. Imperfect acquaintance with Arabia. §3. The Erythreean Sea and Arabian Gulf. N 0 knowledge of the Persian Gulf. § 4. Ignorance of the countries north of the Persian Empire. §5. Confused notions of the Araxes. § 6. The Massagetae; their abundance of gold. §7. Account of India: his knowledge confined to the countries on the Indus. §8. Voyage of Scylax of Caryanda. §9. The Pactyans: their mode of procuring gold. Indian ants. §10. Probable locality and origin of the fable. § 11. Scanty information concerning India in other respects .. .. .. .. .. .. Page 218 SECTION 2.—Persian Empire. Satmpies. § 1. His catalogue of the Satrapies derived from an authentic source. § 2. Erroneous geographical notion of Asia Minor. § 3. Imperfect ideas of the provinces of Upper Asia. No knowledge of the Taurus as a moun- tain system. §4. His enumeration of the Satrapies properly ethno- graphical, rather than geographical. § 5. Satrapies of Asia Minor: the first, second and third, including Cappadocia. § 6. The fourth, Cilicia: the fifth, Syria: the sixth, Egypt and the Cyrenai'ca. § 7. The seventh Satrapy—the Gandarians, 860. §8. The eighth, Susiana: the ninth, Babylonia and Assyria. §9. The tenth, Media: the eleventh, the Caspians, dzc. §10. The twelfth, Bactria: the thirteenth, Armenia and adjacent tribes. § 11. The fourteenth, the Sagartians, Zarangians, &c. §12. The fifteenth, the Sacae and Caspians. § 13. The six- teenth, the Parthians, Chorasmians, Sogdians, and Arians. § 14. The seventeenth, the Paricanians and Ethiopians. § 15. The eighteenth, the Matienians, Saspirians, &c.: the nineteenth, the Moschi, Macrones, &c.: the twentieth, India. §16. Character of this list: a statistical document, without any attempt at geographical arrangement. §17. Curious account in Herodotus of a plain surrounded by mountains and sending forth five rivers. Impossible to identify .. .. 231 SECTION 3.—Royal Road to Susa. § 1. Account of the line of route from Sardis to Susa, introduced as an inci- dental digression. §2. Its details and distances. §3. Its valuable character. Difliculties in detail. § 4. Followed the upper road from Sardis to the Tigris; not that taken by Cyrus and Alexander 249 PAGE NOTE A. Indian Tribute .. .. .. .. 255 ,, B. Scylax of Caryanda .. .. .. “256 coNTENTs or VOL. I. xix PAGE NOTE C. Caspatyrus . . .. .. .. 256 ,, I). Indian ants .. .. .. .. 257 ,, E. Ecbatana . . . . . . . . 258 ,, F. Royal road to Susa .. .. .. 259 ,, G. Uncertainty of measurement .. .. 261 CHAPTER VIII. GEOGRAPHY or HERODOTUS : AFRICA. SECTION 1.— General knowledge of the Continent. §1. Character of his acquaintance with African shores of the Mediterranean. His knowledge of the interior derived through Egypt and Cyl‘ene. §2. His account of the Upper Nile as far as Mero'é. §3. Mero'e. The Automoli. § 4. Supposed the Nile above Mero'é to flow from the West. § 5. Agreement of this view with his account of the Macrobian Ethiopians. § 6. Sources of the Nile unknown : theory of Herodotus. § 7. Narrative of the expedition of the Nasamonians into the interior. Their discovery of a great river, probably the J oliba or Quorra. § 8. The Maerobian Ethiopians: expedition of Cambyses against them. §9. Erroneously placed on the Nile by Strabo. §10. Division of Northern Africa into three zones. § 11. Line of inhabited spots from distance to distance in a direction from east to west. § 12. The Oasis of Ammon. §13. Augila. The Garamantes. §14. The Atarantes and Atlantes. §_15. Desert character of the interior. §16. The Troglodyte Ethiopians. §l7. Libyan tribes adjoining the Cyrena'ica, and along the shores of the Mediterranean. § 18. The N asamones and Psylli. The Lotophagi: the Machlyans: the island of Cyraunis. § 19. Tales of fabulous races rejected by Herodotus. § 20. The Syrtis and the Lake Tritonis. § 21. Account of dumb commerce carried on by Carthaginian merchants on the Atlantic coast. Cape Soloeis. § 22. Foundation of belief that Africa was surrounded by the Ocean. Page 262 ’ SECTION 2.-—- Oircamnavigation of Africa by Necho. §1. Narrative as reported by Herodotus. §2. Differences of opinion con- cerning it in ancient and modern times. §3. Arguments of Major Rennell in its favour. §4. Primafacie improbability of the story: absence of all details. § 5. Argument from the sun being on the right hand not conclusive. Other difliculties. §6. Subsequent neglect of the voyage and discovery paralleled by similar cases in modern times. §7. The voyage cannot be disproved, but must be regarded as ex— tremely improbable. §8. Other statements of the Carthaginians. §9. Unsuccessfpl attempt of Sataspes .. .. .. 289 ‘i b 2 XX CONTENTS OF VOL. I. PAGE NOTE A. Use of camels .. .. . . .. 299 ,, B. Distances on the Nile .. .. .. ib. ,, C. The Dodecaschcenus .. .. .. 301 ,, D. The Automoli .. .. .. .. 302 ,, E. Distances on the Upper Nile . . .. z'b. ,, F. Course of the Nile from west to east .. 303 ,, G. Supposed sources of the Nile .. .. 304 ,, H. Expedition of the Nasamonians .. 305 ,, I. Difficulty of communication with interior 307 ,, K. Erroneous position of Carthage .. 308 ,, L. The Oases . . .. . . .. ib. ,, M. Symmetrical arrangement of Oases .. 309 ,, N. Confusion of Great Oasis with that of Ammon .. .. .. .. ib. ,, O. The Atlantes of Herodotus .. .. 310 ,, P. The Lotophagi .. .. .. .. 311 ,, Q. The river Cinyps .. .. .. 312 ,, R. The Syrtis .. .. .. .. 313 ,, S. The Lake Tritonis .. .. .. 314 ,, T. The island of Cyraunis .. .. .. 316 ,, V. Cireumnavigation of Africa. Opinions of modern writers .. .. 317 C H A P T E R I X. VOYAGE OF HANNO. § 1. Narrative of the voyage transmitted to us. Its first object was colo- nization. Progress as far as Cerne. § 2. First voyage from Cerne to the South. §3. Second voyage. Proceeds as far as the Southern Horn. The gorillas. §4. Authentic character of narrative. Difl'erence of opinion as to extent of voyage. § 5. Position of Cerne. § 6. First voyage reached the Senegal. § 7. Second voyage. Cape Verde. Theon Ochema. Southern Horn. §8. Explanation of fires. § 9. The gorillas really apes. § 10. Erroneous notions concerning the voyage of Hanno found in later writers. § 11. Difficulties in regard to earlier part of voyage. § 12. Carthaginian colonies soon abandoned. Page 318 PAGE NOTE A. Editions of the Periplus .. .. 332 ,, B. Position of Cerne .. .. .. 333 C. Views of Heeren .. . . .. 334 ,, D. Notice of Cerne in Scylax . . .. e'b. ,, E. The river Chremetcs .. . . .. 335 CONTENTS OF VOL. I. xxi C H A P T E R X. wnrrnns AFTER HERODOTUS. SECTION 1.-—Otesz'as. § 1. Progress of geographical knowledge from Herodotus to Alexander imper- fectly known. Its limits little extended. §2. Thucydides. His work throws little light on geography. §3. Antiochus of Syracuse. §4.. Ctesias: his life and works. The Persica. §5. The Indica; full of fables and absurdities. § 6. Marvellous animals. {57. Precious stones. Page 336 SECTION 2.——T/ze Anabasz's of Xenophon. § 1. Historical narrative by Xenophon of the march of Cyrus and the retreat of the Ten Thousand: its charm. § 2. His estimates of distances. § 3. March of Cyrus through Phrygia to Cilicia.- §4. He crosses the Euphrates at Thapsacus. §5. Advance from thence to Babylonia. §6. Battle of Cunaxa and subsequent movements of the Greeks. §7. Their retreat as far as Opis. §8. They quit the Tigris: their march through the Carduchians into Armenia. §9. Their line of route traced thus far. § 10. Difficulty in following it to the Euxine. Steps of the progress according to Xenophon. § 11. Cannot be identi- fied in detail. §12. Independent tribes on the Euxine. §13. Inter- esting notices by Xenophon of the nations through which he passed. 342 PAGE NOTE A. Composition of the Anabasis .. .. 359 ,, B. Computation of distances in Parasangs z'b. ,, C. Rate of marching .. .. .. 361 ,, D. Route of Cyrus through Phrygia .. 363 ,, E. Passes between Cilicia and Syria .. 364 ,, F. March from the Gulf of Issus to the Euphrates .. .. .. .. 365 ,, - G. Thapsacus . . . . . . . . ib. ,, H. Position of Pylaa .. .. .. 366 ,, I. Site of Cunaxa .. .. .. .. 369 ,, K. Sittace .. .. .. .. .. 370 ,, L. The Median Wall .. .. .. 370 ,, M. March along the Tigris .. .. .. 372 ,, N. Ruined cities of Assyria .. .. 374 ,, O. Sources of the Tigris .. .. .. ib. ,, P. Descent through Armenia to Tiebizond 375 xxii CONTENTS or VOL. I. C H A P T E R X I. WRITERS AFTER XENOPHON. SECTION 1.——Eplzoras. The Periplus of Scylax. § 1. Ephorus: his historical writings; contained a general geographical review. §2. His general views. §3. His account of Scythia. §4. Of Asia Minor. § 5. Of Africa. § 6. Of the Nile. §7. Theopompus: his work of comparatively little interest to geography. Mentioned the capture of Home by the Gauls. § 8. The Periplus ascribed to Scylax : its character. §9. Its date: erroneously ascribed to Scylax of Car- yanda. §10. Its plan and arrangement: confined to the Mediterranean Sea: imperfect knowledge of some parts. § 11. Detailed account of the Adriatic: supposed it to receive an arm of the Ister. § 12. Full account of the Greek colonies on the Euxine. §13. The coasts of Asia Minor and Africa. § 14. Notices of the Atlantic coast of Africa : Cape Soloeis and Cerne. §15. N 0 account of Indian Ocean. §16. Distances generally given in days’ voyages. § 17. List of islands. Page 379 SECTION 2.—-Aristotle. § 1. Aristotle has left no regular work on geography; but notices in his works on the Heavens and the Meteorologica. § 2. His cosmical views: derived from Eudoxus of Cnidus. §3. His views concerning the Habitable World. § 4. Speculations connected with physical geo- graphy. §5. Erroneous ideas concerning the great rivers of Asia. §6. And of the West of Europe. Imperfect knowledge of Africa. § 7. Described the Caspian Sea as isolated. § 8. The Atlantis of Plato. 395 PAGE NOTE A. Age of Scylax .. .. . . .. 404 B. Order of islands in the Mediterranean 406 ,1 CHAPTER XII. EXPEDITION or ALEXANDER. SECTION 1.— Campaigns in Central and Western Asia. § 1. Alexander’s expedition into Asia forms an era in ancient geography. It first opened up to the Greeks a real knowledge of the interior of Asia. § 2. His campaign in Europe against the Triballi. He crosses the Danube. § 3. Attacks the Illyrians: recalled to Thebes. §4. He crosses the Hellespont: battle of the Granicus. His operations in Asia Minor. § 5. Advances through Phrygia to Gordium. § 6. March by Ancyra CONTENTS OF voL. I. xxiii and through the Cilician gates to Tarsus. Battle of Issus. §7. Occupies Syria and Egypt. His expedition to the temple of Ammon. Foundation of Alexandria. §8. His advance into Asia: crosses the Euphrates and the Tigris: battle of Arbela. §9. Occupies Babylon and Susa: advances to Persepolis. § 10. Pursues Darius from Ecbatana through the Pylee Caspiee: death of Darius. § 11. Halt at Hecatompylus: descent into Hyrcania. § 12. Subsequent campaigns lay through regions previously unknown. § 13. Conditions of marches determined by nature of the country: imperfection of itineraries. § 14. He pursues Bessus, but turns 03 into Aria, and thence into Drangiana: advances through Arachosia to the foot of the Hindoo Koosh. Founds Alexandria ad Caucasum." § 15. Investigation of route. § 16. Site of Alexandria. § 17. Crosses the Hindoo Koosh into Bactria. §.18. Operations in Bactria and Sogdiana: advance to the Iaxartes. § 19. Impossibility of following military movements in detail. §20. Foundation of cities: general character of the country. §21. Improved knowledge of the rivers Oxus and Iaxartes: but the latter confused with the Tanais. § 22. Both flowed into the Caspian. § 23. Hearsay information concerning other tribes. § 24. The Scythians beyond the Iaxartes .. .. .. .. Page 407 SECTION 2.—Mtli'tary Operations in India. § 1. Alexander recrosses the Hindoo Koosh: engages in the reduction of the mountain tribes: the rock Aornus. § 2. Impossibility of following his movements in detail. § 3. Or determining the position of the several tribes. § 4. He crosses the Indus and the Hydaspes: defeats Porus and advances to the Hyphasis. § 5. The point which he finally reached cannot be determined. § 6. Difficulty in following his operations in the Punjab. § 7. Returns to the l-lydaspes, and descends that river and the Indus to the sea. § 8. Progress of voyage down the river. § 9. Exaggerated account of the Indus: its supposed sources. §10. Its Delta: site of I’attala. § 11. Difficulty of identifying the tribes on the Indus : the Malli, Oxydracae, &c. 12. Alexander designed the permanent conquest of the regions on the Indus. § 13. Accounts of the wealth and populousness of India. .. .. .. .. .. .. .. 438 SECTION 3.—Return to Babylon. § 1. Return of Alexander to the \Vest through Gedrosia. § 2. Great difliculties and sufferings on the march. § 3. His progress through Carmania; rejoined by N earchus and Craterus. § 4. Returns to Susa : visits the Persian Gulf and Ecbatana. § 5. Preparations for exploring the Caspian Sea. § 6. Returns to Babylon: receives embassies from many nations, perhaps from the Romans. § 7. Prepares a fleet for the xxiv CONTENTS OF VOL. I. conquest of Arabia : sends out expeditions to explore the Persian Gulf: his death. § 8. His projects. NOTE A. ,2 27 ,7 was Qzgraefiaaewew M m. N n. O o. P p. Q q- R 1'. SS Gordium .. .. .. Pyla: Cilieiaa .. .. .. Mountain Passes near Issus .. March to the Oracle of Ammon The Oasis of Ammon Thapsacus . . . . . . . . Battle of Arbela .. .. . Retreat of Darius to Ecbatana .. Passes between Susa and Persepolis The Pylee Caspiae Heeatompylus . . . . . . Zadraearta . . . . . . Estimated distances . . . . Supposed measurements of Alexander’s route .. .. . . Rate of travelling on dromedaries The Indian Caucasus .. .. Artacoana and Alexandria in Ariis .. Routes from Herat into Bactria .. Prophthasia . . . . . . . . Indian tribes west of the Indus Climate of Araehosia .. .. .. Site of Alexandria ad Caucasum Passes of the Hindoo Koosh .. Comparative value of the historians of Alexander .. .. .. .. Legends concerning Bacchus and Her- cules .. .. .. The rock Aornus .. .. Peucelaotis . . . . . . . . Taxila .. Passage of the Hydaspes Geography of the Punjab Altars on the Hyphasis .. Sangala and the Catheei . . .. Boats on the Indus .. .. .. Descent of the Indus .. .. Confluence of the Hydaspes with the Acesines . . . . . . Course of the Hyphasis .. The Indus identified with the Nile Width of the Indus . . .. .. Source of the Indus . .. Delta of the Indus .. .. .. Site of Pattala .. . . .. Cities of the Malli .. .. T5 The Oxydracaa .. .. .. .. § 9. Cities founded by him PAGE 466 467 z'b. 469 470 471 472 475 2'6. 477 479 480 481 2'0. 483 484 485 486 488 ib. 489 490 492 493 496 2'6. 498 499 ib. 500 503 505 506 507 508 509 2'6. 510 511 512 513 514 515 Page 454 CONTENTS OF VOL. I. XXV NOTE U u. Changes in the course of the Indus .. 516 V v. March through Gedrosia .. .. 518 X x. Route through Gedrosia and Carmania 519 ,, Y y. March of Craterus .. .. .. 521 ,, Z z. Embassy of the Romans to Alexander 51.2 ,, A A. The Pallacopas .. .. .. .. 524 CHAPTER XIII. VOYAGE or NEARGHUS. § 1. Authentic character of the narrative. Modern writers on the subject. § 2. No statement of point of departure. §3. Port of Alexander. Voyage from thence to the Arabis. § 4. Coast of the Oritze. Cocala, Malana. §5. Coast of the Ichthyophagi: difliculties and privations. §6. Geographical details. § 7. More rapid progress as far as Badis. §8. Battle with whales. § 9. Perplcxing astronomical statement. § 10. Voyage along the coast of Carmania and the Persian Gulf. Meeting with Alexander. §11. Voyage continued along the coast of Persia. § 12. Duration of voyage: its character: its importance . . . . . ° ° ' ' ' I 0 o o PAGE NOTE Comparison with Pliny 542 A. .. B Estimate of distances by sea. Supposed difference of stadia .. .. 544 C. Port of Alexander .. . . .. 546 D The island of Ashtola .. .. .. 547 E ,, Supposed astronomical changes .. 548 ,, F. The island of Ormuz .. .. .. 549 ,, G. Diridotis . . . . . . . . 550 CHAPTER XIV. SUCCESSORS OF ALEXANDER. SECTION L-lllegast/zenes. Increased knowledge of India. §1. Establishment of Seleucus in the dominion of Upper Asia. New division of satrapies. §2. Renewed intercourse with India. Sandra- cottus. His friendly relations with Seleucus. §3. Megasthencs, His work the chief authority of the ancients concerning India. §4. Great Indian monarchy on the Ganges. § 5. Journey of Megasthcnes ; the royal road. §6. His account of the Ganges: and of l’alibotlira. § 7. Other nations of India: constitution of monarchy. §8. (jastcb, xxvi coNTENTs or VOL. I. § 9. Philosophers. § 10. Natural productions. § 11. Fabulous tales. § 12. Geographical information. § 13. First accounts of Tapro- bane. § 14. Daimachus. Notices of India by Patrocles Page 552 SECTION 2.--Bactrian Provinces. Patrocles. § 1. Revolt of Bactria and Sogdiana: establishment of a Greek monarchy there. §2. Very little information concerning it. No extension of geographical knowledge. §3. Wealth and power for a short time. §4. Patrocles: was governor of the provinces from India to the Caspian : asserted the Caspian to be an inlet of the Ocean. His authority on this point universally received .. .. .. .. 569 CHAPTER XV. THE PTOLEMIES IN EGYPT. SECTION l.—The Ptolemies in Egypt. The Red Sea. The Nile. § 1. Extension of geographical knowledge arising from foundation of Alex- andria, and of a Greek monarchy in Egypt. § 2. Extensive trade in the Red Sea, and foundation of cities to promote it. Arsino'e, Berenice, Myos Hormus. § 3. Settlements further south. Ptolema'is Epitheras. §4. Land of Cinnamon and the Southern Horn. §5. Trade with India: not direct, but through the Sabznans. §6. Account of these by Agatharchides. Arabia Felix. § 7. Progress of knowledge of the Nile. Memo. The afiluents of the Nile. §8. The two arms: the Blue and White Nile: the Sembritaz: the Cataracts. §9. No ex- tension by conquests in Ethiopia or in the interior of Libya. § 10. 'l‘imosthenes: his treatise on Ports: one of the chief authorities of ‘ Eratosthenes. His division of the winds .. .. 576 SECTIoN 2.——Progress of Discovery in Western Europe. Pytlzeas. Timoeas. § 1. Extension of knowledge of Western Europe: chiefly owing to Pytheas. His date. §2. His account of Britain and Thule: received by Era- tosthcnes, but rejected by Polybius and Strabo. § 3. Examination of his statements with regard to Gaul and Spain. §4. And to Britain and Thule. § 5. His supposed voyage to the Tana'is. His account of the origin of amber. §6. Other particulars concerning northern lands, founded on truth. His astronomical observations: latitude assigned to Thule. § 7. No mention of Cassiterides. §8. Timaius; CONTENTS OF voL. I. xxvii his historical work. Geographical notices contained in it. The island of Mictis. § 9. Theophrastus. The treatise “ De Mirabilibus ” ascribed to Aristotle, but belongs to third century 13.0. § 10. Curious notices contained in it. First notice of the Fortunate Islands. § 11. General extension of trade and intercourse .. .. .. Page 589 race NOTE A. Myos Hormus .. .. .. . . 607 ,, B. African elephants .. .. .. ib. ,, C. Land of Cinnamon .. .. . . 608 ,, D Monument of Adulis . . .. . . 609 ,, E. The winds as known to the ancients . . 610 ,, F. Dimensions of Britain . . .. .. 612 ,, G. Sir G. Lewis on credibility of Pytheas z'b. ,, ll. Astronomical phenomena at Thule . . 613 CHAPTER XVI. ERATOSTHENES. §1. Eratosthenes the parent of scientific geography. His date and life. § 2. Dictearchus his predecessor: his works. §3. His attempts to measure the elevation of mountains. § 4. State of geographical science in the time of Eratosthenes. He undertakes to reform the map of the world. § 5. His ideas of the position and figure of the earth. § 6. His attempt to determine its circumference. § 7. Defects of his method, and sources of error. § 8. Result after all a near approximation, He reckoned 700 stadia to a degree. § 9. His conception of the Inhabited \Vorld. § 10. His attempts to determine its limits. § 11. Draws one main parallel from the Sacred Promontory to the Indian Ocean. § 12. Points through which it passed in the Mediterranean. Its inaccuracies. § 13. Draws a meridian line through Alexandria and Rhodes at right angles with it. § 14. Errors in latitude. § 15. Still greater errors in longitude. §16. Mode of calculation of distances. His estimate of length of Mediterranean. §17. Other distances given in Mediterra- nean. §18. His imperfect knowledge of Western Europe. Followed Pytheas in regard to Thule and Britain. Ill acquainted with Scythia and Germany. §19. Determination of northern and southern limits of the world. § 20. Other distances along principal meridian. Excess in computation. §21. Continuation of parallel of latitude eastward through Asia. Its coincidence with range of Mount Taurus. § 22, Distances along this parallel. Length of habitable world. § 23, His ideas of the Eastern Ocean: and of the Caspian. § 24. Of India; and Taprobane. § 25. Of Arabia. § 26. Its physical character and tribes. xxviii CONTENTS OF VOL. I. § 27. Of the Red Sea. § 28. His accurate knowledge of the Nile and its tributaries. §29. Ignorant of the rest of Africa. §30. He gave few details of countries. § 31. His division into Sphragides. § 32. His views on Homeric geography. § 33. Imperfect ideas of physical geography. § 34. Speculations on changes in sea level. § 35. Notices of changes in surface of earth .. .. Page 615 PAGE NOTE A. Observations of latitude . . .. 661 ,, B. Gossellin’s theory .. ... 662 ,, C. Southern limit of habitable world .. 664 D. Distance from Alexandria to Rhodes .. 665 E. Promontory of Tamarus .. 666 LIST OF MAPS IN VOL. I. ____..,.__ PAGE 1. MAP To ILLUsTRATE THE WANDERINGs or ULYssEs .. to face 84 2. THE WoRLD ACCORDING To HECATzEUs .. ,, 148 3. THE WoRLD ACCORDING To HERoDoTUs .. .. .. ,, 172 4. SKETCH OF SCYTHIA ACCCRDING To HERoDoTUs .. .. 55;, 206 5. MAP To ILLUSTRATE THE VoYAGE 0F HANNo .. .. 330 6. MAP To ILLUsTRATE THE MARCHES 0F XENoPHoN AND 3 l ALEXANDER IN LowER AsIA .. .. .. .. ,, 358 7. MAP To ILLUSTRATE THE OPERATIoNs BEFORE AND AFTER THE BATTLE OF CUNAXA .. .. .. .. .. ,, 372 8. ALExANDER’s CAMPAIGNS IN AsIA .. .. .. .. ,, 464 9. ALExANDER’s CAMPAIGNS IN INDIA .. .. .. ,, 518 10. MAP on THE WORLD ACCORDING To ERATosTIIENEs .. ,, 660 Note—I am aware that I may appear liable to the charge of inconsistency, in presenting the reader with a map purporting to represent the voyage of Ulysses, when I have expressed in the body of my work my conviction that Homer had no such map present to his mind, and did not attempt to embody in any definite form his vague poetical conceptions of the wanderings of his here from land to land and from one mythical island to another. But such a representation has been so generally inserted in all treatises on ancient geography, and discussions of what is termed Homeric geography, that I thought some readers would complain of its absence; and at the same time it may assist them in following the narrative in the text, and recalling to their minds the order of sequence of the localities mentioned. It will at least bring clearly before their eyes the utter absence of all relation with the real localities in the Mediterranean, with which those mentioned by Homer were brought in connection by local tradition and the perverse in~ genuity of commentators. ‘ With regard to the two maps attempting to represent the idea of the world as XXX NOTE. formed by Hecatteus and Herodotus, I have not deviated from the customary mode of representing the two continents of Europe and Asia, and the Medi- terranean Sea, in accordance with their true position. The change in the direction of the Mediterranean introduced by Dr. C. Miiller does not appear to me to rest upon sufficient grounds to warrant its adoption. At the same time, it must be freely admitted that the map of the world according to Hecataeus is in great measure conjectural, for which reason I have confined myself to a rough and general outline. Even for that of Herodotus, though the historian has furnished us with many valuable materials, which have been fully discussed in the text, there remains much to be filled up by conjecture, and many difficulties that cannot be solved with any approach to certainty. HISTORY OF ANCIENT GEOGRAPHY. CHAPTER I. INTEODUCTOEE § 1. THE study of Geography, like that/ of Astronomy, is to a certain extent a necessary concomitant of civilization, but will be found to have been pursued more or less by different nations according to the circumstances of their position. N 0 people that have made even the first steps in the progress towards cultivation and enlightenment can have failed to direct their attention towards those heavenly bodies which so much in- fluence the condition of all the inhabitants ‘of this earth, which determine the changes of the seasons, and afford the only natural measures of time. Hence astronomy, in a rude and popular sense of the term, must have existed from the earliest ages among all nations that were not utterly. barbarous; and there seems no doubt that it was carried to a great extent and (attained a surprising degree of perfectioII among the earliest nations of antiquity, whose civilization we have any means of estimating. But while the natural situation and circumstances of the Egyptians, the Chaldaeans, and the Assyrians—their open plains and starry cloudless skies—were eminently calcu-. lated to direct their attention to the prosecution of this study, it was otherwise with geography. In all these cases their civilization was eminently local in its character. Derived in the first instance from the peculiar local circumstances in which it grew up, it was to a great extent confined by the VOL. I. B 2 HISTORY OF ANCIENT GEOGRAPHY. CIIAP. I. influence of those circumstances within the same original limits. Egypt especially retained through all the ages of its early greatness the same isolated character that China continued to hold down to a very recent period. The occa- sional outbreaks of ambitious monarchs, who for a time ex- tended their dominion over the neighbouring portions of Asia, produced no permanent result: and the Egyptian monarchy, when it first came in contact with the Greeks, was still limited to the valley of the Nile, as it had been eighteen ,centuries before. The same thing was the case, though to a less degree, with the monarchies of the Chaldaeans and the Assyrians, which successively ruled in the valley of the Euphrates/and the Tigris. The Assyrians, indeed, extended their dominion over a considerable part of Asia, and the adjoining island of Cyprus, but their peculiar civilization was confined to the district in which it arose, and a few isolated monuments alone attest their early‘ connection with any other nations or countries. Geography indeed may, in a certain restricted sense, be regarded as applying even to that limited knowledge of one’s own country and its natural features and boundaries, which every man of cultivated intelligence must inevitably possess. But it is only in proportion as it extends beyond these narrow limits, and comprises other countries, and a more considerable portion of the earth’s surface, that it approaches to the cha- racter of a science. There can be no doubt that the ancient Egyptians not only possessed what may be called an accurate geographical knowledge of Egypt; but that they had acquired at least general information concerning other countries that bordered on the Mediterranean, as well as those that adjoined them on the East. But how far they may have attained to anything like a definite geographical idea of any of those more distant lands, and their relative position, is a question that must be left to Egyptologers to determine; if indeed there exist the materials to enable us to arrive at any con- clusion on the subject. CHAP. I. INTRODUCTORY. 3 Commercial relations, which in almost all countries have been the precursors of geographical knowledge, could have produced but little effect of the kind in Egypt. Maritime com- merce was indeed almost wholly wanting: for the Egyptians appear to have entertained, in all ages, a deeply rooted aversion for the sea and all maritime pursuits; and though the pro- ductions of distant lands were brought to them by their neigh- bours the Phoenicians, and probably also by caravans across the deserts of Syria, this mode of traffic would hardly lead to any increased information concerning the countries from which they were derived. § 2. The earliest nation of which the literature has survived, that of the Jews, was almost as much out off from intercourse with other races, by the circumstances of its position, combined with its peculiar institutions, as were the Egyptians. The special character of their literature was also such as to afford little opportunity for any expositions of a scientific /nature: and while the earlier books of the Old Testament contain numerous geographical details concerning Palestine and the neighbouring districts, there is nothing that affords any idea of the general notions of the Jews on the subject, or the extent of their geographical horizon. The genealogy of the sons of Noah, found in the book of Genesis, is indeed an ethnological document of the highest interest, as embodying the earliest traditions concerning the relations and affinities of the different nations and races of men known to its author; but it conveys no information as to their geographical position; nor must it be hastily assumed that the writer had any definite ideas upon this subject. There can be little doubt, for instance, that the name of J avan was the same as is found in Greek mythology in the form Iaon or Ion, as the founder of the Ionian race. But it had certainly no local connection with the people situated on the coast of the .ZEgean, who were designated by that appellation in the time of Herodotus: and Whether it is intended to apply to the Hellenic race generally, or to the inhabitants of the peninsula now known as Asia Minor,— B 2 4 . HISTORY OF ANCIENT GEOGRAPHY. CIIAP. I. in which sense it appears to have been employed at a later period,-— we have no means of determining. But whatever may have been the nature and extent of the geographical knowledge possessed by the Egyptians or the Hebrews, it may be safely asserted that it had little, if any, influence upon that of the Greeks. The latter, indeed, be- lieved themselves to be indebted to the Egyptians for the elementary principles of geometry;1 they probably derived from the Babylonians the invention of the gnomon, or sun- dial, as well as the division of the day into twelve hours.2 But these scientific inventions were certainly not introduced into Greece until long after the period when our acquaintance with Greek literature gives us the means of judging for our- selves of their attainments in geography. The results of such an inquiry, as we shall presently see, are conclusive as to the fact that their geographical knowledge in the earliest ages was limited to the countries in their immediate neighbourhood, and to the group of nations that surrounded the ZEgean Sea. All beyond was vague and indefinite: derived from hearsay reports, imperfectly understood, or mixed up with mytholo- gical fancies and fables of poetical origin. §3. There was, however, one people which exercised con- siderable influence upon Hellenic culture and civilization, that calls for more especial notice in this place. The Phoenicians, unlike the Hebrews and the Egyptians, were essentially a commercial and seafaring people; and the earliest notices of them that We find, both in the Hebrew and Greek literature, concur in representing them as skilful and daring navigators, capable of conducting long voyages, and bringing back the productions of distant countries to exchange with their more sedentary neighbours. Unfortunately their native literature has utterly perished, and their early history is a blank. 1 Herodotus, ii. 109. to Anaximander, who lived in the 6th 2 This is expressly stated by Hero- century, B.C. Probably he was in dotus (1.0.). Other writers, however, reality the first to introduce it into ascribed the invention of the gnomon Greece. CHAP. I. INTRODUCTORY. 5 Whether or not there be any foundation for the tradition recorded by Herodotus, that the original abode of the Phoeni- cians was on the Erythrzean Sea, from whence they migrated to the tract on the shores of the Mediterranean which they inhabited in historical times, “ and immediately betook them- selves to long voyages, freighting their ships with the wares of Egypt and Assyria;”3 it throws no light upon the origin of this peculiar tendency, which so remarkably distinguished them from all other Semitic nations. To a certain extent, indeed, its causes are not far to seek. Confined to a narrow sea-board, and excluded from all extension of their territory towards the interior by ranges of mountains, which at the same time afforded them abundant materials for shipbuilding, it was but natural that they should devote themselves to maritime pursuits: and the similar instances of Amalfi, Genoa, and Venice, in the middle ages, show how easily even a scanty population, beginning with very limited re- sources, but devoting all their energies to maritime commerce, may attain to a marked superiority over all their rivals. But the extent of that commerce and the length of the distant voyages which we find them undertaking at this early period, as compared with what were customary among the Greeks and other ancient nations, even in a more advanced condition, undoubtedly present a perplexing problem, which we have no means of solving. The earliest notices which we find of them in the Hebrew literature represent them as being, as early as the time of Solomon (about 13.0. 1000), already familiar with the voyage to Tarshish, by which there seems no doubt that we are to understand the region in the south of Spain known to the Greeks in the early ages as Tartessus.4 3 Herodotus i. 1. (The same state- ment is repeated in vii. 89.) This tra- dition has been adopted as authentic, or as resting upon a basis of truth by several modern writers, including Movers (Die l’ho'nizie'r, book i. ch. ii.), while others, among whom is Mr. Kenriek (Phoenicia, p. 52). reject it as altogether unworthy of credit. 4 This subject is fully investigated by Mr. Twisleton, in his article TARSIIISH, in Dr. Smith’s Dictionary of Biblical Antiquities, vol. iii. 6 HISTORY OF ANCIENT GEOGRAPHY. CHAP, I. It is entirely in accordance with this that one of the earliest of their colonial settlements, of which the date can be fixed with any approach to certainty, was that of Gadeira, or Gades, in the district thus designated, the foundation of which is ascribed to a period as early as B.O. 1100. It is singular that, according to the traditions preserved to us, this distant colony preceded any of those in the Mediterranean, where Utica, on the coast of Africa, was established a few years later; while Carthage, which was destined to attain to so proud a pre- eminence over all other Phoenician colonies, was not founded till near three hundred years later.5 Unfortunately the dates thus transmitted to us rest upon very doubtful authority; but the testimony of the earliest extant Greek literature, in the form of the Homeric poems, is conclusive as to the general fact that the Phoenicians were at that remote period the principal traders and navigators in the ZEgean Sea and the neighbouring parts of the Mediterranean. They combined, as was generally the case with the earliest voyagers, the two objects of trade and piracy, especially for the kidnapping of slaves. But whatever may have been their evil practices in this respect, it seems clear that they possessed at this time the whole carrying trade of the seas with which the Greeks were familiar, and thus became the intermediaries through whom the arts and civilization of Egypt, Assyria, and Babylonia passed into Greece. It was through the same channel that the Greeks first became acquainted with various productions of more distant lands, such as ivory and frankin- cense, both of which were familiar to them in the days of Homer. Nor were the Phoenicians engaged only in the transport of 5 According to the anonymous author of the treatise On Wonders, falsely ascribed to Aristotle, but probably about a century later, Utica was founded 287 years before Carthage, a statement he professes to have derived from Tyrian records (Dc Miml).§ 134). Unfortunately the date of Carthage itself is not known with certainty, and the conclusion adopted by Movers (Die Pho'nizier, vol. ii. pt. ii. pp. 150-157) and Mr. Kenrick (Phoenicia, p. 145) of 13.0. 813 is admitted to rest only on a balance of evidence. The statement that Gadeira was settled a “ few years ” before Utica rests on the authority of Velleius Paterculus (i. 2). CIIAP. I. INTRODUCTORY. 7 foreign commodities. Among the productions of their own coasts was the shfiellg-gfish, from which they learnt at a very early period the art of extracting the purple. dye, for which they were famous in all ages, ‘and of which the Tyrians continued to be the chief manufacturers even under the Roman Empire. Glass also was an article extensively manufactured by them, sCThat its invention was by some authors ascribed to the Sidonians; and richly embroidered robes and garments are repeatedly mentioned in the Homeric poems as the work of Sidonian artists. On the other hand, we learn from the Hebrew Scriptures that Hiram, king of Tyre, was able to furnish to Solomon skilled artificers and artisans in almost every depart- ment of work requisite either for the construction or ornament of the Temple at J perusalem.“ In all the ornamental arts indeed the Phoenicians appear to have been at this period as much in advance of their neighbours the Jews, as they undoubtedly were of the Greeks. § 4. That this extensive commerce must have led to a wide expansion of the geographical horizon amongst the Phoenicians, as compared with their neighbours the Egyptians or the Hebrews, may be assumed as certain. But what were the real extent and limits of the knowledge thus acquired, and how far it was embodied in a distinct geographical form, are questions which, from the total loss of the Phoenician literature, we are wholly unable to answer. While we find in the earliest Greek records many vague and dimly-traced ideas as to the wonders of “the far west,” which are in‘ all probability derived from Phoenician sources, it is remarkable that no definite allusion is found to the countries in which that people had already esta- blished themselves, long beforcrthe date that can be plausibly assigned to the earliest remains of the Greek literature. N ot only is the name of Tartessus not found in the Homeric poems, but the whole of that cycle of myths, which was in later times connected with the name of the Greek deity Herakles, but 6 See 1 Kings vi; 2 Uhl'on- ii‘ 8 HISTORY OF ANCIENT GEOGRAPHY. CHAP. I. unquestionably belonged originally to the Tyrian god Melkarth, is conspicuously wanting. N o allusion is found to the island of Erytheia and the triple-headed Geryones with his herds of cattle, or to the golden apples of the Hesperides, or the Columns of Herakles. The name of Atlas is indeed found, but with nothing whatever to connect him with the mountain that subsequently bore his name, or with the local habitation assigned him on'the straits that led into the western ocean.7 Whatever ideas the Greeks may have derived from the Phoenicians concerning the western portions of the Mediter- ranean were of so vague and floating a character, that they can hardly be said to have assumed any geographical form; and they certainly afford us no clue to what may have been the conceptions entertained by the Phoenicians themselves. With regard to the East, on the contrary, they appear to have derived no ideas at all. Though the extent and character of the commerce carried on by the Phoenicians with the distant regions of the East is still a subject of much controversy among Oriental scholars, it may be assumed as certain that they received the commodities of India and other distant lands, either direct from the countries themselves, or more probably from an intermediate entrepo‘t in Arabia, as we shall find was the case with the Greeks at a much later period.8 In this respect indeed they appear to have been very much in the same position as the Venetians in the middle ages, and to have constituted the only channel through which the valuable pro- ducts of the East passed into the hands of the western nations. As we learn from the Hebrew writings that this was the case as early as the time of Solomon (13.0. 1000),9 it is certainly strange that not the slightest trace or allusion in any form is 7 This point will be more fully ex- amined in the next chapter. 8 On this subject I must refer my readers to the learned and able articles on TARSHISH and Drum, by Mr. Twisleton, in Dr. Smith’s Diet. of Bibl. Geogiz, where the whole question will be found fully investigated. 9 This appears from the well-known passage in the First Book of Kings, ch. xxii. The more elaborate descrip- tion of the commerce of Tyre in the 27th chapter of Ezekiel belongs of course to a much later period, about the beginning of the sixth century, B.C. CHAP. I. INTRODUCTORY. . 9 to be found in the Homeric poems to those regions of the far East, which were to the Greeks in all subsequent ages pre- eminently the land of marvel and of mystery. The only geographical notion—vague and floating as it was ~—which must have been derived by the early Greeks from this source, was that of the Ethiopians-J‘ burnt or black men ”-—a nation with whom thrgyptians had long been familiar, and of whom the knowledge probably passed from them through the Phoenicians to the Greeks. Through the Egyptians also must have come the fable of a race of Pygmies,‘/ situated apparently in the South of Africa, on the Ocean stream, and engaged in constant wars with the cranes that visited their country as immigrants from the N orth.1 § 5. It does not fall within the scope of the present work to enter into a detailed examination of the Phoenician commerce, even if there were more trustworthy materials than are really available for such an inquiry. But two of the articles which the Greeks unquestionably derived from them, and which they were supposed to import from some of the most distant regions of the known world, require a special mention in this place, from their connection with geographical questions that we shall find recurring at almost every stage of our future investi- gations. Tin and amber, two products of rare occurrence, and almost wholly unknown within the basin of the Mediterranean,2 were certainly both of them well known to the Greeks ; and in both cases there seems no room to doubt that they had been introduced to them by the Phoenicians. Tin, which was called by the Greeks C(tSS’iLQZOS, is repeatedly mentioned in the Homeric poems,3 and although its use as a separate metal would always be limited in extent, it was employed from a 1 The passages in the Homeric ‘, author, and the fact seems to have poems which refer to the Ethiopians certainly been unknown in ancient and the Pygmies will be considered in times. the next chapter. 3 Homer, Iliad, xi. 25, 34; xviii. 474, 2 Amber is found, though in small 564. Its ready fusibility is alluded to quantities, near Catania in Sicily; but by Hesiod, Theogon. v. 862. no notice Of this is found in any ancient IO HISTORY OF ANCIENT GEOGRAPHY. CIIAP. I. very early period as an alloy of copper, its mixture with which produced the compound now known as bronze, so valuable from its superior hardness, and which on that account was extensively used by all the nations of antiquity. But there exists much difference of opinion with regard to the quarter from whence this tin was procured by the Phoenicians in early ages. At a somewhat later period, but still long before the time of Herodotus, it is certain that the principal, if not the sole, source from which the tin used by the nations of the Mediterranean was supplied, was from certain islands in the \Vestern Ocean, known to the Greeks by name as the Cassiterides or Tinlslands, but the situation of which was w", unknown to Herodotus,4 and probably also to all his contem- poraries. Later information however leaves no doubt that the islands thus designated were the Scilly Islands, together with the adjacent peninsula of Cornwall, which was erroneously sup- posed to be a larger island of the same group, and from which in reality all the tin was procured.5 N o allusion to these islands is however found in the Homeric poems, nor in any author earlier than Herodotus, and it is probable that for centuries the Greeks continued to receive their supplies of this important metal from the Phoenicians without any inquiry as to the locality from which it was derived.“ That people were also from very early ages distinguished for their skill as workers in metals, and there can be no doubt that it was from them the Greeks/‘first, learnt theart of .making bilngaand probably in the first instance imported all articles composed of that metal ready manufactured;6 mainland naturally led to the suppo- 4 He says distinctly that he has no knowledge of the islands called Cassi- terides, from which tin was brought (iii. 115). But, he adds, it is certain that, as well as amber, it was brought from the extreme regions of the world. 5 No tin is found in the Scilly Islands, nor is it probable that it ever was produced there’; but the occurrence of this group of islands so near to the sition that the whole were connected together. Tin is also found in considerable quantities in Devonshire ; but from its inland position on the borders of Dart- moor, this is much less likely to have attracted the attention of traders than that of Cornwall. 6 Homer distinguishes Sidon by the CHAP. I. INTRODUCTORY. I I But however conclusive is the evidence that this was the quarter from which the supply of tin was derived in historical ages, many writers have found it difficult to believe that the Phoenicians could have carried on an extensive trade with such remote islands at the very early period when tin and bronze were certainly known both to them and to the Greeks. Hence some writers have supposed that they “derived their supplies in the first instance from Spain itself, where tin is also found, though in comparatively small quantities.7 The difficulty in the way of this theory appears to be, that tin is not found in the south of Spain, with which alone the Phoe- nicians had any direct trade; but in the interior, and princi- pally in the province of Galicia, from which it would be nearly as difficult to bring it to Gades, as from Cornwall or the adjacent islands.8 I‘ But, moreovCi', it is impossible to see how on this supposition the idea could have arisen of its being brought from “ the TinIslands ;” a notion which we find gene- rally established at the earliest period when we have any geo- graphical information on the subject at all. epithet of “ abounding in bronze” (1r0A15Xa7ucos, Odyss. xv. 425); and the ' great works in that metal presented by Hiram, king of Tyre, to the temple of Solomon bear the strongest testimony to their proficiency in the art. At a later period the Tyrrhenians (Etruscans) were also renowned for their skill as workers in bronze, and objects manufactured by them were to some extent imported into Greece; but no such intercourse can be carried back to the Homeric age. 7 Concerning the actual produc- tion of tin in Spain, see The Cassz'te- rides, by Dr. G. Smith (8V0. Lond. 1863), p. 45. In ancient times Posi- donius asserted that it was found in abundance among the Artabri, who occupied the north-western angle of the peninsula (the modern Galicia); and Pliny states that it was found in Lusitania and Gallaecia, which he erroneously regards as disprov-ing the popular notion that it was brought from certain islands in the ocean. (Plin. H. N., xxxiv. 16, § 156; Posi- donius ap. Strab. iii. 2, p. 147.) 8 Tin was also found in Lusitania, and is still met with in Portugal, but only in the interior of the country, where it would be little likely to attract the attention of the Phoenicians. The only other country in Europe where tin occurs in any quantity is Bohemia, the forests and mountains of which, as well as its inland position, rendered it one of the most inaccessible parts of Europe: so that it was un- known, till a late period, both to the Greeks and Romans, and even when they became acquainted with it they had no suspicion of its metallic wealth. The supposition that the Phoenician tin was brought from India is now generally discarded, and may be safely set aside as untenable. It is not found in any part of India nearer than the Malay peninsula. 12 . HISTORY OF ANCIENT GEOGRAPHY. CHAP. I. Whether or not their supplies were really derived in the first instance from Cornwall, or from some part of Spain, there is no doubt that Gades always continued to be the centre of the tin trade; it was from thence that the Phoenician colonists, and in later times the Carthaginians, made their adventurous voyages to the British Islands; and from this circumstance it came to be supposed that the Cassiterides were connected with Spain, a notion which became so firmly fixed in men’s minds that they were described by geographers in much later times as a group of islands in the Ocean, off the coast of Spain, without any hint of their connection either with Britain or Gaul. Strange as it may appear, it is thus that they are represented not only by Strabo, but even by Ptolemy, long after Britain itself was familiar to the Romans? This fact appears in itself conclusive against the view adopted by some modern writers that the British tin was in very early times brought overland through Gaul to the neigh- bourhood of Narbo and Massilia, and was imported from thence by the Phoenicians.1 That such was the case in later times, when the Greek republic of Massilia had gradually extended its commercial relations through Gaul to the Ocean, there can be no dispute; but to maintain that this was the route followed by the trade in the time of the Phoenicians is contrary to all the information we have on the subject. All accounts} point to Gades as the mart for tin,2 and the port from which the long voyage to the Tin Islands was undertaken; while we have no account of the Phoenicians trading with the southern coasts of 9 Strabo, iii. 5,'§,11, p. 175; Ptol. ii. 6, § 7 6. 1 This view was especially main- tained by Sir G. C. Lewis in his History of Ancient Astronomy, (pp. 450-455). The arguments on the other side are well brought together by Mr. Twisleton, in the article TARSHISH already cited, p. 14 39. 2 In the account of the Phoenician trade given by Ezekiel (chap. xxvii.) more than 100 years before the birth of Herodotus, Tarshish is spoken of as the mart for silver, iron, lead, and tin. The authorities followed by Avienus represent “ the Tartessians ” as trading from an early period with the Tin Islands, where probably the Phoenician colonists of Gades are meant (Avienus De Ora Maritima, v. 113); but little value can be attached to the expres- sions of so late and inaccurate a writer. CHAP. I. INTRODUCTORY. I 3 Gaul, or having any settlement near the mouths of the Rhone, which could serve as the emporium for so important a trade as that in question. §6. Another product which was certainly known to the Greeks in very early days, and the introduction of which is generally ascribed to the Phoenicians, jgamhfér; an object that was in the earliest days much sought after as an ornament, and that continued even in later times to attract an amount of attention wholly disproportioned to its estimation at the present day. Here again, while the substance is repeatedly mentioned in the Homeric poems,3 not the least clue is afforded to the locality from which it came. But before the time of Herodotus the tradition had come to be firmly established that it was brought from the banks of a great river called the Eridanus, which flowed into the northern ocean; and though the historian is disposed to reject this statement, as not resting upon ade- quate authority, he admits as an unquestionable fact that amber, like tin, was brought from the most distant regions of the earth.4 The fact was undoubtedly true ; for amber is found almost exclusively on the northern shores of Germany, and much the most extensively on those of the Baltic, though it occurs also in considerable quantities on the western coast of Sleswig and the adjacent islands.5 But it is much more difficult to believe that the Phoenicians actually made the voyage to the Baltic, and brought amber from thence, than that they visited Cornwall in quest of tin. In later times the 3 That the substance called in Homer “ electrum” ’ ‘Opxfipov Kai 1131/ iiAAwz/ 6poA01/ofiyueyou whim. And again in the same section he asks how it was possible that the poet 'rzfiu 1repl ’Ido'01/a o'vpfidv'rwv rcal 'r'hv ’Ap'yo‘v Kai 'roi‘is ’Ap'y01/az’n'as T6511 duoixwyovuéuwv wapdz wEio'u/ dw'filcoos fill; Demetrius 0f Scepsis denied altogether Homer’s knowledge of the voyage to the Phasis : dpxv’zv (P1101 #176’ ei‘o‘s'vaz 7971/ sis ‘Mimi! dvrofinpiau 1'05 ’Ioia'ozlos "O/mpov (1.0 ). He appears to have placed the abode of ZEetes on the shores of the Ocean, as was done by Hesiod and Mimnermus. 4 Aiiirao mikw, 7661. 'r’ duce'o; firs/Vol.0 arrive; Xpvo'écp xeia'rat éll 9aAo'quc9 duceavofi 'n'apd XeZAea', Z'v‘ 0§X€TO 66209 ’Ifio'wv. Mimnermus ap. Strab. i. 2, t} 40, p. 47. Mimnermus flourished about 13.0. 600. 5 Concerning these legendary attri- butions see the next chapter. The attempt to explain the legend of the golden fleece by a practice of the bar- barians in the neighbouring Caucasus to collect the gold dust in the torrent beds by placing fleeces of wool in them (Strabo, xi. 2, § 19, p. 499), belongs to a class of rationalizing explanations CHAP. II. VOYAGE OF THE ARGONAUTS. 2I to conceive how such a notion should have acquired currency. As soon as the Greeks began to extend their navigation into the Euxine, they would soon learn that at its eastern extre- mity lay the land of Colchis; and as this was the remotest region towards the east of which they had any knowledge, it would be a natural conclusion to assume that here must be that far-distant land of the rising sun, to which the celebrated voyage of the Argonauts had been directed. §2. \Vhen at a later period the Milesian and other Greek colonies gradually spread themselves along the shores of the Euxine, they would continually seek to identify themselves with the interest that had gathered around so celebrated a legend, and thus we find the tale of the Argonauts mixed up with a number of local traditions, or poetic fictions, which have no more real connection with the original story, than have the settlement of Antenor and ZEneas in Italy with the primitive Tale of Troy. At the same time the traditional tale adapted itself to the realities of the geography; and the voyage of the Argonauts from the Symplegades to Colchis, as we find it described in Apollonius Rhodius (in the third century 13.0.), has almost the accuracy of a geographical treatise. §3. But the case was very different with the return to Greece. For some reason, which it is not easy to determine, it seems to have been very early assumed that they could not, or at least did not, return by the same route which they had followed in the first instance;6 and the ingenuity of the poets and logographers, having a wide field afforded them by the prevailing vagueness of geographical not-ions, was exer- cised in devising various routes—all equally imaginary, and equally impossible, by which the ship Argo was supposed to of mythical legends of all others the most unsatisfactory, though highly popular with the Alexandrian critics and their successors. Strabo himself introduces the statement with a sus- picious Ael'ye'rat. ‘5 Mr. Grote suggests (History of Greece, vol. i. p. 314) that “it became necessary to devise another route for them on their return,” in order to account for the numerous local legends connected with the visit of these mari- time heroes to Libya and other places in the IVIediteI'r-anean, wholly apart from the Euxine. 22 HISTORY OF ANCIENT GEOGRAPHY. CHAP. II. have effected her return to Thessaly. The original idea seems to have been that followed by Hesiod (or by one of the poets whose works were extant under his name), that the Argonauts, after attaining the object of their voyage by possessing them- selves of the golden fleece, sailed up the Phasis, and thus passed into the Ocean stream, which was universally con- sidered as flowing round the whole world.7 Once embarked upon this circumfluent stream, it was not difficult to carry them wherever it was desired, and they were supposed to have followed it till they found themselves on the south coast of Libya, opposite to the Mediterranean. Here they were in- structed by Medea to quit the Ocean, and they carried the Argo “ over the desert surface of the land” for twelve days, until they launched it again at the mouth of the Lake Triton. This is the form in which we find the story told by the earliest extant poet who has dwelt upon it at any length, in the fourth Pythian ode of Pindar;8 but, unfortunately, the lyric cha- racter of the composition prevents it from presenting us with anything like a continuous narrative. Another version of the story represented them as sailing up the Tanai's, instead of the Phasis, and passing, by means of that river, the sources of which were still unknown, into the great northern Ocean, and thus sailing round till they re-entered the Mediterranean at its western extremity. Later writers, who were aware of the impossibility of this mode of proceeding, introduced an addition similar to that found in the other form of the legend, and described them as transporting the ship upon rollers from 7 'Ho'fodos 5% did otor§os ar’r'robs eio'rre- 1r7tevrce'vat Aé'yer, Schol. ad Apollon. Rhodium, iv. 284. In another pas- sage (ad v. 259) the same Scholiast associates Hesiod with Pindar and Antimachus as carrying the Argonauts through the Ocean to Libya, and thence over land to the Mediterranean. ‘Haio'o‘os 5% Ital Hiz/Bapos r311 l'Ivdtoi/l’nars Ital ’Av'rl’- ,uaxos s’x/ A6537 did 'roi) ’.Qxeauofi (pao'u/ éhfle'iz/ airroils sis AtBt'mu Kai Bao'rdo'au'ras 'r'bu ’Ap'yc‘u 6Z5 1''?) fpué'repm/ 1re'Aa'yos 'yeve'o'Qaz. 8 Pindar. Pyth. iv.—~According to the Scholiast on Apollonius (iv. 259), the historian Hecataaus introduced a varia- tion on this version, and represented them as ascending the Phasis to the Ocean, then following the Ocean to the Nile, and descending the latter river to the llIediterranean. This strange hypothesis was gravely con- , troverted by Artemidorus and Eratos- I thcnes. CHAP. II. VOYAGE OF THE ARGONAU’I‘S. 23 one navigable river to another, and thus reaching the outer sea.9 § 4. Apparently the latest form of the legend was that with which we are in modern times most familiar, in consequence of its having been adopted by Apollonius Ithodius in his well- known poem, but which had previously found little favour with the Greeks.1 According to this, the Argonauts, in order to elude the pursuit of the Oolchians, sailed across the Euxine to the mouth of the Ister (Danube), and ascended that river as far as the point where it divided into two branches or arms, one of which flowed into the Euxine, the other into the Adriatic or Ionian Sea. This strange geographical error was, as we shall hereafter see, widely prevalent among the ancient Greeks, even at a period when such a misconception would appear impossible, and was believed even by such writers as Eratosthenes and Aristotle. It would, therefore, be readily adopted by the rationalizing critics and poets of the Alex- andrian school; but it could obviously not have formed any part of the old legend, being an outgrowth, though an erro- neous one, of more advanced geographical knowledge. As if this absurdity had not been enough, Apollonius having thus brought his heroes into the Adriatic, must then conduct them up the Eridanus (which was in his time already identified with the Padus, the great river of Northern Italy), and from thence by a bifurcation similar to that assigned to the Ister, into the Rhodanus or Rhone, which they then descended to the Tyrrhenian or Sardinian Sea.2 The object of this strange 9 This was the story followed by the geographer Scymnus (ap. Schol. ad Apollon. iv. 284), as well as by the historian Timaeus (ap. Diodor. iv. 56). It is the same version, though in a very confused form, that was adopted by the author of the poem on the Argonautic expedition ascribed to Orpheus. 1 According to the Scholiast, the only writer before Apollonius who had adopted this route was one Timagetus, the author of a work on Harbours (wept Ame’vwv); whom he cites re- peatedly (Schol. ad Apollon. iv. 259, 306, 324, &c.); but his name is other- wise unknown, and we have no clue to the date at which he lived. He was, however, probably a late writer. The notion of the bifurcation of the Ister was unknown to Herodotus: and does not appear until a date long subse- quent. _ 2 Not content with this, the poet 24 HISTORY OF ANCIENT GEOGRAPHY. CHAP. II. addition to the legend was obviously to bring them to the dwelling of Circe, whose place of abode had been long regarded as fixed on the western coast of Italy, adjoining the Tyrrhenian Sea; while her name was so inseparably associated with the legend of the Argonauts that it was thought absolutely neces- sary to represent them as paying her a visit.‘ From thence they passed by the promontory of the Sirens, Scylla and Charybdis, and the Planctae or Moving Islands, on their way home; touching also at Phaeacia, as well as at Thera, Anaphe, and other points which were connected by local legends with the tale of the Argonauts. It is natural to suppose that these Homeric localities, with their accompanying fables, were merely introduced by the poets in imitation of the Odyssey, and formed no part of the original legend. But there seems much reason to believe that there were very early legends connected with the Argonauts in the west as well as the east; and it is remarkable that the only allusion to the voyage of the far-famed ship Argo that is found in Homer is in connection with the dangers of the rocks called Planctae, which were connected by the general consent of ancient writers with the western part of the Mediter- ranean, in common with Scylla and Charybdis, and the islands of ZEolus and Circe.3 It was precisely the conflicting character of these two sets of legends, and the necessity of attempting to reconcile them, that involved the logographers and poets in such hopeless confusion; and led to their adopting such strange geographical theories for that purpose. They had attained just suflicient geographical knowledge to see the tells us that there was a third arm, by Apollonius (iv. 858, 922). Strabo leading ofi‘ direct into the Ocean, which the Argonauts were on the point of following had they not been warned in time by Hera (iv. 625-642). 3 They are distinctly placed by Homer himself (Odyss. xii. 55-72) be- tween the rocks of the Sirens and those of Scylla and Charybdis: and the same view of their position is taken also tells ‘us that Homer invented the Planctee in the western sea in imitation of the Symplegades or Cyanean Rocks at the entrance of the Euxine. Later writers identified the two, and gave the name of Planctze to the Symple- gades; but this is wholly at variance with the Homeric conception. (See Nitzsch’s note on the passage.) CHAP. II. VOYAGE OF THE ARGONAUTS. 25 difliculties that arose, when they attempted to combine into one narrative stories originally quite unconnected with one another, and to give a definite form to what the earliest poets and their hearers were contented to leave wholly vague and unsubstantial. § 5. It would be a mere waste of time to attempt to extract from these different versions of the Argonautic legend, in the form which it ultimately assumed, any indications of the primitive geographical notions with which it was associated, for we are wholly unable to distinguish these from the almost equally confused and irrational views, which were still current among the Alexandrian poets. The earliest writers, so far as we know, who treated the subject at any considerable length were Eumelus of Corinth, and the author of the poem called Naupactica; the former of whom may be assigned to a period about 13.0. 761~744, while the latter may probably be referred to the same century.4 From the very scanty fragments of them that are preserved, it is pretty clear that the general framework of the fable had already at this period assumed the form with which we are familiar; but no geographical details are cited from either poet; and the scope of their works, which was in both cases a genealogical or mytho-historical one, ren- ders it extremely doubtful whether they troubled themselves to relate the voyage in such a manner as would have thrown any considerable light upon the subject, even if they had been preserved to us. rl‘he hypothesis of some modern writers that the poem called Argonautica, still extant under the assumed name of Orpheus, was really the production of Onomacritus, who flourished in the time of the Pisistratids, is wholly destitute of support from any ancient authority ; and the work in question may safely be assigned, on its own internal evidence, to a period not earlier than the second century of the Christian era.5 4 See Mure’s History of Greek Lite- however, assigns them to a period not mture, vol. ii. p. 261, pp. 447-450; less than acentury later. and Marckschefi'el, Hesiodz' Frag/manta, 5 See Note A (p. 28). Lips. 1840, pp. 218-223. Mr. Grote, . 26 HISTORY OF ANCIENT GEOGRAPHY. CHAP. II. Of course it is impossible to assert that there may not have been some voyage or naval expedition at a very remote period, ' which attained to such a celebrity as to become the nucleus around which crystallized so many local and poetical legends; and it is difficult to explain on any other hypothesis why the original legend should assume a form so totally difi‘erent from all others transmitted to us from the early ages. This is the more remarkable because the people to whom'the enterprise is uniformly ascribed by the poetical traditions, in the form in which alone we possess them, is that of the Minyans, a race inhabiting Boeotia in very early ages, whose power and wealth are attested both by the Homeric poems6 and by existing remains, but who were certainly not in historical times con- nected with maritime pursuits or naval enterprise. But it may be safely aifirmed that, if there ever was a voyage of the Argonauts which gave rise to the traditionary tale, it was of a comparatively very limited character; and that the idea of Colchis and the Phasis was not connected with it till long afterwards. It may also be regarded as certain that it was to a great extent interwoven with legends and traditions that arose after the great extension of Greek navigation and commerce in the Euxine, which did not take place till the seventh century B.C. But to suppose the original legend to have grown up in consequence of these'exploring voyages, and that the tale of the Argonauts is merely a mythical repre- sentation of the progress of Greek discovery in the Euxine,7 not only takes no account of its inseparable connection with 6 Homer, Iliad, ix. 381, where race towards the East about twelve cen- Orchomenus is cited as renowned for its wealth, in the same manner as the Egyptian Thebes. He elsewhere (ii. 511) attaches to it the distinctive epithet of “Minyan.” 7 This view has been adopted by Humboldt (Cosmos, vol. ii. p. 140, Engl. trans), and by Mr. Edward James in Smith’s Dictionary of Ancient Geo- graphy, art. Bosronus); but to speak of the pressing forward of the Hellenic turies before our era appears to me a mere anachronism. There is no evi- dence of any such tendency in this quarter until a period at least four centuries later; and the only ground for assigning it to so early a date is the wholly unsupported date affixed by the ordinary system of chronology to the Argonautic Expedition, considered as a historical fact. can». II. VOYAGE or THE ARGONAUTS. 27 the Minyans, but is at variance with the fact that we know the legend of such a voyage to have been already familiar to the Greeks in the age of the Homeric poems, long before either the Milesians or Megarians had penetrated into the Euxine. All that can be said is, that, as the legendary traditions of the Trojan \Var implied of necessity a state of things in which navigation had already become sufficiently familiar among the Greeks for them to transport a considerable army across the ZEgean to the shores of the Hellespont, so there existed another set of traditions, unquestionably also of early date, that pointed to some enterprise of a more distinctly naval character, of suffi- cient importance to be connected with the names of heroes and demigods, and to become in like manner the basis upon which was accumulated a mass of mythical fictions. 28 HISTORY OF ANCIENT GEOGRAPHY. CHAP. II. NOTE A. ARGONAUTICA OF ORPHEUS. I SHOULD have been content to leave the consideration of the supposed antiquity of the Argonautica, as was done by K. O. Muller and Mr. Grote, as a question that had been decided beyond appeal by the successive investigations of Schneider, Hermann, and Ukert: had it not been for its having been brought forward afresh by M. Vivien de St. Martin in his recent Work on the historical pro- gress of Geography.1 Admitting that the arguments of the German critics, derived from grammatical and metrical details, may be con— elusive against assigning an early date to the poem in its present shape, he still maintains that it may be merely a rifaccz'amento of an earlier work, and that the poem now extant is in substance the same as that of which he ascribes the composition to Onomacritus. Two arguments appear to me conclusive against this hypothesis: the one, that, as stated in the text, this supposed redaction by Onomacritus of a poem on the Argonautic voyage is a pure fiction : that is to say, a mere arbitrary hypothesis, assumed without a particle of evidence. There is some ancient authority, though very vague and indefinite, for Onomacritus having composed hymns in the name of Orpheus, or worked up previously existing poems of a religious character into a more definite shape ; and it is not improbable that the poems current under the name of Orpheus in the time of Aristophanes belonged to this class. But there is absolutely none for Onomacritus having handled the subject of the Argonautics, a poem of a totally different character.2 Nor, in the 1 Histoz're de la Geographic et des Décou'vertes Géog'raphiques. 8vo. Paris, 1875. 2 Suidas, indeed, mentions a certain Orpheus of Crotona as having written a poem, called Argonautica (s. v.) And this Orpheus is evidently the same who is mentioned by another grammarian as having been associated with Onomacritus in the task of re- vising the poems of Homer. (Schol. ad Plaut. cited by Ritschl Die Alex- andrim'sche Bz'bl'iothek, p. 4.) He therefore flourished in the time of the Pisistratids. It is singular that more use has not been made of this state- ment of Suidas (valeat quantum), by the advocates of the early date of the poem. NOTE A. ARGONAUTICA OF ORPHEUS. 29 second place, is there any mention of the existence of any such poem before the Alexandrian period, or indeed till long after ; and the existing Scholia on Apollonius, which are of unusual fulness and value, while repeatedly referring to the different versions of the tale found in different authors, never allude to the existence of a poem on the subject under the imposing name of Orpheus. This consideration alone appears to me conclusive against its being of older date than the late Alexandrian period. The internal evidence appears to me equally decisive. M. de St. Martin finds in it the primitive simplicity and didactic character of the earliest poets. To me it appears, in common with several distinguished critics, to have the jejune and prosaic tameness so characteristic of the declining Greek poetry of the second and third century after the Christian era. And this character is as strongly marked in the conception and mode of treatment of the subject as in the details of style and diction. For these last I must refer my readers to Hermann’s elaborate dissertation, appended to his edition of the Orphica; an excellent summary of the whole subject, from the critical point of view, is given by Bernhardy in his Gru-nclm'ss der Griechz'schen Literatur (2nd edition, Halle, 1856, vol. ii. pp. 847-353). From the geographical point of view it matters little whether the poem is to be ascribed to the Alexandrian or to the Christian period. In either case it is equally worthless, and unworthy of careful examination. But the evidence that it is not (as M. de St. Martin maintains) “ certainly anterior to Herodotus,” appears to me overwhelming. The confusion of the writer’s geography, which is regarded by M. de St. Martin as arising from his great antiquity, bears a striking resemblance to that found in several of the later geographers. It is not merely that he has erroneous ideas, even in regard to regions like the north coasts of the ZEgean; that he represents the Araxes, Thermodon, Phasis, and Tanais, as all having a common origin; and that in describing the voyage from the Maeotis to the Northern Ocean, he jumbles together the names of Scythian tribes derived from all kinds of sources, and enumerates the Geloni, Sauromatae, Gretae, and Arimaspians, among the nations dwelling around the Palus Maeotis, while he transfers the T-auri, noted for their human sacrifices, to the shores of the channel leading into the Northern Ocean. But he describes the Argonauts as passing through a narrow channel into the Ocean, “ which is called by the Hyperborean tribes the Cronian Sea and the Dead Sea.” 3O HISTORY OF ANCIENT GEOGRAPHY. _' ‘NOTE A. Both these names were familiar to the geographers and poets of later times ;3 but no trace of them is found before the Alexandrian period. Here they visit in succession the Macrobians, Cimmerians, and the land of Hermionia, where is the mouth of Acheron and the descent into the infernal regions : but they are especiallywarned to avoid the island of Ierne, in order to do which they by great exertions double the Sacred Cape, and after twelve days’ voyage reach the fir-clad island, sacred to Demeter, where the poet places the fable of the Rape of Persephone. Thence in three days they come to the island of Circe, after which they pass by the Columns of Hercules into the Sardinian Sea. Here we find mentioned the customary legends of the Sirens, Charybdis, &c., but mixed up with the names of the Latins, Ausonians, and Tyrrhenians, as inhabitants of its shores: and the mention of Lilybaeum in Sicily is associated with the burning Etna, and the fable of Enceladus. It is remarkable that the “ far-stretching Alps” (mvvfikees "AM-etc) are mentioned among the ranges of mountains—associated with the‘ Rhipeean mountains and the Calpian ridge—Pthat overshadowed the ' land of the Cimmerians, and helped to shut out from it the light of the sun. Absurd as is this statement, it shows a familiarity with the name of the Alps as a great mountain chain, though it was certainly unknown as such to the Greeks in the days of Herodotus. The mention of Ierne (or, as it is called in one passage, the Iernian Islands) is still more decisive. There is no evidence of any know- ledge of the British Islands among the Greeks before the time of Pytheas, while the name of Ierne (Ireland) is not mentioned till a considerably later period} Confused and extravagant as are the geographical notions con- tained in the above narrative, it does not appear to me possible to pronounce upon their evidence alone, that the poem cannot belong to the Alexandrian period instead of the Roman Empire. Its assignment to the later age must rest upon considerations of style and language, as well as upon the all-important fact that no allusion to its existence is found in any ancient author, or even in any of the scholiasts or grammarians down to a very late date. 3 As for instance to Dionysius Peri- early date of the poem, are fully given egetes (v. 31-35), whose description of by Ukert (Geographz'e der Grieehen and this northern ocean bears a close re- Ro'mer, vol. i. part ii.), to whose elabo- semblance to that of the Orphic poet. rate dissertation on the subject I must 4 The geographical arguments refer my readers. against the authenticity, or supposed (31) ’ CHAPTER III. HOMERIO GEOGRAPHY. SECTION 1.-——~Geneml Views. §1. YVE may now proceed to consider the notions of geo- graphy entertained by the Greeks, at the earliest period when they can be regarded as possessing any real geographical notions at all. In this investigation our principal-indeed, it may be said our sole—source of information is to be found in theHomeric Poems“. _ \Vhatever opinion may be entertained as “ to the date or mode of composition of the Iliad and Odyssey—~ whether we regard them as the work of one author or of several ——it is certain that they. represent to us the ideas and senti- ments of a very early age; and we shall find as we proceed in our inquiries this character of antiquity and primitive sim- plicity attested not less strongly by the nature and limits of the geographical knowledge which they display, than by their evidence as to the manners, arts, and institutions of the heroic ages which they describe.l In attempting to represent to ourselves, and to retrace as far as possible, the dim and vague outlines of the geographical picture that floated before the mind of the poet, and that may have been present more or less distinctly to those of his hearers, it might naturally have been thought that we should be materially assisted by the labours of the numerous com- mentators and interpreters who devoted themselves in ancient times to the elucidation of his ideas on this as well as on other subjects. But so far from this being the case, it'will be found 1 See Note A, p. 75. 32 HISTORY OF ANCIENT GEOGRAPHY. CHAP. III. that great part of the difficulties which surround the inquiry have arisen from these very commentaries, embodying, as they do, traditional interpretations of the Homeric legends, derived from a much earlier period, but of the origin or authority of which we have no knowledge whatever. There were not indeed wanting during the Alexandrian period some writers who had the boldness to asp/ail the conclusions thus generally received, and dispute the accuracy of the traditional attribu- tions ,2 but they appear to have found few followers, and the great authority of Strabol-a writer who in many other respects possessed a real insight into scientific geography—was unfor- tunately thrown into the opposite scale, and tended, to an extent that can hardly be overrated, to consolidate the fabric of the system which he found established in his time. Yet it will be found on examination that that system rests on no substantial foundation, and is in many instances opposed to the most simple and natural interpretation of the poet’s meaning. It is only quite in recent times that modern writers have shaken themselves clear of that blind reverence for the opinions of the ancients which led men to accept without hesitation the conclusions of Strabolor Plinyfand receive as established truths the interpretations that had been put upon the ideas of the poet by traditions of much later date, or by the attempts of rationalizing critics to bring them into accord- ance with the known facts of geography. The simple and vague ideas of Homer have been disfigured and disguised by the desire to reconcile them with the scientific knowledge of after ages, and to adapt them to a system of which neither the poet nor his contemporaries had the slightest conception.3 2 The most prominent among these and Homeric commentator, who was a was Eratosthencs, to whose opinions contemporary of Aristarcnus, was con~ on the subject we shall repeatedly have spicuous by his attempts at this kind to refer. His views were also adopted of interpretation (see Strabo, i. pp. by Apollodorus, and to some extent at 3, 5, 32, 8:0.‘ for which he is justly least by Demetrius of Scepsis. reproved by Gominus. (Elem. Astron. 3 Crates, the celebrated grammarian c. 13.) SE01‘. 1. HOMERIC GEOGRAPHY. 33 Hence in all inquiries into the Homeric geography, it is above all things necessary to divest ourselves of all these subsequent additions, and to confine our attention as entirely as possible to the words of the poet himself, and the conclusions that may be legitimately drawn from his own language.‘i § 2. There can be no doubt that Homer, in common with all his successors down to the time of Hecataeus, believed the earth to be a plane, of circular form, surrounded on all sides by the Ocean, which they conceived, not as a sea, but as a vast continuous stream, for ever flowing round and round the earth.5 The vault of heaven, which he terms “ brazen” or “ of iron,” epithets used to denote its solidity, he conceived as a solid concave surface, like the “ firmament ” of the Jews, of equal extent with the earth so as to rest on it on all sidesf“ But at the same time he seems to have supposed it to stand in need of some additional support, and therefore to rest upon tall pillars “.which keep the heaven and the earth asunder.”7 These pillars are in the charge of Atlas, but it is not stated that Atlas himself supported the heavens, according to the notion prevalent in later times, and adopted by almost all subsequent poets. Nor is there the slightest indication of the situation of these mighty pillars; or anything whatever to 4 It is one of the great merits of two I agree with it, and it was undoubtedly of the leading writers on this subject, I the idea entertained by the Greeks Volcker and Ukert, that they have ' adhered steadfastly to this system. Mr. Gladstone has adopted the same line: and I cordially assent to his remarks upon the course to be pursued in the investigation (Homer and the Homeric Age, vol. iii. pp. 262-265), widely as I difi’er from him in respect to the results. 5 See Note B, p. 75. 6 This is distinctly stated by Hesiod (Theogon. vv. 126-7). I‘a'ia. 8e’ 70!. 1rp¢Irrov ,uév é-yeiva'ro laov éavrfi, Oirpavbv do'repdeve’, 'two. [My 'n'epi. rrdv'ra Icamirrrol. There is no passage in Homer where the conception is stated with equal distinctness; but his expressions all VOL. I. down to a much later period. 7 . . - 5X64. 8e’ Te_r<(ova9 afrrbs Moucpds, a‘i. yato'w 76 Kai. ofrpavbv 01mph; é'xovo'w. Odyss. i. 64. That the sense given in the text is the true meaning of stars é'xovo'w, I feel no doubt, notwithstanding the different interpretation proposed by some of the commentators: and it seems most natural to translate 5X61 in the first line as “keeps ” or “ guards.” The idea that Atlas himself held and supported the columns seems to have arisen from a desire to explain the passage in accordance with the myth as generally received in later times. D f 34 HISTORY OF‘ ANCIENT GEOGRAPHY. CHAR’ III. ~ connect them with the mountain chain in the distant regions of the west, to which the name was applied by theGreeks of plater ages. The statement appears, so far as Homer is con- " "" jcerned, ‘as a purely mythological fiction, which has nothing to \do with geography'." - § 3. It is repeatedly stated in the Homeric poems that the sun ‘ rose out of the Ocean streamfand again sank into the same at its setting.8 How it was carried back to the point from which it was ‘to start afresh on its course it is probable that no one in his day ever troubled himself to inquirei It is certain that no trace is found of the absurd inventions by which later poets and meteorologers endeavoured to overcome this difficulty.9 The stars also are represented as following the same course and bathing every day in the waters of the Ocean.” There was however one exception. Even the rudest observers could hardly gaze on the beautiful starry skies of Greece without noticing that there was one conspicuous group of stars—the constellation of the Great Bear—which never set, but, in the words of Homer, “keeps turning round in the same place, jfland alone has no share in the baths of Ocean.”1 Besides the “Great Bear-which was already known by the two names of “ the Bear ” and“ the Waggon ”-—Homer mentions other con- 8 Iliad, vii. 421-423; viii. 485; but that its course was not visible, on xviii. 240; Odyssey, xix. 433, &c.' It was preceded by the Dawn (’Hc6s), which in like manner “ arose from the streams of Ocean, to give light to immortals and to men.” —— Iliad, xix. l. account of the elevation of the northern parts of the earth. (Arist. Meteorol. ~- ii. 1, § 16.) 9 It is diflicult to believe that the story of the Sun being carried back along the Ocean stream, reclining in a golden bowl, which is found not only in Mimnermus, but even in Stesichorus, and the logographer Pherecydes (ap. Athcnaeum, xi. p. 469-70), could ever have been regarded as anything more than a mythological fiction. But the earlier meteorologers, according to ,,.Aristot1e, got over the difl‘iculty by gsupposing that the sun travelled round Lin the night by the north to the east,. l 'Apm'ov 9', fiv Kai. firm-Earl é'n'ixhna-w KaAé- 01:01:’, i)?" ab'rofi o'Tpe'rbea-at, Kai. 1" 'Qpz'wva Soxefia, 011; 8’ (imwpé; e'an Aoe'rpiw ’Qxeavoio. Odyss. v. 273-275. The same three lines occur also in the Iliad (xviii. 487-489). It is to be observed that in the Odyssey (1.0.) Ulysses is represented as steering by watching the Bear (Le. the Great Bear). It would appear, therefore, that the refinement of steering by the Little Bear, or Pole Star, which is ascribed to the Phoenicians, was not yet familiar to the Greeks in the days of Homer, or at least was not yet known to the poet. SECT. 1. HOMERIC GEOGRAPHY. 35 stellations; “the late-setting Bootes,” and “ the Pleiades, and the Hyades and the mighty Orion;”2 but it would be absurd to suppose that because these are all that he notices by name, they were the only ones with which he was acquainted. Indeed the mention of so obscure a group as the Hyades sufficiently shows that the nomenclature of the stars was in his time pretty well advanced. The only single star that he notices is the clog-star, with the baleful influence of which he was familiar, as well as with its peculiar brightness.3 §4. \Ve do not find in the Homeric poems any distinctive terms for the cardinal points, or four quarters of the heavens, as such: that is to say, he has no words answering to the simple terms, North,v South, East, and West. He indeed repeatedly contrasts the two latter—which must always have been clearly marked by the course of the sun—designating the one as “ t_q- wards the dawn and the Sun;,” the other as “ arkness ”—on account of the setting sun being immediately followed by dark- ness.4 But his use of the terms is, as might be expected of a poet, especially in a rude age, somewhat vague and general, and this has actually led several writers, both in ancient and modern times, to dispute the signification of the terms, and to maintain that by the quarter or region of darkness Homer meant the north; and the south by that of the sun; a theory which is at once met by the unanswerable objection that it takes no account of the mention of the Dawn, which is in these phrases invariably associated with that of the Sun.5 The 2 Hkql'ci'o‘ag 0’ ‘Yo't'da; 1'6, 1'6 7'6 0'96'1/09 ’Qpiwvo;. Iliad, Xviii. 486. Compare Odyssey, v. 272. ' 3 He calls it “the dog of Orion” (Iliad, xxii. 29). It had apparently not yet acquired the distinctive appel- lation of Sirius, z'.c. “the burning,” which is already found in Hesiod (0}). et Di. v. 607). The (20797;) 51rwpwb$ mentioned in v. 5 is supposed by some commentators to be Arcturus, but it is more probable that here also Sirius is meant. 4 The one is always 1rpbs 3743 'r’ 'he'Mo'V 're: the other, as opposed to it, 1rpbs (64ml), or 'n'o'rt §6¢0V 'hspdev'ra (Iliad, xii. 239; Odyssey, xiii. 240, &c.) 5 This paradox was put forward in ancient times by Strabo (x. p. 455), and apparently before him by Ephorus (ap. Strab. i. p. 84). It has in modern times been adopted by Voss, who made it the foundation of his whole system of Homeric geography; but is, in my opinion, fully refuted by Volcker (Homerz'schc Geographic, pp. 42-46), and has been generally rejected by the most recent inquirers (Ukert, Nitzsch, D2 36 HISTORY OF ANCIENT GEOGRAPHY. CHAP. III. ' people. question could not have arisen, but for the absence of any .Qorresponding terms to designate the North and South. The expressions employed by Homer to distinguish what we Lzsihould term the quarters of the heavens, are almost always erived from the winds, the names of which must have been from the earliest ages familiar to the Greeks as a seafaring Of these he knew only four, which he describes as “opposed to one another in pairs, and which may fairly be taken-as representing in a general way the four cardinal points I I of the compass.6 Borgas‘f' the north wind, blowing from Thrace, assumes in Homer the same prominent position that it does at the present day to all navigators of the ZEgean. It was the bringer of fine weather and a clear sky, but nevertheless a strong and violent wind “ that rolled mighty waves.”7 .NQIJLS, - the south wind, which was directly opposed to it, was a stormy wind, bringing sudden squalls dangerous to navi- gators. Zepflhyrus, the west wind, was dreaded on the same account, and is repeatedly represented as ,a stormy wind ;8 while E_1_1_rus, the east wind, is but seldom mentioned, and bears a subordinate part: very different from what it would have done, had the poet lived in England. § 5. Homer abounds in descriptions and descriptive epithets of the sea, from which a large part of his similes are taken. No Buchholz, &c.). Mr. Gladstone, after pointing out the true meaning of the phrases, and justly adding that they must at the same time be received with a certain amount of vagueness, proceeds to argue as if they are to be taken definitely as corresponding to north-west and south-east (vol. iii. p. 266), a pro- position for which I cannot see the slightest foundation, though he thence- forth relies on it as an established fact. The deviation from the ecliptic, to which he appeals (even if its influence be admitted), would of course give an equal amount of variation in each direction, to the south as well as to the north, of the equinoctial line of sunset and sunrise. 6 Note C, p. 77. 7 Kai Bopé-q; aiepnyeve'ms, Ité'ya xiii/.11 KvM'v- 8001/. Odyssey, v. 296. But it was generally associated with fine weather and a clear sky (whence the epithets aifipn'yev'hs and aidpn'ye- ue’rns), and is described in aremarkable passage in the Odyssey as “breaking down before it” the waves that had been raised by the other winds—— <3pcre 8’ e'rri. Kpaurvbv Bopéiqu, up?) as mipar’éi'ézsfev. V. . a fact familiar to all who have had much experience either of the .ZEgean or Mediterranean. But as coming from the north it was naturally the wind that brought frost and snow (Iliad, xix. 358 ; Odyssey, xiv. 4:75). 8 See especially the passage in the SECT. 1. HOMERIO GEOGRAPHY. 3 7 poet has ever equalled him in the knowledge he shows of its varied moods and aspects, with which every Greek was neces- sarily familiar. But he affords us vscarcely any geographical statements concerning it. He never applies the name of ZEgean, so well known to the later Greeks, to the sea between Greece and Asia: for which he had no distinctive appellation, for the same reason that the Greeks in later ages had none for the Mediterranean: it was “the sea,” the sea which everybody knew, and had he wanted to distinguish it, he would doubtless have called it in like manner “ our sea.” But in two passages he designates smaller portions of/it by the local names of the Thraeian Sea/ and the Icarian Sea: and other such appellations were doubtless in use in his time. Hence he speaks of “ every sea,” as if there were many such, and tells us that all seas, as well as all rivers, were the offspring of Ocean, which was indeed the source of all the waters on the earth, even fountains and wells.9 But the distinction between the mighty river of Ocean, and the sea which was enclosed within it, is through- out clearly maintained. The idea of the Ocean as an outer sea, separated from, but communicating with, the inner sea, was wholly foreign to-the mind of Homer, and many of the difficulties and errors of his commentators have arisen from their attempting to introduce into the Homeric geography this conception of the Ocean, as it was familiar to their own minds.1 I . It must be added that no indication is found in Homer of any knowledge of the Euxine as a separate sea. He was of course familiar with the Hellespont, from. its proximity to Troy, and must have known that beyond that narrow strait Odyssey (xii. 288-290) where the that the Zephyrus blowing from the Notus and Zephyr-us are spoken of as raising sudden squalls and gales such as were the cause of shipwrecks. Again in the same book (v. 408) there comes on suddenly “a shrilly west wind with a violent squall,” Kelmn'yo‘vs Ze'4wpos, ,ue'ydkp o‘i'u AatAa-rn 615ml. It is only in describing the Elysian Fields Ocean is represented as a gentle cooling breeze QOdyssey, iv. 568). 9 Baevfafau'rao p.éya atk'vog 'Qxeavoio, ’ ozirrep 'n'oiv're; 1ro'rap.oi.. Kai. mica. 6011\(10'011, Kai n-cio'al. rcpflvac, Kai. (,bpeia'ra (aaxpo‘t voiovo'w. v Iliad, xxi. 195-197. 1 This difiioulty was apparently en- 38 HISTORY or ANCIENT GEOGRAPHY. arm. 111. there was (again a broad open sea—the Propontis or Sea of Marmorai—but there is nothing to show that he had any knowledge of the Bosphorusfor of the far more extensive sea beyond.2 He supposed the Achelous to be the greatest of rivers, which sufliciently shows that not even any obscure rumour had reached him of such mighty streams as the Ister and Borysthenes. § 6. It is hardly necessary to mention that the primitive geo- graphy of the Homeric times knew nothing of a division of the world into three continents: and that the names of Europe, Asia and Africa, in this sense, were wholly unknown to the poet. He indeed mentions the name of Libya (as Africa was always called by the Greeks) as that of a particular region of great fertility, doubtless referring to the country west of Egypt which always continued to be known by this special designation. But he never mentions the name of Asia, even as that of a country: and though he speaks of the Asian meadows on the banks of the Oayster, this was evidently a mere local appellation.3 The name of Europe does not occur in the poems at all.4 countered by the Alexandrian critics who, like Eratosthenes, transferred the wanderings of Ulysses to the outer Ocean, while they still retained the idea of the Ocean, such as it was known to them, instead of the deep, gently flow- ing, stream that Homer describes, encircling the whole world, earth, and sea alike. 2 He was, indeed, acquainted with the river Sangarius which flows into the Euxine (Iliad, iii. 187; xvi. 719), but only as a river of Phrygia, and there is no reason to suppose that he knew anything of its real outflow, or of any distinction between the Pro- pontis and the Euxine. He knew generally that there was sea somewhere to the north, and that was all. 3 'Ao‘irp év Aeqtdwt, Kaiia'vpiov dla¢i fie'eapa. Iliad, ii. 461. 4 It is first found in the Homeric Hymn to the Pythian Apollo (v. 73, 112), which is probably at least two centuries later than the Iliad or Odyssey. But even there the name of Ebpu’nm is used only to designate continental Greece, in contradistinction to the Peloponnese and the Islands. Its extension to the continent gene- rally was probably of much later date, though it was, as we know, already well established in the time of He- cataeus. SECT. 2. . HOMERIO GEOGRAPHY. / SECTION 2.—Geogmphy of the Iliad. § 1. In proceeding to investigate the local geography of the Homeric Poems, and to point out the extent and character of the knowledge which they really display of different seas and lands, we may safely start from the principle which was long ago laid down by Eratosthenes, that the poet was well acquainted with the regions near at hand, but ignorant of those afar off,5 This conclusion indeed appears at first so obvious, or rather so self-evident, that it is difiicult to suppose it could ever have been contested: and it is- strange to find it rejected with scorn by such writers as Strabo and Polybius.v But the reverence for the supposed authority of Homer, which had arisen into a kind of blind superstition among the later Greeks, led them to regard with indignation any attempt to curtail the domain of knowledge attributed to him. Hemwas the father of geography as well as of history, and it was held to be equally heretical to dispute his statement as to a sup- posed geographical faet, as to deny the reality of the Trojan ‘War, or to doubt the personal existence of Ajax or Patroclus. At the same time, strangely enough, this unquestioning belief in the accuracy of the poet was coupled with an equally ready admission of the popular legends which identified the localities described in the Odyssey with certain definite spots that had become familiar to Greek colonists and navigators in later times. How these attributions had at first arisen, we are inmost cases utterly unable to divine; but, once established, they maintained their ground with a singular tenacity, even where they were in palpable contradiction with the statements of the poet himself. Thus the abode of ZEolus gvas considered as fixed beyond dispute in one of the Lipari, or, as they were called by the ancients, the ZEolian Islands, though Ulysses is 5 Eratosthenes ap. Strab. i. p. 18, 'EAMp/ucd side/vat, T5511 5% 1rdppw 1r02\2\'}711 Vii. p. 298: e’rrouveT 'yc‘zp ’Epa'rco'6éuovs é'xew dmrezpl'av. This principle was durdqraa'w, (‘in qmatz/ éxeTz/os Kai"0,u1;pou adopted by Apollodorus, which excites Kat 'roils iiAAovs robs 1raAazobs 'rez ah! the indignation Of Strabo. 4O HISTORY OF ANCIENT GEOGRAPHY. CHAP. III. described as sailing from thence with a west wind in a direct course to Ithaca.6 Such a course was obviously impossible, unless we suppose him to have sailed right across the mountains of Calabria, or frankly admit that Homer was unaware of the existence of any such ‘obstacle. ‘We are met by no such difficulties nearer home; and as long as our attention is confined to the shores of the Egean Sea, and the countries that immediately adjoin it, we find the poet generally well informed, and we have sometimes clear evidence of accurate personal observation. I It is only when we wander beyond this limited range, that we begin to meet with contra- dictions and absurdities, and statements bearing the stamp either of vague ignorance, or of palpable poetic fiction. The distinction thus established between what have been well called the Inner and the Outer Geography of Homer, coincides nearly, though not entirely, with that of the two poems. As the action of the Iliad is confined within the limits of the inner circle, and the nations represented as engaging in conflicts on .the plains of Troy are the inhabitants only of the regions bordering on the ZEgean, the allusions to anything beyond those limits are_ few and incidental, and it will be found convenient to take the geography of the Iliad in the first instance by itself, as representing the more definite portion of what was known to the Greeks in these early days, before entering on the wider field of the unknown regions opened out to us in the Odyssey. But in adopting this division of the subject, I desire not to be understood as resting it upon any other basis than convenience, or adopting the opinion of those who would assign the Iliad and Odyssey to different authors, and even consider them as separated by a considerable interval of time. So far as geographical evidence goes, I can see nothing to necessitate or to warrant such a separation.7 “ Odyssey, x. 25-30. graphy of Homer, especially that of 1 The treatises that have been the Odyssey, are so numerous as to written in modern times on the geo- form almost a literature in themselves. SEW-2. HOMERIC GEOGRAPHY. 4r §2. The most prominent place in the geography of the Iliad must undoubtedly be assigned to the Catalogue of the Ships, as it is commonly called, though it would certainly be more accurately termed the Catalogue of the Forces, with the names of the leaders under whom the several contingents were arrayed. Such an enumeration naturally gave occasion to the poet to show his accurate and minute acquaintance with the divisions of the populationf/such as they existed, or as he supposed them to exist, at the time of the Trojan War, as well as with the names of the cities and towns that they possessed; and as the reverence for the name of Homer conti- nually increased, this Catalogue gradually came to be invested with a kind of official authority. At the same time this very cause afforded great inducement to interpolation, for which the peculiar character of this part of the poem presented the greatest facilities.8 But while it cannot be denied that such interpolations may have been introduced—perhaps even to a considerable extent-two considerations appear to prove that the Catalogue in its/ present state is substantially the work of a very early period. i The one of these is the ‘close agreement of the names and attributes of the leaders and chiefs enume- It may suffice to refer to those of Ukert (Be'merkzmgen iiber Homer's Geographic, 8vo. VVeimar, 1814, and in his Geographic der Griechen a. Ro'mer, vol. i. pp. 13-33) ; Volcker (iiber Homerische Geographic and Welthunde, Hanover, 1830), and to the valuable notes of Nitzsch on the Odyssey (Anmerlmngen zu Homer’s Odyssee, 3 vols. 8V0. Hanover, 1826-40) as well as the more recent investigations of Dr. Buehholz. (Die Homerischen Realien. 8V0. Leipzig, 1871, vol. i.) The subject has been reopened in this country by Mr. Gladstone (in his Homer and the Homeric Times, vol. iii. pp. 249-365), whose views differ more widely from those of all previous in- quirers than any other. Their unten- able character has in my opinion been fully shown in an able article in the Edinburgh Review (for October, 1858), which I have reason to know to have been written by the late Mr. Herman Merivale. The reviewer, however, while successfully combating the novel theories of Mr. Gladstone, has gone much further than I am disposed to follow him in support of the tradition ary system, as handed down to us from the commentators and geographers of antiquity. 8 A well-known instance of this is mentioned by Strabo (ix. 1, p. 394), in regard to the two lines describing the contingent of Ajax: Ala; 6’ e'lc 'iaAa/a'ivo; dyer! Svoxafdexa. Vijaq, 017706 8’ a-ywv, w’ 'Admrauuv LO'TO-VTO qSdAayyes. where the second line was asserted by the Megarians to have been fabri- cated by the Athenians in support of their pretensions to the island of Salamis. 42 HISTORY OF ANCIENT GEOGRAPHY. CHAP. III. rated, with the subsequent notices of them scattered through the Iliad: the other is the fact that in the enumeration of the towns from which they came, we find the names of pbsvqlll‘e- pm which had dwindled into insignificance, or whose very . sites were unknown in later days, freely intermingled with ‘those of illustrious cities, whose names were familiar to all.9 / A considerable portion of Strabo’s geographical description of Greece is taken up with investigating cases such as these, or explaining topographical difficulties, which could not have arisen unless the state of things described by the poet had been separated by a considerable interval from the historical period of Greece. / i \ As might well have been exvpectedkthgwQatalogue contains )yery little geographical information, properly. SO Balledi it is wreyegngmreration ofnamesrandplacesxwjthnhut rarely any reference to their geographical position, beyond what is im- plied by their being included in a given nationality.1 But that by which it is especially characterized is the number and variety of the epithets- applied to the difi'erentxtowns ,emnume- ratediepithets‘ sometimes of an ordinary character, and fre- quently repeated; in others singularly distinctive and appro- priate. Such are the terms of “ well-walled ”'(or more literally “wally ”), applied to Tiryns;2 of “abounding in doves,” to Thisbe, in Boeotia;3 and the “hollow Lacedaemon cleft with 9 Thus Strabo himself says in regard t0 Laconia: 113V 6’ EiAAwz/ 7561/ {me 1'05 wom'roii na'rwvopao'pe'vwv ‘rd. ,utu dzqipn- 'raL, 1'61! 5’ 1x111] Aei'lre'rat, 'rc‘z. 5% yea-(0116- naarraz : while he admits that some could not be pointed out at all (oboanofi fiche/1:06:11 (Pam), viii. 5, p. 364. 1 Sometimes, however, the poet adds a few words indicating the position of a town with reference to a mountain or river. Thus he describes Lilaea as situated at the sources of the Cephissus (ii. 523), Tarphe and Cronium on the banks of the Boagrius (2'1). 533), Pheree by the lake Boebei's (2'12. 711), and the Magnesians as dwelling around the Pencus and the wooded Pelion (ii). 757). 2 Tipvz/Bo'r 're ruxcéeo'aav. Iliad, ii. 559. The only other city to which the same epithet is applied, is Gortyna in Crete (ib. (546), where, however, there are no remains of Cyclopean walls. a nokv'rpfipwvoi 're @t'crfinll. Iliad, ii. 502. The abundance of wild pigeons in the rocks of this locality is attested both by Colonel Leake (Northern Greece, vol. ii. p. 507) and Sir Thomas Wyse (Impressions of Greece, p. 85), as well as by Strabo. The same epithet is applied to Messe in Laeonia (v. 582), but the exact locality is in this instance uncertain. Snc'r. 2. HOMERIG GEOGRAPHY. 43 glens;”4 a description which, from its recurrence in the Odyssey, was probably become one in general use? Indeed it may always be doubtful whether the distinguishing epithets that we find in Homer were selected by himself, or had already become traditional from their employment by earlier poets.5 Be this as it may, we must take the case as it stands. It is impossible for us now to determine how far the Homeric Catalogue was based upon previously existing materials, as well as to what extent it may have been interpolated in later’ ages. If we cannot accept its authority with the blind defe-\ rence shown by the later Greeks, we must receive it as a document of the highest interest from its antiquity, but which , it is impossible for us to criticize or dissect, from the total l‘ absence of the requisite materials. '1 §3. Of a very different character from the Catalogue of the Greek forces, is that of the Trojans which follows it, and which has very much the appearance of being a subsequent addition to the poem. Its meagre and jejune treatment of the subject presents a striking contrast to the fulness and richness of the Greek Catalogue, and it is as inferior to the latter in geographical interest as in poetic merit. But it was universally admitted in ancient times as an integral part of the Iliad; so much so, indeed, as to be made the subject of an elaborate commentary by Demetrius of Sceps_i_:_s_,_a work extending to not less than thirty books.6 We must, therefore, such as must have preceded the Iliad or the Odyssey. 4 Koikqv Aaxesat'uova Krrra'reo'o'av. Iliad, ii. 581. The signification of this last epithet, which occurs both in the Iliad and Odyssey (iv. 1), but only as applied to Lacedsemon, would be uncertain, but for the very peculiar character of the scenery there, which (as Mr. W. G. Clark justly observes, Pclopomzesus, p. 156) at once helps one to the mean- ing of the word. 5 This is suggested by Mr. Clark (Ibid. p. 210), and appears to me almost certain. Such an habitual ap- plication of distinguishing epithets is 6 Strabo not unnaturally expresses his wonder at the difi'useness of a. writer who devoted thirty books to a commentary on little more than sixty lines of Homer (xiii. 1, p. 603). But from the quotations preserved to us it is evident that it comprised much valuable matter; and Demetrius ap_ pears to have been one of the writers who shared in those more sceptical views regarding the Homeric geo- graphy, which were viewed with so 1 much disfavour by Strabo himself. characteristic of all early ballad poetry, I 44 HISTORY OF ANCIENT GEOGRAPHY. CH.AP III. . farther. be‘ content to include this also in our estimate of the Homeric geography, though not without a passing protest.7 § 4. The most interesting question, in a geographical point of view, that arises from the two Catalogues is, the extent of the knowledge which they evince of the countries adjoining the Egean, for all that is comprised within their limits may be assumed to have been well known to the poet. It cannot, on the other hand, be affirmed that all beyond these limits was unknown, or at least imperfectly known, though we shall find reason to believe that his real knowledge extended but little continental Greece, inc uding the‘jleloponnese and Thessaly, but exclilding Epirus and Aharnaniar~ It comprises, more- over, the islands on the west, opposite to Elis, Cephallenia, Ithaca, and ,Zacynthus, as well as Dulichium and the Echi- nades nearer to the mainland; and all the islands in the south of the ZEgean—not only the important islands of Crete and ‘ Rhodes, but Carpathus, Cos, and even such insignificant islets as Casos, Syme,,.Nisyros, and the Calydnee—being distinctly mentioned by name. On the other hand, none of those which might be thought especially the Greek islands, viz. the Cyclades, and those connected with them, are mentioned as sending a contingent to the Greek forces, though it cannot be doubted that Homer was well acquainted with them.8 In like manner we know that he was familiar with those near the coast of Thrace, Lemnos, Imbros, and SQLHOtllI'BQCGf as well as The description in the Catalogue extends over all. ...~-“-..~< m1 <\.-. A _...,, ‘4" ‘Der Epische Cycles, vol. ii. 7 It is a strong argument against the authenticity of this portion of the poem, that the Cypriot of Stasinus, a work which seems to have been com- posed with especial reference to the Iliad, contained also a catalogue of the Trojan forces ; a strangely inappro- priate addition, if the one now extant was then found in the Iliad. (Welcker, p. 508; Mure’s History of Greek Literature, vol. ii. p. 281.) 8 N o allusion is found in the Iliad to the island of Delos and its celebrated sanctuary. But it is incidentally re— ferred to in the Odyssey (vi. 162), and its non-occurrence in the Iliad may well be accidental. 9 It is scarcely necessary to refer to the well-known passage in Mr. King lake’s Eothen, in which he points out the accurate local knowledge displayed by Homer in selecting Samothrace as the point from whence Neptune looked down upon the plains of Troy; an idea wholly at variance with the natural presumption from its position on a. map. SECT. 2. HOMERIC GEOGRAPHY. 45 with the great mountain promontolypfflAdthos. If we now take up the Trojan Catalogue we shall find it describing in detail only the Tread and its immediate neighbourhood, but enumerating also contingents from the opposite coasts of Thracefiincluding thegPeeonians from the banks of the Axius, a stream which the poet describes as “the fairest that flows upon the earth.” 1 The auxiliary forces from the Asiatic side were naturally more ‘,numerous: these were the Mysians,w Phrygians, llIzeonians-ii—whose site in Lydia is marked by the Gygt'ean Lake and Mount Tmolus—the Carians, with whom are associated “ the streams of Maeander and the lofty summits of llfygfglleiland the Lycians, who came “from the far Lycia and the_ba_nksof_the eddyi—ngr Xanthus._’_’_ These were the farthest people to the south. To the north we find mention of the Paphlagonians, occupying apparently the same reglon as m. . historical times; and beyond these the Halizones “from Alybe afar, where is the birthplace of silver.”2 Whether these were the same people subsequently known as the Cha- lybes, was a point much disputed among ancient authors: at all events we must clearly place them to the east of the Paphlagonians, and may therefore safely assume that Homer (or, at least, the author of the Trojan Catalogue) was ac- quainted by name with the nations on the south coast of the lipxinegas n far as ptllemlnlalys, though the name of that river does not occur in the poems. § 5. But if the limited area thus circumscribed may be taken as including all, or nearly all, of which the poet of the Iliad can be shown to have had any definite knowledge, there are not wanting indications that point to an acquaintance, how- ever vague, with far more distant regions. And the very fact that these notices are for the most part merely incidental, l 'ASaoiLoii KziMco'rov 5800p e’n-mr'dr'a'rar alfav. ‘ it, See Strabo, Xli- pp- 549-553- The Iliad. ii- 850- I name of the H alizones had disappeared 6951/ dpytipov émi in later times, but it is found in another , n k : passage in ‘the Iliad (v. 39). so that Ibid‘ “5" i it does not rest only on the dubious For the discussions raised concerning authority of the Trojan Catalogue. 2 TnAéBev ég ’A}\1§BY]§, yevédiuq. I 46 HISTORY or ANCIENT GEOGRAPHY. GHAP. III. points the more strongly to their being notions already familiar to the poet himself, and which he might conceive to be equally well known to his hearers. One of the most remarkable of these passages is that in which he describes Zeus as turning away his eyes from the battles before Troy to gaze on “the land of the horse-loving Thracians and the close-fighting Mysians, and the noble Hippemolgi, who live on milk alone, and the Abii, the most virtuous of mankind.” 3 It is a question that has been much disputed, and that cannot be decided, whether the terms Hip emo'lghj‘milkers of marea',” Glactophagi, “feeders on and Abii, “men without property,“ are used by the poet as proper names, or only as descriptive adjectives,5 but in any case it is clear that the poet was vaguely acquainted by hearsay with the fact that I beyond the mountains of Thrace were to be found extensive plains, over which roamed wandering tribes, having the pecu- liarity that they subsisted mainly on the milk of their mares, instead of that of cows or sheep. This characteristic has in all ages distinguished the Scythian tribes, that is, the nomad races of Central Asia, or who have passed over from thence into Europe, and was familiar to the Greeks in later times. But neither the name of Scythians, nor that of any of the 3 vdo'duv ért' i1r1r0‘n'67lwv (DPymInI Kadopa'anevos av Mvo'dw 'r' dyxepdxwv Kai. dyavdw ‘Imm- o (In! yhax'rogbdywv, ’ABiwv ‘re, Smatordrwv riv- apu'mwv. Iliad, xiii. 4—6. The Mysians here mentioned are evi- dently the European people of the name, the same afterwards known to the Romans as the Moesians, from whom, according to the belief univer- sally prevalent among Greek writers, the Mysians of Asia Minor derived their origin. 4 ‘ABror, literally, “men without the means of life,” tie. with poor and scanty means.” dM'ydBior, ei’rreAeTs 'yoip eia'z Kai d'rrAoT 'r'hv ‘o‘r’arrau, Eustath. ad Zoo. The idea of their justice and virtue was evidently derived from their simple and frugal habits of life; and continued to prevail among the Greeks through all ages, though applied some- times to one nomad tribe, sometimes to another (see lEschylus, Prometh. Solnt. Fr. 184; Arrian, Anabasz's, iv. 1). 5 See the subject fully discussed by Strabo, vii. 3, §§ 2—10, and by Eusta- thius, ad Iliad. 1.0. The analogy of such names as Ichthyophagi, Troglo- dytae, Macrobii, &c., which were cer- tainly used by the Greeks as Gentile appellations, is in favour of their being so in this case also; and accordingly we find that the Hippemolgi and Abii were generally regarded as such. But little or no value can be attached to the usage of later writers in this respect, in interpreting the language of Homer. SE01‘. 2. HOMERIC GEOGRAPHY. 47 subordinate divisions of that great people which we meet with in Herodotus and Ephorus, is to be found in Homer. § 6. In regard to the countries lying to the south and east, there are many more indications of knowledge, and even in a certain sense of familiar intercourse. .withthencirilized nations at the south-eastern. angle of the Mediterranean—the Phoe- nicians-‘and Egyptians. To the former people, indeed, as has been already shown, the Greeks were indebted for almost all their foreign trade; and they must have been therefore quite familiar with the name, as well as with the articles exported from thence.6 But not only does Homer represent Paris as possessing many valuable things of Phoenician manufacture, but he distinctly describes him as having brought them him- self from Sidon, “after having sailed over the broad sea.”7 He was apparently regarded as having been driven there by a violent storm, in the same manner as Menelaus was in the Odyssey; but in both cases it is clear that the poet looked on such an adventure as not improbable. In the case of Egypt again, though accidentally the name of the country does not occur in the Iliad, we find an incidental reference to the Egyptian Thebes, which shows that exaggerated rumours of its wealth and grandeur were already familiar to the poet. In the ninth book Achilles alludes, as the acme of all imaginable wealth, to that which was stored up within the houses of the Egyptian Thebes, “ a city that had an hundred gates, through each of which there passed every day two hundred men with 6 It is remarkable, as has been fre- quently pointed out, that the name of “Phoenician,” though of frequent occurrence in the Odyssey, is only once found in the Iliad; the manu— factured articles are always called “Sidonian,” and Paris is described as bringing them from Sidon. But this is easily accounted for by the prepon- derance of the capital. The name of Tyre is not found in either poem. 7 Eve‘ é'a'av 0i. 1ré1rAot wau'n'oc'nmot, Ep'ya. 'yv- VG-LKLOV Etsoviwv, 7&9 afrrb'; 'AAésou/Spos‘ 9606.8?” fi'yovye Eider/[179611, ézrmkcbq efzpe'a. adv-rev. lliad, vi. 288-2911 As the poet places this visit on his return voyage from Sparta, after having carried off Helen (2b. 292), it has been sought to connect it with the story, adopted by Euripides and Herodotus, of Paris having touched in Egypt also on his return; but there is nothing to warrant this assumption, except that it is more probable he should have been driven, like Menelaus, to Egypt, than direct to the Phoenician coast. 48 HISTORY or ANCIENT GEOGRAPHY. crap. III. their horses and chariots.”8 Even these fabulous tales are sufficient proof that the Greeks in the time of Homer had intercourse, more or less direct, with Egypt, as we shall find more clearly shown in the Odyssey. § 7. Through Egypt also was unquestionably derived the notion of the Ethiopians (burnt or black-faced men), who must have been regarded as dwelling beyond the Egyptians to the south, on the borders of the Ocean stream. The utterly vague and fabulous character of all that was known concerning them is sufficiently shown by the fact that they are never mentioned except in connection with the gods, who were supposed to repair to the banks of the Ocean to feast on the sacrifices offered by them.9 Equally vague is the indication of another people, the Pygmies, who were also supposed to dwell by the shores of the Ocean stream, where they were engaged in con- tinual wars with the cranes, who migrated thither to fly from the winter and the storms of Greece.1 The notion of the existence of such a race of dwarfs in the interior of Africa, which was long prevalent among the Greeks, and has recently been shown to be not altogether without foundation,2 could only have reached the Greeks of Homer’s time through an Egyptian channel. alluded to is one that the poet assumes to be familiar to all. 0138’ 30’ £9 ’Opx6,u€v0v 1ro-rwida'erat, 0138’ 300. (9153a; Ai-yvrrn'as, 30L 'n'Ae'io'ra $651.01.; e’v x'nj/ra'ra. “27m, 01 9’ éxa'rdpm'vkoi elm, 8L11K617LOL 8’ all!’ éiczio'ras dwépes eio'otxveiidc ow 'trr-rroto'w Kai 5x60‘- 11. Iliad, iX. 381—384. Diodorus (i. 45) asserts, as if he had undoubted authority for the fact, that Thebes, in the days of its greatness, really did possess 20,000 war chariots. 9 Iliad, i. 428; xxiii. 206. Pre- cisely the same notion is found in the Odyssey also (i. 22-25; v, 282). l fid're ‘n‘ep Khwyyiy yepoivwv 'n'éAei. oi’Jpavo'Gt rpm at r’ é-rrei. 051/ Xeqrdwa dniyov Kai. deéo'rba'rov Fmfipov. Icka'yyfi ‘mi 76 1ré'rov'rai. e'rr’ 'Qlceowoio fioéwv, dwdpdm. IIv'ypat'oto't qSdvov Kai Kfipa. (1)6’- pOUO'CLL. Iliad, iii. 3-6. It is evident that the tradition here 2 Rumours of the existence of a race of dwarfs in the interior of Africa have from time to time reached the ears of modern travellers, as they penetrated farther and farther into the continent. Quite recently (1870) M. Schweinfurth has established beyond a doubt the fact that such a race is really found adjoining the tribe of N (yamnyam in Central Africa. They are known by the name of Akka, and have many points of resemblance with the Bush- men of Southern Africa.V It is not im- probable that such tribes were once more widely diffused, and may well have come within the cognisance of the Egyptians. (See Schweinfiu'th’s Travels in the Heart of Africa, vol. ii. ch. xvi., and Keith J ohnston’s Africa, p. 241.) SE01‘. 3. HOMERIC GEOGRAPHY. 49 SECTION 3.-—Geography of the Odyssey. § 1. Few and scanty as are the allusions in the Iliad to dis- tant regions and countriesjthey are still sufficient to show that the author of that poem was well aware of the existence of an outer world, beyond the limits of that to which the view of the Greeks was in his day habitually confined. This outer world is far more fully opened to our view, though still enveloped in the dim haze of uncertain distance, in the Odyssey, and espe- cially in that part of the poem which has always been the most attractive and the best known. The wanderings of Ulysses, as related by himself to the Phgiacians, have had a charm for all readers of all ages from the times of Homer to our own: and few, very few comparatively, have been those _who have paused to enquire what slender foundation of truth might underlie these delightful fictions, and whether the scenes of his adventures were real localities disguised under a thin veil of poetic ornament, or were mere creations of fancy, as shadowy and unreal as the Elysian Fields or the gloomy realms of Hades. One point, indeed, may be taken as univer- sally admitted. The legendary and mythical tales, which received their definite embodiment in the Odyssey, were not the creation of the poet. The names of the Sirens and Circe, of _Scylla and Charybdis, of the Lotus-eaters and the Laestry; genes, were already familiar to the Greeks before they were.. wedded to immortal verse by Homer: and there can be no doubt that the \Vanderings of Ulysses, like the Voyage of the ship Argo, had formed the subject of many a rude ballad, before they were worked up into a polished and coherent narrative by the author of the Odyssey. It is especially necessary to bear this in mind, in attempting to investigate the questions concerning the geographical foundation of the story as we find it there related: or rather to enquire how far there existed any such foundation at all. \Ve must represent to ourselves the poet as having to deal with a mass of hetero- geneous materials: sometimes obscure reports gathered from VOL.L E 50 HISTORY or ANCIENT GEOGRAPHY. GHAP. III. ’’ Phoenician mariners of the marvels of the “ far west; others purely mythological fables of unknown origin; others again traditional tales belonging to that border-land of truth and fiction, in which it has become impossible for after ages to discern the boundary. . All this he had to work up into one poetic whole, and no one can dispute his success; but when it has been maintained, both in ancient and modern times, that he undertook also to work it up into one geographical system, it may well be [asked whether he had any geographical system at all? The ancient writers, indeed, for the most part believed that in this respect, as in all others, Homer was the master-spirit and guide to all learning,3 and Strabo expressly describes him as the founder of scientific geography and the precursor in this _ department of Anaximander and Hecataeus,4 But it is strange hat the same erroneous estimate of the scope and character of such a poem as the Odyssey should prevail even in our own day, and that one of the latest attempts to investigate its geography should be based upon the assumption that Homer had present to his mind an elaborate and complicated map of the wanderings of Ulysses, utterly without foundation in fact, but as distinct and definite as a chart of the voyage of Ma- gellan or Vasco de Gama.5 3 “ Il maestro di color che sanno.” Dante, Inferno, canto iv. v. 131. The expression is, however, applied by the Italian poet with better judgment to Aristotle. . 4 Strabo, i. 1, where he distinctly asserts that geography is a branch of philosophy, and that all those who had taken it up—among whom he gives the first place to Homer—were 'rowb‘roi 'rwes, z'.e., men of a philosophical turn of mind. 5 This appears to me to be the funda- mental fallacy, pervading all Mr. Gladstone’s system of Homeric geo- graphy. According to his own state- ment: “The question before us is what map of the earth did Homer shape in his own mind, that he might adjust to it the voyages of his heroes, Menelaus and Ulysses” (vol. iii. p. 250). But what right have we to assume that he shaped any such map at all? The use of maps was not in- troduced into Greece till centuries after the time of Homer; and all ex— perience shows that it is only by the habitual use of maps that men come to form clear conceptions of the geo- graphical relations and relative posi- tions of different countries. Without this training I believe it to be impos- sible for any one to frame in his own mind any system of such geographical relations, as between distant countries ; though he may clearly picture to himself those of neighbouring lands and islands, such as the shores of the SE01‘. 3. HOMERIC GEOGRAPHY. I SI §2. We have already seen how deeply rooted among the ancients, both Greeks and Romans, was the belief that attached the names of the Homeric localities to particular places, well known in later times. How these attributions had grown up in the first instance, we are in most cases utterly unable to conjecture. In some they were probably local legends or myths that had some kind of resemblance to those that were found in the Odyssey, and they were in consequence eagerly laid hold of in order to connect the particular localities with names that were celebrated throughout the Greek world. It is certain that they rarely, if ever, arose from any geogra- phical attempt to explain or account for, the wanderings of the hero, in accordance with the statements of the poet.6 On the contrary, it has been already observed that these popular attributions were in many cases in direct contradiction with the text; and later writers in vain struggled to reconcile the two. The only safe course that can be pursued in this enquiry is to discard in the first instance all those commonly received ' identifications, and simply follow the statements of Homer, pointing out where they really afford some clue to their own interpretation, and where they are clearly at variance with the ideas that have been commonly attached to them. § 3. The voyage of Ulysses, on his return from Troy, began of course in the well-known waters of the Egeanfhnd here the account is, as might be expected, clear and consistent. He was driven in the first instance, by unfavourable winds, to the land of the Cicon'espon the coast of Thrace _;7 a people who appear in the Iliad among the Trojan allies,8 and with whom ZEgean, which lie within the range of his own experience and observa- tion. 6 The only exception would be the fixing on Gaudos (Gozo) as the island of Calypso, if we are to understand Strabo's words (vii. 3, p. 299) as im- plying that Callimachus was the first to make this attribution; but it ap-~ pears probable that in this case, as in that of Scheria, he merely adopted the popular tradition. 7 'IMo'Oeu ,ae (,bépwv oil/sue; Kucoveo'cn 1re')\ao'— o'eu. Odyss. ix. 39. 8 Iliad, ii. 846. According to Herodotus (vii. 59), the Cicones in- habited the coast land about Doriscus and the mouth of the Hebrus, but they were extinct as a people in the time of the historian. E 2 52 HISTORY or ANCIENT GEOGRAPHY. CHAP. III. Ulysses engaged in hostilities. Sailing again from thence, he had a north wind (Boreas) which would have been favourable for his course, but its increasing violence drove him to seek shelter on the land, where we are not told: nor have we any hint as to his course from thence, except that it was a direct course with a fair wind (evid ntly, therefore, still a north wind), as far as CapeflMalea, where the violence of the wind, and of the waves_and current, drove him out to sea, and pre- vented his seeking shelter undgilugypheoraf Here we have a very characteristic trait, Cape Malea having been renowned in all ages for sudden and violent storms, so that its circumnavi- gation has been always regarded with dread both by Greek and other marinersi/ In the Odyssey itself it is twice again mentioned as having caused similar disasters.1 From thence they were driven “ by wicked winds ” for nine days over the open sea.2 A voyage of that length would undoubtedly carry a Greek navigator of those days altogether ‘out of the sphere of his knowledge, and accordingly when 9n the tenth day they arrived in the land of the Lotophagifwe find ourselves at once within the ‘limits of that outer world, all statements concerning which were more or less mixed with fable. But in this case the existence on the north coast of Africa of a people who subsisted mainly on the fruit of the lotus-tree, was found in later times to be a well-established fact; and the sweetness of the fruit was supposed to have given rise to the fable of its making people forget their country.3 Some notion of this might easily have been gained from Phoenician navigators, nor is it impossible that even a Greek ship might have been really driven there, as Ulysses 9 dAAoi ire Kiiira [3609 “re weptyvoin‘rr'rovra Mo'iAeLav. Kai. Bopéiqg oirréwo'e, wapérrhayé'ev 8% Kvefipwv. Odyss. ix. 80-81. 1 First Menelaus, then Agamemnon, is described as being driven out of his course by violent squal ls while doubling Cape Malea (iii. 287; iv. 514). The same 1ncident is introduced in the fictitious narrative of Ulysses (xix. 187). Its evil repute in ancient times is attested by numerous passages; and in modern days will be familiar to all who have ever navigated the ZEgean. 2 51/6611 8’é1/v1'7ptap ipepd/wqv bkooizs (ilfiétLOLG'LV, miv'rou e'rr'ixGuéevr" afrrdp Sexa'ry en-éfimrev, yai'qs Aw'roipd-ywv, oi: 'r'éivdwov eISap Edovmv. lbid. ix. 82-84. 3 See Polybius (xii. 2), who de- scribes it from personal observation. SE01‘. 3. HoMERIo GEOGRAPHY. 53 t was supposed to be. And as he was driven off in the first instance from Cape Malea by a north wind, the natural inference would be that the land of the Lotus-eaters was somewhere away to the south Hence there is nothing improbable in the supposition, which has been adopted by almost all commenta- tors and geographers, that in this instance the fable had a foundation in fact, and that the Lotophagi were really situated] somewhere on the north coast of Africa. But we are not entitled to go farther, and assume that Homer had a definite idea of the position of the Lotus-eating people. Later writers either placed the Lotophagi on the coast between the Lesser and the Greater Syrtis,4 or restricted them more specially to the island of Meninxt/5 the beauty and fertility of which would appear to correspond to the poetical idea of the country of the Lotus-eaters, and where the Lotus shrub actually grows in great abundance.6 But it is most un- likely that the poet had anything more than a vague and general idea that the Lotophagi dwelt to the west of Libya, with the name of which, as we have seen, he was also acquainted. § 4. With the next step we plunge into complete uncertainty. \Ve are told only: “ From thence they sailed on, grieving in their hearts, and came to the land of the Cyclopes ;”7 whom the poet proceeds to describe as a lawless, cruel race, ignorant of agriculture and of all the other arts of life, dwelling in caves, but inhabiting a fertile land, which produced all kinds of grain without the need of cultivation. They were wholly unacquainted also with navigation, so that when Ulysses and his companions landed on a small island, opposite to the mainland, they found it full of wild goats which lived there flie Kiistenliincler cles Mittelmecres, 4 $<1y1aX,§110- The assumption of Volcker that it 5 Strabo, i. 2,§ 17. was only one day’s voyage from the 6 See Barth, ll'anderungen durch land of the Lotophagi is certainly not justified. The poet indeed tells us vol. i. pp. 259-265; Guérin, Voyage that they arrived at the Island of Arche’ologique clans la Re'gence cle Goats in the night (v. 143); but there Tunis, vol. i. pp. 203-207. is nothing to show that this was the 7 é'ueev 8% 'n'po're'pw nAéopxmdKaxfipez/mij-rop. fiTSt night after thPy left the LOtO- K'uKAai-n'wv 5’ é; yafav inrepcpuihwv, olequia- I phagi_ ‘9 ’ lK'Zil-Gléi. Odyss. ix. 105-107. 54 HISTORY or‘ ANCIENT GEOGRAPHY. can». III. ‘ ‘through his memorable adventures inthe cave of Polyphernus. 1": isouthern extremity or ice of Italy.9 unmolested. Here also the greater part of his ships remained ' in safety, while Ulysses, with the crew of one vessel, went _ n n , i That the land of the Cyclojies—wasR'Sicily, and that they\ dwelt on the slopes of Mount‘ Etna facing the sea, has so generally been believed, both in ancient and modern times, that it had become, as we know from the Cyclops of Euripides, as fixed an article of popular, faith in his day, as it was in that of VirgiLs Yet it is certain inn, th917§;i.§_-p0thing in the Ofilfiiellgléailgnihah.,.Qonclusion._ There is no indication “either of the distance or the direction of the voyage from the land of the Lotus-eaters thither: and it is scarcely necessary to add that neither the name of Sicily, nor that of ZEtna, is found in Homer. Nor is there any island adjoining this coast at all corresponding to that described as abounding in wild goats. To this it must be added, that Homer elsewhere (as we shall see) speaks of an island ealled Thrinakia, which has been almost universally identified with Sicily : and not only is there nothing to connect this with the land of the Cyclopes, but the two appear in the mind of the poet to have been wholly separate. These difficulties have led some writers in modern times to reject the view, which seems to have been universally adopted in antiquity, placing the Cyclopes on the east coast of Sicily, and to transfer them to the south-west coast of that island, or else to the opposite mainland, the Both hypotheses are equally devoid of any adequate support. All we can say is, that Homer} conceived thggQyclopes as dwelling in a moun- taino'us, ,IandlofWcQnsideIabIQ ,extent,f_g__sgnievwlier~e_ , to the west 8 The Cyclops of Euripides is full of allusions to the site of his cave among the rocks of jEtna (sec vv. 20, 56, 82, 101., &c.). In the .ZEneid its proximity is noticed at the very outset of the narrative. "Portus ab accessu ventorum imntotus, et mgens Ipse, sed horrificis juxta tonat [Etna ruinis.” iii. 570. -" Viilcker (I-Imnerz'sche Geographic, pp. 110-113) maintains the former the- ory; Ukert (Geographic, i. p. 23) the latter, though with a reasonable ex- pression of doubt. Forbiger and Buch- holz follow Vdlcker. 1 Not only does Homer expressly call it “The land of the Cyclopes,” like “the land of the Lotophagi," not the “ island"; as he terms the abodes of ZEolus, Circe, and Calypso; but he elsewhere tells us that the Pheeacians, SE01‘. 3. HOMERIC GEOGRAPHY. 55 oLGngece, and...ap.parently.--to~theenorth-"of the land of "the ,I'igtophagifi. .. .-/ _ § 5. The next stage in the progress is equally obscure and indefinite.’ Again “ they sailed on, grieving in their hearts;” grieving for the loss of their companions, though rejoicing to have themselves escaped from death.3 But not a word as to the direction or duration of the voyage. \Ve are only told that they arrived at the island of ZEolus, where that here dwelt in patriarchal state with his six sons and six daughters, in a floating island, which was girt all around with an impene- trable wall of brass, and with smooth rock rising up on high.4 He was appointed by Zeus to be the guardian of the winds: and being favourably disposed to Ulysses, after entertaining him hospitably for a month, he sent him forth on his voyage again, having tied up all the winds in a bag, which he com- mitted to his charge; except only thg west wind, Zephyrus, which was to send him on his way. With this favouriiidgfwirid they sailed for nine days and nine nights, and on the tenth day they were already within sight of their native land, when Ulysses fell asleep, his companions opened the bag, and all the winds rushing forth produced a storm, which drove the ships back all the way to the island of ]Eolus.5 There is perhaps no incident in the whole poem more glaringly fabulous than that just related. Yet several modern inquirers have attached importance to the statement that the voyage from the island to Ithaca was one of ten days’ duration with a west wind: and it must be admitted at least that it shows the marvellous island to have been situated in the poet’s mind at a considerable distance to the west of Ithaca.6 It was before migrating to Scheria, had dwelt real voyage; but of very questionable in the open country of Hypereia, near application in regard toafietitious one. to the Cyclopes, whose violence had 3 Odyss. ix. vv. 565-6. compelled them t0 I'QDJOVG. (Odyssey, ‘ AtoM'nz/ 8’ 6'9 vijo'ov dubc'lcoued" 51160. 8’ é'uaLev V11, 4-8,) Ai'mmg‘lmrorciSns, du'Aos cider/drown 0e02- 2 This last inference rests upon the AW" . . V,” min” 86, 76, MW MP. 7 . O ‘ . l . TrwmevL urlt" \ r 1 L/el'xg assumption that the nayigators, having XOILMEW, appnnov. N00,, 8 auasespwe been driven out of their course to the 'n-é'rpy], south, would try to rectify their posi- Ibld- X-1-4- tion by steering towards the north; a 5 Ibid. x. 14-55. _ legitimate inference in the case of a 6 Those who consider that every 56 HISTORY OF ANCIENT GEOGRAPHY. CHAP. III. almost as settled a conclusion among the ancients that the island of ZEolus was one of the remarkable group of volcanic islands‘ to the north of Sicily, to which they in consequence gave the name of the ZEolian Islands, as that the Cyclopes dwelt at the foot of Etna; though the two assumptions were in fact incompatible :7 and it was equally impossible to reconcile the position thus assigned to the magic isle of Eolus with the account of the voyage from thence to Ithaca, as given by the poet himself.8 It cannot indeed be denied that some of the Lipari Islands—especially Stromboli, the one selected by popular tradition in the time of Strabo and Pliny as the abode of the ruler of the winds 9—have something in their form, and in the §i16§pnesswith which they rise out of the middle of the “sea, ‘that may serve to'recall, if not to suggest, the idea of the island girt with a wall of brass and a smooth precipice of rock alLround. But even if it be- admitted that some exaggerated account of this appearance, as transmitted through Phoenician traders, may have given rise to one part of the fable; there can be no doubt that Homer had no real idea of its situation, and simply conceived it as lying a long way off, in the middle of the sea, to the west of Ithaca. Of its position with respect to Sicily or to the land of the Cyclopes, he had apparently no idea whatever. § 6. With the ill-fated return of the hero to the island of ZEolus begins a new series of adventures. That personage having refused to furnish any further assistance to one who was statement of the poet is to be construed the ZEolian Islands, without either literally and strictly, naturally point ‘ passing through the strait between out that the duration of the voyage Scylla and Gharybdis—which he did from the island of ZEolus to Ithaca is I not do until a much later period—or the same as that from Cape Malea to ; circumnavigating the whole of Sicily, the land of the Lotophagi; so that a most improbable assumption. But Ulysses would have exactly “made up P this difliculty, like all others connected his lee way.” I cannot believe that i with the vulgar attributions generally the “nine days and nine nights ” is in l adopted, was overlooked by almost all either case anything more than a vague '; ancient writers. and general expression. 8 See above, p. 39. 7 It is evident that had the Gyclopes ‘ 9 Strabo, vi. 2, § 6, 11, p. 276; Plin. dwelt at the foot of ZEtna, Ulysses l N. H. iii. 9, § 94. could not have passed from thence to , SEo'r.3. . HOMERIO GEOGRAPHY. . 57 evidently hated by the gods, the winds were unfavourable, and it was only after six days and nights of laborious rowing that Ulysses and his companions reached the civtywof Lamus, the king of the Laestrygones, which had a land-locked port, guarded at the entrance by two projecting rocks.1 Here again ' nothing is told us of the direction of the voyage: and the utter vagueness of the data for determining the place of abode of the Leepstrygones-a people of giants and cannibals, as obviouslyfabulous as the Cyclopes—is sufficiently proved by the fact that the prevalent opinion among the Greeks assigned them to Sicily,2 while the Romans transferred them to Forgrpia: on the coast of Campania.3 Both suggestions may bevsafely dismissed as utterly without foundation. Nor is there any clue to afford grounds for a more plausible conjecture.4 § 7. After the destruct'on of the greater part of his ships by the giant Leestrygones, Ulysses with one vessel only “sailed on ”-as usual without affording us any indication of the direc- tion of his course, or the length of his voyage—till he came to the island of ZEa, the abode of the goddess Circe, the sister of “ the wickedzfiilded ZEetes.”5 The purely mythical character ' of everything connected with Circe and her island, is evident, not only from the adventures encountered by Ulysses and his companions, but from her being styled a goddess, like Calypso, and from her connection with ZFietes, a being as clearly mytho- 1 Odyssey, x. vv. 80-90. Nothing : unlike as can well be to the description can be more distinct or graphic than in the Odyssey. On the other hand, the description of this haven, which is I was told by Colonel Mure that the evidently the original copied by Virgil little bay of Sapri, on the coast of in the first book of the ZEneid (vv. Calabria, not far from l‘olicastro, was 159-169). Yet the one is in all pro- g the very image of the port of the bability as purely a creation of the ' Laestrygonos as described by Homer. poet’s fancy as the other. Others have found a striking similarity 2 \Vhat is still more extraordinary, 5 in that of Balaklaval the current tradition placed them near E 4 Note D, p. 77. Leontini, in the interior of the island. 1 5 mi'rorcao'uyvrjm O/\06<1>pouo; Airj-rao. (Strabo, i. p. 20; Plin. N. H. iii. 8, z 0dyss..x.137, §89.) The epithet dikoo'q5pwv is applied also 3 Plin. iii. 5, § 59. The same tra- to Atlas, as the father of Calypso dition is frequently alluded to by the ' (Odyss. i. 52); in both cases it appears Roman poets. Yet the beautiful bay ' to indicate a being of supernatural of Formiee (the Gulf of Gaéta) is as ' powers, to be dreaded as such. 58 nrsronr or ANCIENT GEOGRAPHY. onnp. III. logical as Atlas. And of all the attempts to give a “local habitation” to the legends related in the Odyssey, none is more absurd than that which attached the name to the Circeian promontory, on the coast of Italy, which is not an island at all.6 Whether the idea arose from the identification of the magician goddess with some local divinity of similar attributes, or from a mere casual resemblance of name, we are wholly unable to decide.7 Equally impossible is it to attempt any other determination of an island of which nothing is told us that is not on the face of it purely fabulous. We can only assume that, as the poet represents the abode of Circe as the place from whence Ulysses took his departure for the gloomy T ‘realms of Hades, he must have conceived it as situated towards “ the region of darkness ;” that is, the west or north-west.8 It is certainly not worth while to enquire what geo- graphical idea the poet formed in his own mind of this visit to the regions of Hades, or to attempt to define the locality of thenCimmerians,~--whom , he -describesnas wrapped up in mist and_,,.cloud,- _..and._.,_.Qrershedorved- by perpetual night, never beholdingqthe ‘sun either as liewawsfcends the heavenanonashe ,,___\__m. ...__._~--—--—---—: — declines towards setting?‘ They dwelt upon the Ocean 6 Yet we find Pliny seriously writ- 8 See Note E, p. 78. The only ing: “ Circeii quondam insula immense quidem mari circumdata (ut creditur Homero), at nunc planitie” (H. N. iii. 5, § 57). The same statement was already made at a much earlier period by Theophrastus (Hist. Plant. v. 8, § 3), but he judiciously reports it only as the tradition of the natives; while Pliny gravely infers that all the inter- mediate and surrounding land was a recent addition. 7 It is certain, however, that tra- ditions connected with Circe had at- tached themselves to Latium at a very early period, as we find her introduced as the mother of Agrius and Latinas in a passage of Hesiod (Theogon. v. 1011-1013), which, though it is almost certainly a subsequent addition to the poem, cannot be referred to a late date. thing we are told of it is that it was surrounded by a boundless sea :— Vijo'ov, Tijv 'n'epi. ‘Ii-61110; drreipvro; e'cr'redioivw'rat. Odyss. x. 195. ii) 5’ e’; wec'pad’ 't'xave fiadvfipéov 'Qxeavo'io. 5110a as Ktmaepiwv dudpdw 517mm; 76 1115M; 're, ipépr. Kai. veq‘ie'An xencakzinnevot - 01366’ 'n'o'r’ ai’i'roiis w’jémos ¢aé6wv Karade'pxe'rat dK'rL'i/eo'a'w, 000’ omir’ div areixno't 1rpbs oz’lpavbv do're- pdevra, '_ 0170’ 31" 5.11 dill e’1ri. yacav d'rr' or’lpavéfiev n'poerpdrm'rar, 6AA’ é'n'i. viii: (Moi) TéTfl-TG-L Suhoio'i. fipo-ro'iow. Ibid. xi. 13-19 It is remarkable that Circe, when she was indicating to Ulysses his route to the abode of Hades (x. 506-510) says nothing of the Cimmerians, with whom also the hero and 1118 companions have no intercourse; they are only men- tioned in passing, in the lines above cited. 9 $120123. HOMERIC GEOGRAPHY. 59 stream, but'whether on the nearer side, or beyond its waters, is a point upon which the ancient interpreters were divided, and the language of the poet appears ambiguous. One thing only is clear; the voyagers had to navigate the Ocean stream itself, into which they passed, and from which again they passed into “the sea,” on their return, without any difficulty, or. any explanation.1 The supposition that they must have passed through a strait, like that which really leads from the Ocean into the Mediterranean, rests upon no foundation at all, and belongs to an order of ideas wholly different from that which was present to the mind of the poet.2 § 8. From the island of Circe, Ulysses set out on his home- ward voyage, after having been duly instructed by the goddess’ with regard to the dangers that awaited him, and beset his course. The first of these was that of the Sirens, who seduced all passers-by with the sweetness of their songsi’fAfter this came the choice between two routes; the one passing by the ___Pnlanctze, rocks of the most formidable character, from the terrific‘ surf that broke at their foot, with which is associated, though in a very obscure manner, the mention of fire and smoke.3 Their aspect was indeed so alarming, that Ulysses and his companions preferred passing wide of them, and following the other route, although this involved the necessity of passing between S/cgyllaand Charybd-is, of the perils of which the here had been sufficiently warned by Circe. They however succeeded in accomplishing the passage with the loss 1 xi. 9-13, 639—640; xii. 1—2. The passage into the Ocean is very ob- scurely indicated in the first of these passages, as well as in the preliminary directions given by Circe (l. 0.), but the contrast at their exit is more clearly marked: aim-zip é-n'ei. 770704.].020 Ai'rrev 660v 'Qxeavo’io vrjfig, 6.1a. 8’ Exam xilpa Oakalo'mjq eupmrdpoto. Odyss. xii. 1, 2. '~’ This appears to be the error in which Eratosthcnes and other ancient investigators found themselves, who transferred the wanderings of Ulysses . into the Ocean, that they could not divest themselves of their more scien- tific ideas on the subject, and were con- sequently obliged to explain how he got back again into the Mediterranean, a difficulty that evidently never pre- sented itself to the mind of the poet. The same fallacy pervades still more strongly the views of Mr. Gladstone and the marvellous map in which he has embodied his conception of this part of the Odyssey. 3 See Note F, p. 79. 60 HISTORY or’ ANCIENT GEOGRAPHY. can». III. of only six of their number, and arrived without further obstacle at the sacred island of Thrinakia, which was occupied by the consecrated herds of the Sun ~(Hyperion), under the guardianship of two nymphs, his daughters. These herds Ulysses had been especially warned not to touch, but being detained in the island a whole month by contrary winds,4 his companions yielded to the pressure of hunger, and slaughtered and ate some of the sacred cattle: an act of impiety, which was soon punished when they again set out on their voyage, their ship being caught in a violent storm, and shivered to pieces by a thunderbolt. Ulysses alone contrived to float upon a portion of the wreck, and was again driven through the strait, where he narrowly escaped the dangers of Charybdis; after which he was carried along for nine days and nights, till he arrived at the island of _Qa~lyqpsuo.gw No portion of the Wanderings of Ulysses is more familiar to all readers than that just related. The dangers of Scylla and Oharybdis are become as proverbial as the cup of Circe and the songs ‘of the Sirens? The locality assigned to them by tradition was as definitely established as in most other cases; but in this instance alone could it be shown to be based upon physical phenomena, which had a real existence, and which might easily have given rise to exaggerated fables such as those related by the poet. The navigation of the Straits of Messina lisllsiléelllhsxapresdiiéil iééllhifilfillélble difll-CuliiQSIand dangers to the mariners of early days : and these dangers were in part of a character wholly unknown elsewhere, arising from the complicated action of the tides and currents produced by the meeting of the waters of two great seas in a narrow and crooked ‘ was the violence of these winds, as much vented them from sailing. When they at last set out with fair weather, and are out of sight of land, they are caught by a sudden gale from tho 4 In this instance the winds are spe- cified with unusual definitencss: the south wind (Notus) blew for a whole month without ceasing, nor was there ever any wind, except from the east or west (Zephyrus, ib. 408). and, after south (Eurus or Notus) (xii. 325-326). that abates, there arises a south wind But it appears from verse 400 that it (Notus), by which Ulysses is driven back to Gharybdis (ib. 427). as their direction, if not more, that pre- Seer. HOMERIC GEOGRAPHY. 61 channel.5 Nothing like the eddies resulting from these causes is to be found elsewhere in the Mediterranean, as the Straits are the only place where the tides are felt with such strength and regularity: - a phenomenon frequently noticed by the ancients, and which they correctly ‘sought to bring into con- nection with the Homeric accounts of Charybdis.6 It may be added that anything in the nature of a whirlpool appears in all ages to have been peculiarly subject to exaggeration, and the fables related of the Maelstrom in Norway, and even of Corrievrechan in the Hebrides, are scarcely inferior to those current among the Greeks concerning Charybdis.7 We may, therefore, safely assume that in this case the descriptions of the poet—exaggerated and extravagant as they undoubtedly are—rested upon a certain basis of fact: and that in asso- ciating the dangers of Scylla and Charybdis with the Straits AQLMessi-na, the tradition of later days was not a mere random attribution, like that of the islands of Circe and Calypso, but was derived, like that of the Lotophagi, from real information, however vague, and disfigured by fabulous additions. But it would be wholly unwarranted to assume that because Homer had really heard some vague account of Scylla and Charybdis, he had therefore any definite idea of their situa- tion, and knew that the strait between them was that which separated Italy and Sicily. He appears to have placed them 5 The real nature of these dangers is stated with great clearness by'Thu- cydides (iv. 24): é'orr 5% a nope/abs‘ i7 ,ae'ragil 'Pn'yr'ov Ootkao'a'a Kai Meo’o’rjvns, firrep Bpaxri'ra'rov Eucekfa rijs 'hrrefpov drre'xer, Kai E'o'rw r‘) XQIPUBOLS KAnOeTo'a difficulties of the navigation are such as might reasonably have given rise to the reputation of its dangerous cha- racter. ‘ 6 Strabo, i. 2, § 36, pp. 43-44. - 7 In both these cases the danger is, 'roz'iro, i‘; ’05vo'o'ei1$ Aé'ye'rar Olarrkeiio'ar' however, a reality, though, as in the as oral/6mm 8% ml e’lc ,ae'ydAwu weAa- case of Charybdis, varying much with 76w, r05 re Tvpo'nvucofi Hal r017 Eurem- the circumstances of Wind and tide. x08, e’o'rrfrr'rovo'a '7) Gdaaaaa e’s aim-b, mi See the description of Corrievrechan in riooé'o‘ns 050a, eilcd'rws XaAevr»); e’vonfatin. Anderson’s Guide to the Highlands The present phenomena of the Strait, of Scotland, p. 76, 3rd edit.; and an and the action of the tides and currents excellent account of the Maelstriim, by within them, are fully described by Dr. Charlton, cited in Murray’s Hand- Admiral Smyth (The Mediterranean, book for Norway, 3rd edit. 1871. pp. 178-182), who admits that the 62 _ HISTORY or ANCIENT GEOGRAPHY. can). III. somewhere in the “far west,” like the islands of Eolus and Circe; to enter into any more precise examination of the locality was as foreign to the mind of the poet as to that of his hearers. Hence it would be idle to take this identification as a fixed point from which to attempt to determine the others that are more or less connected with it. All these must indeed be left in a state of utter uncertainty. The name of’Thrinakia —the sacred island of the Sun—was generally regarded by the ancients as identical with that of Trinacria, which, according to a tradition generally received, had been the ancient appella- tion of Sicily ; 8 and they therefore did not hesitate to suppose that the island meant by the poet was Sicily itself. But to say that the voyagers, after passing through the Straits between Scylla and Charybdis, came to the island of Thrinakia, would on this supposition be as strange a mode of expression as to say that a modern captain, after sailing through the Straits of Dover arrived at the island of England! And, what is more conclusive, Thrinakia is distinctly described as a small island, devoid of inhabitants, and having no natural resources, except the sacred flocks and herds, so that the companions of Ulysses were in actual danger of starvation, when they had recourse to the sacrilege.9 Hence it seems clear that, whatever may be the explanation of the name, the island of Thrinakia had no connection with the large and fertile island of Sicily. § 9. Of the island of Calypso, to which the poet gives the name of Ogygia, it may safely be asserted that it is altogether as 8 This tradition is referred to as an established fact by Thucydides (vi. 2), and was followed by Strabo and all later writers. Yet it may well be doubted whether the name of Trina- cria was ever in use, except among the Greek poets and logographers. It could certainly never have been a na- tional appellation, and it is most un- likely that such a name as “ the three- cornered ” was ever in use, even among mariners, for so large an island as Sicily, the triangular shape of which could only have been known when it came to be much visited, and its coasts fully explored. Timaeus, it appears, maintained that its original name was Thrinakia (ap. Schol. Apollon. iv. 965), and it seems very probable that both names were in reality mere inventions, in order to connect it with the Ho- meric Thrinakia. The name of _ Si- cania for Sicily is found elsewhere in the Odyssey, but only in the last book (xxiv. 307), which is generally re- garded as of later date than the rest of the poem. 9 Odyss. xii. 325-373. SE01‘. 3. HOMERIC GEOGRAPHY. 6 3 vague and visionary as that of Circe. It was situated “in the centre of the sea;”1 far out of the way of all resort, and unvisited alike by deities and mortal men? Ulysses is repre- sented as arriving there after drifting for nine days and nine nights (the ever recurring number) on a fragment of wreck from Charybdis.3 On the other hand, it is apparently more than twice that distance from Phaeacia; for on his departure he is represented as sailing for seventeen days with a favour- able breeze, and on the eighteenth coming in sight of “the shadowy mountains of the Phaeacian land.” 4 This much only may be affirmed with certainty, that the poet conceived the Pheeact'ans. - island of Calypse-as- situated a long way off to the west of the For he distinctly tells us that the nymph had warned the hero, in steering his course across the vast expanse of open sea that he had to traverse, “ to keep the constellation of the Great Bear steadily on his left hand ;” 5 that is to say, to steer from west to east. This is the only allusion in either poem to steering by the stars; and it is the most definite indi- cation to be found in the Odyssey of the course pursued, or supposed to be pursued, by the wandering hero. Yet, strange to say, it has hardly been accepted in its plain and obvious I mic-(p év dadupurg, 86L 1" biutahos éo'n Oomdo'mys. Odyss. i. 50. The use of 5,u.q>a7\ds in this sense as the centre or middle point of the sea, is obviously derived from its application as the boss in the centre of a round shield. In like manner in later times Delphi was called “the navel of the earth” (5,u¢a2\es 713$ 713$, Strabo, iX. p. 419). 2 Ibid. v. 101, 176. 3 He was driven by a south wind (Notus) to Charybdis. If we are en- titled to assume that he afterwards continued the same course, we must place the island of Calypso to the north of the Strait; and this has been generally done by the modern writers who have attempted to arrange the localities of the Odyssey (see Viilcker, p. 121), though wholly opposed to the popular tradition which identified it with Gaulos, or Malta. 4 ex'rwlcatdexoirp 5' é¢aiin7 5pm. a'méev'ra, 'yat'r); (Paujxwv. Ibid. v. 280. It may be observed that Homer always uses the expression “the land of the Phzeacians.” He never calls it an island, like those of ZEolus, or Circe, or Calypso, which shows that he con- sldered it as a tract of considerable extent. This does not, however, ex- clude the idea of its being a large island like Sicily or Crete. 5 15711 ydp dfi ptw o’ivwye KaAvd/u'), 620. 965.001’, 1rov1'orropevépevac. én" dpia'repa xetpbg E'Xov'm. Ibid. v. 277. The lines immediately preceding con- tain the description of the constella- tion of the Great Bear, which has been already quoted (note to p. 34). 64 HISTORY OE ANCIENT GEOGRAPHY. CHAP. III. sense by any of the geographical commentators on the poem; one has understood it as indicating a course to the north-east, another to the south—east, in order to accord with their precon- ceived hypotheses concerning the relative positions of Ogygia and Scheria.6 It seems much more simple to understand the words as pointing only to the broad fact that Ulysses was returning from the far west, and that his voyage must there- fore have had a general direction from west to east.7 The unusual length of the voyage across the open sea would render it especially necessary to steer by watching the stars, and this may account for the practice being mentioned in this instance only. § 10. We are now arrived at the last stage of the protracted wanderings of Ulysses. It was a predetermined arrangement of the gods (as the poet tells us), that he should not be con- ducted to his native home either by gods or men, but by the @eaeianga semi-divine people, who partook of the qualities of both.8 Hence, when he was at length permitted by Calypso to quit her enchanted island, his voyage was directed, not straight to Ithaca, but mthe land of the Phzeacians, where he ultimately landm'safety, notwithstanding the storm raised by Poseidon. Here, therefore, the difficulties of the hero’s voyage wewtwan end : and here also, if we adopt the popular view, there is an end of all geographical difficul- ties in connection with it. That the island of Scheria was the same that was afterwards known as was the established belief in ancient times ;9 and has continued to be as generally Lreceived in modern days: and though it is admitted that the 6 See Note G, p, 81, 8 Qawixwv é; ya'iav, oi. d'yxieeot yeyéaaw. 7 I say a general direction, because Odyss. v. 35. there is no ground to believe that in In the same manner Alcinous speaks this, any more than in other passages in of the Phaeaoians as favoured by the the Odyssey, did the poet mean to de- gods ;-- signate any precise quarter of the . . . érrei (npw'w éyywev einév, heavens, 01' quadrant Of the 0112018, (130175;) Kiikhwrre'; 're Kai. o'iypta ¢5Aa_I‘vyoi_vrm1/. with the accuracy of a modern manual Ibld- V"- 206- of sailing directions, as has been too 9 We find it already referred to as often assumed by the commentators. such by Thucydides (i. 25). SECT. 3. HOMERIO GEOGRAPHY. 6 5 ‘ people of the Phaaacians had disappeared before historical times—a conclusion, indeed, not obscurely intimated by the poet himself—it is still supposed that the place of their abode can be determined without a doubt, and that local peculiarities which occur in the poet’s description‘ may be still traced in the island of Corfu.x Even in ancient times, however, there were to be found a few critics who raised the voice of scepticism with regard to this popular attribution, as well as to most similar identifica- tions;2 and attention has been repeatedly drawn by modern scholars to the strong tinge offmpythical colouring which is east around everything connected with the Phaeacians.3 Not only were they closely connected with the gods, but they were in the habit of receiving frequent visits from them, and associat- ing with them on equal terms.‘ On the other hand, they dwelt afar off, out of the way of all mortal visitors;6 their skill in navigation was of a distinctly supernatural character ;6 and the circumstances of the voyage when they conduct 1 This is stated particularly strongly 'ypappa'mcfis ' bs 'r'hv [rev P1108011 KaAv- by the Edinburgh Reviewer (p. 522), ipofis vfiadu 4mm, Thu 5% Kdplcvpav Exe- but very few of the visitors to Corfu pram—Strabo, vii. 3, § 6, p. 299. share his enthusiasm on the subject. 3 This has been most ably and fully Colonel Mure is sceptical. Dr. Schlie- worked out by Welcker in his memoir, mann, with his usual enthusiasm, finds Die Homerischen P/zéiaken and die In- everything in precise accordance with seln der SeZz'gen (reprinted in his Homer. The result of my own obser- Kleino Schriften, vol. ii.), but the vation is entirely in accordance with same view has been adopted by several that of Colonel Mure. Without deny- other writers in recent times, Nitzsch, ing that some resemblance may be Bursian, &c. The Pheeacians are, in- traced to the Homeric description, if deed, placed by Homer himself on the we assume the identity, I certainly can same level with the Cyclopes and the see nothing of a character that would Giants (Odyssey, vii. 206), and all in itself lead one to such a conclusion. three races must be regarded as equally 2 Eratosthenes appears to have led mythical. (See Preller, Grieehz'sche the way, and was followed by Apollo- Mythologz'e, vol. i. pp. 387—395.) dorus, who reproached Callimachus ‘‘ Odyssey, vii. 201-206. with having identified the island of 5 Ibid. vi. 204*205. Gaudos with that of Calypso, and 001'- 6 Ibid. viii. 556-564. Their ships cym with Schema; a thing excusable, had no need of steersmen or rudders, as he considered, in the vulgar, but but knew of their own accord where not in a writer like Callimachus, who they were to go, and held their way claimed to be a man of learning (m2 wrapped in mist and darkness, per- 7013‘ prev iiAAois av'y'yva'iimv six/at, KaMu- forming even the longest voyages ,uaixq: 8% ,u1‘7 mix/v, MGTGFOlOU/LEIX/(p 76 ;v withinasingle day. VOL. I. F 66 HISTORY OF ANCIENT GEOGRAPHY. carp. III. Ulysses to his native land—the nocturnal voyage and the landing him while fast asleep—seem as if expressly designed to mask the transition from the land of fable in Phaeacia to the realities of Ithaca. In the same manner the prophecy as to their future extermination and the destruction of their city,7 is calculated to transport them into the realm of the ideal, and to show that they were to be regarded as a mere poetical creation, not a really existing people. As usual, we are wholly at a loss to explain the reasons that led the Greeks in early times, notwithstanding all these peculiar characteristics, to treat the Phaeacians as ‘a historical people, and to identify the Homeric Scheria with the island of Corcyra, which was situated within less thTaIfi'MIllO miles of Ithaca, and directly opposite to the coast of Thesprotia, which is repeatedly mentioned in the Odyssey, but never in connection with the Phmacians. But whatever may have been the origin of this tradition, it was strongly encouraged and perpetuated by the Oorcyraeans them- selves, after the establishment of the Corinthian colony in that island (about 13.0. 735). Having speedily risen to a prominent position as daring and successful navigators, they asserted their claim to be recognized as the representatives, if not the lineal descendants of the Phaeacians of Homer.8 N o clue is afforded us in the Odyssey to the position of Scheria with respect to Ithaca: indeed, the passage between the two appears (as already pointed out) to be intentionally veiled in mystery. The only statement that brings the island of the Phaeacians into connection with any known point in geography, is that of Euboea being the most distant land to which they had ever had occasion to convey a wandering guest :9 an expression which certainly appears to point to their being situated on the western side of Greece. We must, therefore, be content to banish the kindly and hospitable Pheeacians, as well as the barbarous Cyclopes and 7 Odyssey, xiii. 172-184. cydides in the passage already cited 8 This is expressly stated by Thu- (i. 25). 9 Ibid. vii. 322. SECT. s. HOMERIO GEOGRAPHY. 67 Laestrygones, to that outer zone of the Homeric world, in which everything was still shrouded in a veil of marvel and mystery. We can as little explain in the one case as in the other, what gave rise to the original legend that has been amplified by the creative genius of the poet into the form with which we are all familiar. The attempt to find a substantial basis of reality, either for these, or for the other legendary tales introduced into the Odyssey, is in almost all instances utterly futile, and it is still more idle to endeavour to arrange them in accordance with any preconceived geographical system, or assign a definite and rigorous meaning to expressions which were doubtless employed by the poet in a mere vague and general—that is to say, in a poet/teal sense. In the very few cases in which it is probable that some slender basis of fact did really under- lie the fictions that he presents to us—thge ,Lotophagiandjthe Stigait ‘of ,Scylla andCharybdis are the only instances‘ in which I its existencecan plausibly traced—it is clear that any such notices were accompanied by no definite indications of locality, and there is no reason to suppose that Homer attached any more correct geographical notions to the one set of legends than to the other. ‘f § 11. In passing from the land of the Phaeacians to Ithaca, we might fairly presume that we were passing also from the ideal to the real-—from the unknown to the known. It might, indeed, be naturally supposed that the poet of the Odyssey would be as familiar with the island home of his hero and the geography of the group of islands that surround it, as the poet of the Iliad undoubtedly was with that of the Troad and the adjacent islands. It is strange, therefore, to find, that instead of his notices of them being marked by the same clear and definite conception of their position and geogra- phical relations, which is displayed in many other instances, we are met by difi‘iculties which have been a stumbling-block to the commentators in all ages, and have never yet received a satisfactory solution. Ithaca itself, is, indeed, correctly de- scribed as pre-eminently rugged and rocky, abounding in _.___,. F 2 68 HISTORY or ANCIENT GEOGRAPHY. one». III. goats, but altogether unsuited for rearing or keeping horses.1 In it was a mountain, called N eriton, “ waving with foliage ;” an epithet wholly inappropriate at the present day, but which may probably have been applicable in early times. But in the only passage in which the poet undertakes to describe its position with respect to the surrounding islands, “Dulichium and Same and the wooded Zacynthus,” which lie, as he tells us, all round it,'very close together,.he adds: “ Ithaca itself lies Alo,w,sthefarthest 0s‘ Oran, in the sea: towards the west, and the Qtllelisiaway towards the dawn and the rising of the sun.”2 This certainly appears to be the obvious sense of the words, though, as usual, they have been distorted by the grammarians and commentators both in ancient and modern times; in the vain hope to reconcile them with the real circumstances of the case. Ithaca certainly is not a low island; as the poet himself tells us, it contains a conspicuous mountain, or rather two,3 and instead of lying to the west of the others of the ‘same group and afar off, it is situated to the north-east of Cephallenia, between the larger island and the mainland, but almost close to the former, from which it is separated only by a narrow strait. The only real solution of the difficulty appears to be, to admit thatgljiomer was not personally ac- quainted with the group of islands in question, and that, though familiar with their names, and with some of their leading natural characteristics, he had an indistinct and 1 Hence Telemachus declines the offered present of Menelaus of horses and a chariot, as wholly useless to him. év 8"190ixy 051" Zip’ 8p.v. . of beginning with the Pillars of Hercules as the farthest limit of the known world, he started from home, and beginning apparently at the Hellespont, described first in detail the shores of Greece and the European coast of the ZEgean; thence proceeding westwards by the Adriatic, Italy, and Ligu- ria, to Spain and Tartessus; then returning again to his starting-point in order to describe Thrace, Scythia, and the north coast of the Euxine, as far as the Caucasus. In like manner in his second book he appears to have begun the description of Asia from the Hellespont, and proceeded along the south coast of the Euxine to the Caucasus; then returning to the Hellespont again, and following the shores of the YEgean and the Mediterranean to Syria, Egypt, and Libya. The notices of the Medes, Parthians, Persians, and Indians, were perhaps annexed to those of the Colchians and other nations adjoining the Euxine. But this is very uncertain. § 5. So far as we can gather from the imperfect remains that have been transmitted 9 to us, the geographical knowledge of Hecataeus was very much what might have been anticipated‘ from the extent and distribution of the, Greek, ‘colonies and - v"settleIlilellts. He was well acquainted not only with the shores of the ZEgean and Ionian Seas, and with those of Southern Italy and Sicily, but with the eastern coast of the Adriatic, where he enumerates various obscure tribes of the Illyrians and Liburnians, with which the Greek colonies of Epidamnus and Apollonia had probably brought them in connection.1 Near the head of that sea he placed the Istri, “a people on the Ionian Gulf”—-but without any allusion to their gene- rally-received connection with the river Ister—and the city of Adria, with a river of the same name, which he describes as situated in a region of ' surpassing fertility.2 In Southern Italy, or Magna Grwcia as it was called in later times, he enumerates a crowd of names, including not only Greek settlements, but many towns of the CEnotrians, or native . 2 Note A, p. 153. 1 Fr. 60-69. 2 Fr. 58. SECT. 1. HECA‘I‘ZEUS. I tribes in the interior 3-——for the most part otherwise unknown, having probably disappeared during the revolutions that those countries subsequently underwent, which brought about a complete change in the population. In the Tyrrhenian Sea he was not only familiar with Corsica and Sardinia, but mentioned Elba under the name of ZEthale, an island of the Tyrrhenians, already celebrated for its mines of iron,4 as well as the less important island of Capreae. Of the towns in the interior he mentioned Capua and Nola, both of which were in the immediate neighbourhood of the Greek colonies in Campania, and were doubtless in constant commu- nication with them. But no reference is found in the existing fragments to any of the cities on the mainland of Italy north of this; and the whole coast of Western Italy, occupied by the Latins and Campanians, is a blank.5 It may be con- sidered certain that he did not mention the name of Rome, otherwise so important a testimony could not have been emitted by Dionysius of Halicarnassus. The existence of so considerable a gap in this part of his geography is the more remarkable, as the next place we find mentioned is the insignificant seaport of Monoecus (Monaco), on the coast of Liguria.6 It was. probably frequented, if not already occupied, by the Massilians; and it was doubtless through the same channel that Hecataeus had derived his knowledge of N arbo, which he terms a Gaulish city and emporium, thus indicating that it was already known as a centre of trade with the interior.7 The adjoining coast was apparently occupied by the Helisycians, who were noticed by Hecataeus as a Ligurian tribe.8 A number of names, many of them obscure or otherwise unknown, are mentioned by him on the east coast of Spain, and in the neighbourhood of Tartessus and the Pillars of Hercules: a circumstance which appears to indicate that the 3 Fr. 30-40. ‘ Fr. 25. 5 See Note B, p. 153, 6 Fr. 23. 7 Fr. 19. 8 Fr. 20. I40 HISTORY OF ANCIENT GEOGRAPHY. CHAP. V. Massilians carried on more trade, and held‘ more intercourse with these countries, than was the ease in later times, when they were almost entirely driven out by the Carthaginians.9 But no trace appears of any knowledge of the Atlantic shores of Spain. Even the name of Gadeira, or Gades, does not seem to have been mentioned by him. It was, perhaps, con- founded by the Greeks with the city of Tartessus, a name which was in these early days employed very vaguely.1 § 6. It is almost certain that Hecataeus had no knowledge whatever of the western and northern shores of Europe; and if, as appears probable, he assumed the existence of a continuous ocean in that direction, this was merely an inference from the established notion, so deeply implanted in the Greek mind, that the whole world was surrounded by a circumfluent ocean.2 Of the countries north and west of the Euxine, on the other hand, he had certainly considerable means of acquiring infor- mation through the Milesian colonies that were scattered all around its shores. But we have hardly the means of judging how far this information was actually embodied in his work. The citations from this part of it are few and scanty, and for the most part contain only the names of obscure or unknown Scythian tribes; but it seems impossible that he should have omitted to mention the numerous and flourishing cities on the shores of the Euxine,3 more particularly as he is cited as an cules in this quarter, and denied that 9 Several of these are placed among Geryones and the island of Erythea the Mastieni, whom he described as a nation adjoining the Pillars of Her- cules (Fr. 6) : their name is again inci- dentally mentioned by Polybius (iii. 33), but appears to have been lost, as an ethnic appellation, after the Roman conquest, and is not found in any of the later geographers. Its occurrence in Polybius is, however, a strong con- firmation of the accuracy of Hecataeus. 1 Gadeira is, however, mentioned by Herodotus (iv. 8), who terms it a city on the Ocean beyond the Pillars of Hercules. But it appears that Heca- taeus rejected altogether the ordinarily received account of the labour of Her- I had anything to do with Spain (Arrian, Anal). ii. 16). 2 The statement of Pliny (Hist. Nat. iv. 13, §94), that the northern ocean was called “ Amalchian ” by Hecatzeus, probably refers to Hecataeus of Abdem, not to our author. See Note 7, p. 148. 3 The emission of all such names in the existing fragments is a strong instance how little we can rely upon any negative evidence derived from such imperfect sources of information as we possess. Even on the south coast of the Euxine, which Klausen and Colonel Mure suppose him to have SECT. 1. _ HEOATZEUS. r41 authority for the small town of Carcinitis, at the head of the gulf of the same name,4 which is mentioned also by Hero- dotus,5 but not noticed by any later geographer. Of the Scythian tribes of the interior he mentioned the Melan- chlaeni,6 who are noticed also by Herodotus; the Dandarii,7 a tribe adjoining the Caucasus, whose name is still found in Strabo; and the Issedones, a name that, as we have seen, had been already rendered familiar to the Greeks by Aristeas of Proconnesus.8 Even these few names are sufiicient to show that his knowledge was not confined to the nations bordering on the coast, but that he had collected at least a certain amount of information concerning the remoter tribes of the interior. § 7. In Asia also he was not only well acquainted with the Colchians, the Chalybes, the Moschi, and other barbarian tribes in the neighbourhood of the Euxine, but was familiar also with the name of the Matieni, a people of the interior, adjoining the Armenians;9 as well as with that of the river Araxes, which flows into the Caspian Sea.1 This sea itself he described, under the name of the Hyrcanian Sea, as sur- rounded by lofty mountains, which were covered by dense forests} a statement which is true only of a part of its shores. In its neighbourhood'he placed the Parthians, and to the east of them the Chorasmians,3 two‘ well-known names, which here appear for the first time. But it must not be assumed that he was acquainted with the true position of the latter people, beyond the Caspian towards the east, or that he had any true idea of the configuration of that inland sea. visited in person, the citations refer I tween the two writers in regard to the principally to obscure barbarian tribes, f limits assumed for the two continents. while all the more important Greek See p. 147. 9 Fr. 189. These Matieni were sepa- colonies are omitted. 4 Fr. 153. rated from the Phrygians by the river 5 Herodot. iv. 99. Halys in the upper part of its course 6 Fr. 154. 7' Fr. 161. (Herodot. i. 72). They must not be i | l l 8 Fr. 168. It is to be observed that I confounded with the more powerful Hecataeus placed the Issedones in Asia, ! nation of the same name that dwelt east while Herodotus (iv. 13, 25‘) included i of the Tigris. them in Europe. But this discrepancy 1 Fr. 170. 2 Fr. 172. evidently arose from the difference be- 3 Fr. 173. I42 HISTORY OF ANCIENT GEOGRAPHY. CHAP. V. It is more remarkable that he appears to have collected some information, not altogether untrustworthy, concerning India—the name of which, as well as that of the river Indus, is for the first time found in the extant fragments of his work!‘ He mentions, indeed, several names of tribes and cities in that country, among which the Gandarii are known to us also from Herodotus, and appear toihave occupied the country about the upper Indus and the valley of Cabul.5 Caspapyrus, or Caspa- tyrus, their capital city, the name of which was also known to Hecataeus, was situated on the river itself; and it was from thence that, according to Herodotus, Scylax of Caryanda embarked on his expedition to descend the Indus.6 It is not improbable that some account of that voyage—conducted as it was by a Greek of Asia Minor—might have already reached Hecatmus, and been one of his sources of information con- cerning these remote countries. Of the other tribes' men- tioned by him the Calatians are also noticed by Herodotus, but we have no clue to their position. The Opians, as he tells us, dwelt on the river Indus; 7 their name is otherwise un- known, but is perhaps preserved in that of Alexandria Opiane, the city founded by Alexander at the foot of the Indian Cau- casus.8 Argante, “an Indian city,” the name of which is also cited from our author,9 is wholly unknown. Meagre and scanty as are these earliest notices of India, we are almost surprised to find that so much was known to Hecataeus, when we remember that the Indian ‘provinces were first annexed to the Persian Empire by his immediate contemporary Darius. But the India of Hecataeus, like the Persian dominion, was doubtless confined exclusively to the provinces west of the Indus, and did not extend beyond that river. 4 Fr. 174-178. 8 See Chap. XII. 5 F1‘. 178. Herodot. iii. 9], vii. 66. 9 F1‘. 176. ’Ap'ydz/'r17, miMs ’Iz/8fas, Herodotus, however, did not consider as ‘Ema-alas. Steph. Byzant. s. v. them as being properly an Indian This is a specimen of the class of nation. notices, out ‘of which we are to at- 6 Herodot. iv. 44. See Chap. VI. § 1. tempt to reproduce the lost work of 7 Fr. 175. Hecataeus! SE01“. 1. HECATZEUS. 143 Of his general knowledge of the interior of Asia we have hardly the means of judging, very few citations having been preserved to us from this part of his work. But there can be no doubt that he had a general acquaintance, by name at least, with all the provinces of the Persian Empire. Herodotus represents him as enumerating to the assembled Ionians (in order to dissuade them from their intended revolt) “ all the nations that were subject to the Persian king; ” 1 and it is certain that such knowledge must have been readily attainable in his day. But what amount of information he possessed concerning them, or what ideas he had formed of their relative geographical position, we are unable to judge. He appears, however, to have had at least a vague notion of the existence of the Persian Gulf;2 and he mentioned the name of the Myci, or Mycians,3 a people noticed also by Herodotus, and who probably dwelt in the neighbourhood of the Erythraean Sea. It is remarkable that no reference is made to any mention by him of Babylon, or of any of the great cities in the valley of the Euphrates and Tigris (except Sitace, the capital of the district afterwards called Sittacene 4), a circumstance which certainly tends strongly to negative the possibility of his having himself extended his travels into those countries, so fullof interest to the Greeks, and which had been rendered comparatively easy of access in his day, from the union of all WVestern Asia under the Great King. § 8. ‘Egypt appears to have occupied a considerable place in his worlaa‘fidgto have been described in unusual detail. We know with certainty that Hecataeus had not only himself visited the country, but had ascended the Nile as far as Thebes, and there conversed with the priests of Ammon, as did his successor, Herodotus, after him.5 \Ve are told, indeed, I Herodot. V. 36. 'Erca'rai‘os 5’ 6 Ao'yo- Kai 71):! Sui/aural aim-oi). 7rozbs . . . 1rpdi'ra ,w‘sv oz’m E'a miAquozl Ba— 2 F1‘. 182, with Klausen’s note; 017%’: 7661/ Hepae'wv dvazpe'eo'fiar, Ka'raAé- 3 F1‘. 170. 4 Fr. 1841: ' If: 'ywu 'roi 're gel/ea wail/Ta 'rciiu iipxe Aapei‘os 5 Herodot. ii. 143. ‘ ‘ * I44 HISTORY or ANCIENT GEOGRAPHY. GHAP. v. that the later writer copied many things from his predecessor literally—among others the history of the phoenix, the descrip- tion of the hippopotamus, and the account of the manner in which the natives caught crocodiles.6 It appears therefore that, in this part of his work at least, Hecataeus was far from confining himself to a dry geographical description of the country, but dilated at considerable length ‘upon its natural curiosities, and the manners and customs of the inhabitants. He, however, certainly added a number of mere geographical details, such as are not found in Herodotus, as we find the names of not less than fifteen cities of Egypt cited from him by Stephanus.7 The remaining part of his work, comprising the coast of Iiilgyacfrom the confines of Egypt to the_-.Straits of Hercules,8 was a mere Periplus, or coast-description, noticing many ports and small islands; while the only tribes of the interior he is known to have mentioned are the Mazyes and Zygantes— obviously the same with the Maxyes and Gyzantes of Hero- dotus, both of which lived within a short distance of the sea, near the Tritonian Lake.9 The most distant places that he enumerates towards the west are Metagonium, a name which . we find in later times applied to a promontory and people not 1 far from the Pillars of Hercules; and Thinga, evidently the same place that was subsequently called Tinga, or Tingis (the- modern Tangier), just without the Straits.2 It is pro- bable also that his river Liza was the same with the Lixus of later geographers, on the Atlantic coast of Mauritania, but this is far from certain. § 9. In regard to the general notions of geography entertained by Hecataeus we are, unfortunately, very much in the dark. It is certainlyqprobahle that-he-was ~one~~of these-writers whom Hverojlb'tiis had in vie_w___When he censuresM“ those who repre- 6 Porphyrius ap. Euseb. Przep. 9 Herodot. iv. 191, 194. Evangek. x. 3, p. 166, cited by Klausen, 1 Fr. 324. The name is again found erylii‘r. 292. in Scylax (§ 110), and in Polybins r. . ;-1_‘Fr,‘269-2ss. 8 Fr. 299-328. (iii. 33). 2 F 326 SECT. 1. HECATZEUS. 14 5 sented the earth as “ exactly round, as if drawn with a pair of compasses, and the Ocean flowing all: around it.”3 This was clearly the popular idea, derived originally from the Homeric poems; and, from all we know of the progress of the Greek mind, there can be no doubt that they would be very slow to emancipate themselves from the influence of an error once established upon such authority. Anaximander of Miletus, the countryman of Hecataeus, as we have seen, was the first that drew up a map of the world, and there can be little doubt that this formed the foundation of that of Hecataeusf’“ But though the latter is said to have introduced material improve- ments on the work of his predecessor, there can be no doubt that both would be still very rude and imperfect attempts, which might well excite the ridicule’ of Herodotus. In this case, as in so many others, it is probable that the scientificr, tendency of the Greek mind came into play, and that theyl assumed the round form of the eafih'ia'nd the circumfluenti. Ocean as first principles, without any actual knowledge of the facts. Weare told also that they placed Greece in centre l Oflheriverld-and Delphi as the‘ central point of Greece.‘ This last notion appears to have obtained a fixed hold on the Greek mind, and is frequently alluded to by the poets of the fifth century B.0., as a received article of popular faith.6 Just in the same manner the geographers of the middle ages assumed that Jerusalem was the centre of the world, and arranged other countries accordingly. §10. It was evidently the same symmetrical turnpof mind that led Hecatzeus to divide the world into two great continents or primary divisions of equal extent, But this question of the division of the continents is not free from difficulty. Hero- dotus, in the passage already referred to, ridicules those who made Asia of equal size with Europe, in terms which seem to 3 Herodot. iv. 36. See Chap. VI.§ 1, 5 Agathemerus, l. c. l. p. 160. “ ZEschyl. Choé'l'rh. v. 1036: Eurip. "1 Agathemer.Geogr.i. c. 1; Strahi. 1011. v. 462; l’ind. l’yth. iv. 74 (183), p. 7. Vi. 3. See also Strabo, 1X. p. 419. VOL. I. l.- .146 HISTORY OF ANCIENT GEOGRAPHY. 01111». v. vkexclude all consideration of a third continent; and Hecatmus, ,by including all Libya, as well as Egypt, under the head of ikAsia] appears to have sanctioned this arrangement. On the other. hand, it is clear that the divisionuinto three continents wasv well established in the time of .HeIodQmfiAfiQ that he himself tells us that he continues- to .usethe divisions and the namesj, because they are sanctioned by custom,” though he thinks them unreasonable, and without good foundation. And in another place he censures “ the Ionians,” who divided the world into the three portions of Europe, Asia, and Africa, but considered the two last as separated by the Nile; thus, as he points out, leaving the Delta unaccounted for.8 It is difficult to suppose that among these “Ionians ” Herodotus did not mean to include Hecataeus—the most recent as well as the most eminent of Ionic writers on geography—or that, if Hecataeus had departed from the generally received doctrine on so important a subject, this would not have been noticed by Herodotus. It seems therefore probable that, although Hecataeus undoubtedly divided his work into only two books or parts, the second ‘of which included the description of Libya as well as that of Asia, he never- theless recognized the established division of the three con- tinents, regarding Asia and Africa together as equal in size to Europe. At the present day we are so accustomed to our modern maps, and to the small size of Europe, as compared to the other two great continents, that we find it diflicult to repre- sent to ourselves the opposite view. But Herodotus undoubt- edly regarded'Europe as greatly exceeding in size both Asia and 7 This appears to be clearly esta- of Egypt. It is probable, therefore, blishcd by the numerous citations of that these were separate subdivisions names of places in Libya from “ He- ~ or portions of the work, included under oataeus in his description of Asia.” ; the main head of Asia. A regular (li- There are, however, many others in vision into books, as usual with later which the “Periegesis of Libya” is writers, is not to be thought of at so cited as if it were a distinct work; but early :1. period. the same thing applies to his description I 2 Herodot. ii. 16. SECT. 1, HECATEUS. 147 Africa together,9 and therefore treats it as a gross blunder on the part of Hecataeus to have considered it as only equal to Asia. One point that doubtless affected the comparison was, that Hecataeus regarded the Cimmerian Bosporus and the Tanai's as the limit between Europe and Asia 1----a view gene- rally adopted in later times—while Herodotus extended the confines of Europe to the river Phasis. Both systems were current in their time, as we learn from the poet ZEschylus, who in one passage adopts one view, in another follows the other.2 § 11. A traveller who had visited Egypt could hardly fail to have formed or adopted some theory concerning the much con- troverted questions respecting the Nile and its annual inunda- tions, a subject which, as we have seen, had already exer- cised the ingenuity of so". cial of the Ionic philosophers. But on this point Hecataeus appears to have acquiesced in the view which, if we may trust to Diodorus, was that of the Egyptian priests:3 that the Nile derived its waters from those of the circumfluent stream of Ocean—a theory which Herodotus justly sets aside as unworthy of refutation.4 A similar want of judgement was displayed by him in accepting, as he appears to have done without scruple, the fabulous tales that were current in his day concerning the Pygmies and the Sciapodes,5 both which nations he placed in Ethiopia, in accordance with the opinion prevalent among the 9 Id. See the next chapter. I Prom. Sol. Fr. 1. Sec § 2, p. 150. 1 This is clearly established with re- 1‘ 3 Diodor. i. 37, § 7. gard to the Cimmerian Bosporus by 4 Herodot. ii. 28. He there includes the fact that Hecataeus placed Phana- i this among the theories set up by goria, which wasjust to the east of the i Greeks “who wished to be thought Strait, in Asia (Fr. 164, 165). The wiser than their neighbours” (ii. 20); Tana'is is not mentioned in the extant and there can be little doubt that he fragments, but, besides that the two had Hecatzeus in view, as he certainly were generally combined by all later had Thales in the first theory which writers who follmved Hecataeus, the . he refutes. See Klauseu on Fr. 278. fact that he placed the Issedones in ‘ But this seems directly contrary to the Asia (Fr. 168) is a strong presumption ‘ statement of Diodorus just cited, no— that he also adopted the Tana'is as the cording to which the same view was l)01111(l=l.1‘_\'. entertained by the Egyptian priests. 2 l/Eschyl. Prom. Vinct. vv. 729-735 ; . 5 Fr. 265, 266. i [.2 148 HISTORY or ANCIENT GEOGRAPHY. can». v. Greeks.‘ It would be interesting to know whether he had really heard these fables in Egypt. On the other hand, the detailed account of the Hyperbo- reans, generally ascribed to Hecatmus of Miletus, undoubtedly belongs to a later writer of the same name, Hecateeus of Abdera, who lived in the time of the first Ptolemy, and com- posed a book expressly on the subject.6 Whether the older writer made any mention of such a people we have no definite information, but the existence of a northern ocean was cer- tainly consistent with his geographical system, as well as with the current belief of his day ;7 and with this belief, as we have seen, the tales of the Rhipaean Mountains and the Hyper- boreans were almost inseparably connected. SECTION 2.-Hecataeus to Herodotus. § 1. Of the progress of geographical knowledge from the time of Hecataeus to that of Herodotus, we have hardly any infor- mation. The work of HELLANIoUs appears to have been purely historical, and we are told that he did not occupy himself much with geographical questions.8 DAMASTES of Sigeum, on the contrary, who was nearly contemporary with Hero- dotus,9 is mentioned as having left a Periplus, and perhaps another work of a specially geographical characterfl Both 6 This work is cited by the Scholiast same time Strabo repeatedly mentions on Apollonius Rhodius (ii. v. 675), as him with contempt as a collector of well as by Elian (Hist. Anim. xi. 1). fables, and unworthy of serious atten- See Wesseling’s note on Diodorus (ii. tion (Strabo,'i. p. 43 ; xi. p. 508). 47), and Klausen on Hecataeus (Fr. 9 Suidas (s. e.) calls him cn'ryxpovos 73), 'l-lpodd'rcp; and both he and Dionysius 7 The statement of Pliny that he of Halicarnassus (Jud. de Thucyd. c. 5) distinctly mentioned the northern place him before the Peloponnesian ocean under the name of the Amalchian War. His date cannot be more accu- Sea Plin. H. N. iv. 13, §94), in all rately determined. See Clinton (F. H. proba ility refers to Heeataens of Ab- vol. ii. p. 371) and C. Miiller (Fragm. dera, though the name has been in- Hist. Graze. vol. ii. p. 64). serted in the maps of the world accord- 1 It is not clear whether the work ing to Hecataeus both by Klausen and referred to as wepl (361161: was distinct Colonel Mure. from the Periplus or not. 8 Agathemer. Geogr. § 1. At the D CCORDIG .__._ . ' H 0. __.__—__.__._.__.___——-—-—-— M —-___—. __________g__ e‘? -=' ‘g— I. ~ London. John Murray. T Sizer. 2. HEGATEUS. 1 49 works are entirely lost; but as we are told that their author copied for the most part from Hecataeus,2 and he is censured by Strabo as a careless and inaccurate writer,3 there is, perhaps, little reason to regret their loss!‘ § 2. But before proceeding to investigate the ample materials furnished by Herodotus for estimating the extent of geogra- phical knowledge to which the Greeks had attained in his time, it will be well to advert briefly to the works of two poets, JZflSQHXLUS and wigs, both of whom may be con- sidered as intermediate in age between Hecateeus and Here- dotus.5 \Ve cannot, indeed, receive their statements as indi- cating in all cases-the same simple belief in the legends which they related, as may fairly be ascribed to the poets of an earlier age; nor are we entitled to assume them asrepre- senting the limits of geographical science in their day. They wrote as poets, not as geographers, and must be criticized as such. But several passages, as well as incidental notices in their works, are interesting as showing the popular notions on the subject, as compared with the 'more definite and elaborate information of writers like Hecataeus and Herodotus. Of all the passages in the extant plays of fEschylus that may be regarded as showing his geographical knowledge, the one that has most exercised the ingenuity of critics and com- mentators is unquestionably that which relates to the wander- ings of Io in the Prometheus Vinctus. Much laborious subtlety has been wasted on the vain attempt to explain these in a manner to accord with some definite geographical system; while the probability is that the poet had no such system present to his mind. He was dealing with a wholly mythical 2 Agathemerus, i. 1. el'ra Aaacta'rns 1 Greek author who mentioned the name 6 Kvr'riei/s ($01‘. Zt'yezez‘zs) 'ra wheio'ra e’lc Of Rome (Dionys. Halicarn. i. 72). 1'00 ‘Exa'rai'ov ,ue'ra'ypoiillas Hepirrkouv I 5 Eschylus obtained his first prize e-ypail/e. I in 13.0. 484, and died in 456. The ex- 3 Strabo, i. p. 47. , tant odes of Pindar belong to a period 4 The most interesting of the few Q extending from no. 502 to ‘152. (See notices that have been preserved to us Clinton, F. H. vol. ii.) from his work, is that he was the first 3' I 50 HISTORY OF ANCIENT GEOGRAPHY. CHAP. V. subject, and by far the greater part of the names that he introduces were of a mythical or fabulous character; hence it probably never occurred to him to consider their geographical position, or arrange them in geographical sequence. Even the few statements that have an apparently definite character will be found as difficult to arrange in any systematic order, as those of a more vague and fabulous ‘description. Thus we are distinctly told that Io, when crossing the Cimmerian Bos- porus would “leave the plains of Europe, and enter on the continent of Asia.” 6 Yet we find Eschylus himself in another passage as distinctly terming the river Phasis the boundary between Europe and Asia.7 Both views, as we have already seen, were currently entertained in his day, and the poet obviously did not trouble himself to reconcile the two. Again, the iron-working Chalybes, who are represented as “ a savage people, not to be approached by strangers,” 8 are placed in the north, adjoining the “ wandering Scythians who dwelt in wattled huts, elevated on wheeled cars ; ” and both these nations are described‘ as situated on the northern side of “ the star-neighbouring summits ” of Caucasus.9 Even the dreaded bay of Salmydessus, the terror of navigators in the Euxine -— “ the step-mother of ships,” as it is termed by the poet 1—is transferred by him from the coast of Thrace to the southern shore of the Euxine, adjoining the river Thermo- don, and the abode of the Amazons. The course of 10’s wanderings after crossing the Cimmerian Bosporus is, if possible, still more extraordinary. The first people to Whom she comes are the Phorcides and Gorgons— mythical beings whom the tradition generally current placed 6 Prom. V. v. 729-735. 7 Prom. Sol. Fr. 1. 8 Prom. V. v. 715. 9 Ibid. vv. 709-722. The river ‘T- Bpm'rv‘ys, which he describes as pouring down its waters from the highest sum- mits of the Caucasus, cannot be iden- tified: the name is not found in any other writer, whether poet or geo- grapher. The name of the Caucasus , here appears for the first time, though it must have been known to the Greeks long before. It is termed by the poet “the most lofty of mountains” (6pr3u ililaa'rov), a statement, as we shall see, confirmed by Herodotus. I 'rpaxe'ia n'évrov Eahpwd-qo'ia yvcieos e’XBpziIEevOQ uaurawt, rmrpvui vedw. Ibid. v. 727. SEcT. 2. HEGATZEUS. I 5 I in Africa; next to these come the Griffins and the Arimas- pians, whom she is especially told to avoid, and after doing so she will come to “a race of black men, who dwelt at the sources of the sun, where is the river Ethiops,” the banks of which she is directed to follow till she arrives at “ the descent where the river Nile pours its grateful waters down from the Bybline Mountains.” 2 It is impossible to believe that in this confused and unintelligible jumble of names and ideas Eschylus had the map of Hecataeus, or any other, present to his mind. He was evidently familiar with certain geogra- phical names, such as the Caucasus and the Cimmerian Bos- porus, and introduced them in the midst of such as were purely mythical and fabulous, without any attempt to com- bine the whole into any intelligible form, or any idea that his hearers would trouble themselves to do so. The place of punishment of Prometheus is itself very obscurely indicated, but it is clear that the poet did not conceive him, according to the tradition current in later times, to have been affixed to a rock on Mount Caucasus, but somewhere on the far borders of Scythia, perhaps to the' Rhipaean Mountains.3 §3. From the fragments of the Prometheus Solutus that have been preserved to us, it is evident that that play contained a prophetic narrative of the journey of Hercules from the Caucasus to the land of the Hesperides, of a somewhat similar character to that of 10 in the sister play. Such an episode could hardly have failed to contain some notices of interest in connexion with those distant regions of the West, which were still so imperfectly known to the Greeks.4 But it was probably of as mythical a character as the corresponding episode in the extant play, and with as little pretension to Kamgmmbv, 5,9,, BUBMVW 5P5”, 5m, casus. (See the passage alrt ady cited.) 2 211m oerr'rhv Neiaos ei'lrrorov beef- 4 Among the few such notices pre- Ib‘d-v' 812' served to us is that of the curious 3 This is evident both from the open- legend invented to account for the ing lines of the play and from the fact stone-covered plain of the Gran near that it was not till after long wan- Marseilles, which is placed hythe poet dcrings that 10 was to come to the Cau— I among the Ligurians (Prom. Sol. Fr. 6). 152 HISTORY OF ANCIENT GEOGRAPHY. CHAP. V. anything like geographical accuracy. Among the few state- ments cited from it is that the Ister took its rise in the land of the Hyperboreans and the Rhipaean Mountains,5 an idea obviously founded upon a mere assumption, like that of the Nile descending from the Bybline Mountains in the south. The notices of geographical names that occur in the Persae ——-a play of a purely historical character—are of a very dif- ferent description, and are interesting as showing that the Greeks, as might have been expected, after the expedition of Xerxes, were familiar not only with the names of the great cities of the Persian Empire—Susa, Ecbatana, and Babylon— but with those of remoter tribes and nations that were subject to the Great King, such as the Parthians, Mardians, and Bactrians. But unfortunately these are nothing but mere names. From the nature of the subject, the play could hardly contain any indications of the real extent or character of the poet’s geographical knowledge. § 4. The same remark applies with still greater force to the poems of PINDAR, the lyric character of which precludes the possibility of anything more than merely incidental allu- sions to geographical names or facts. The only passages that can be referred to as of any interest in this respect are his narrative—if such it can be called—of the voyage of the Argonauts, from the Phasis to the Lake Triton,6 which has been already noticed; and the striking description of the land of the Hyperboreans, “behind the breath of the cold north wind,” from whence Hercules was supposed to have transplanted the olive to Olympia.7 But he himself adds in another passage that no mortal can find the wondrous way to this blessed people either by land or sea.8 It may be added that he elsewhere alludes to the Columns of Hercules,9 to the Phasis and the Nile, as figurative expressions for the extremities of the known world. 5 Schol. ad APOllOIl. RhOd. lV. 284:. \ 8 yavai. 8’ 0576 mega; 111w 5w eiz'pots 6 Find. Pyth. iv. VV, 20-43, e’; 'Ywepfiopc'wv 01751110. Qavpa'rdv 6861/. 7 Zymp. iii. vv. 14—31 ; Pyth. x. vv. Pym X- 30- 31-44. 9 Olymp. iii. 44. News A, B. nncA'rrnus. 153 NOTE A, p. 138. CHARACTER OF EXTANT FRAGMENTS OF HECATZEUS. THE extent and importance of these appears to me to be greatly overrated by Colonel Mure (Hist. of Gr. Lit. vol. iv. p. 151), who says: “The fragments of the Periodus are so numerous as to warrant the belief that they represent a large portion of the substance of the original text, and might perhaps admit of being fashioned into a skeleton of what was, even when entire, a meagre and fleshless body.” But the extant fragments, though numerous, are, as has been already pointed out, with few exceptions, mere names, which, not being arranged in geographical order, afford us no clue to what may have been omitted. There can, indeed, be little doubt that the work was meagre and jejune enough, as is the case with the Periplus of Scylax, to which it has been already compared; but the extracts found in Stephanus of Byzantium—— or rather in the miserable epitome that is now extant of his treatise—certainly cannot be taken as representing the original; and there are not wanting indications that Hecataeus in some parts of his work introduced notices of the character and productions of the lands he described. (See Fragments 58, 72, 172, 173.) NOTE B, p. 139. INTERCOURSE OF THE GREEKS WITH ETBURIA. Colonel Mure, in commenting upon the omission in Hecataeus (to judge from the existing fragments) of all notice of the flourish- ing Pelasgian or Etruscan cities on the west coast of Italy, as well as of the rising State of Rome, adverts to the similar absence of “all special notice of central Italy” in the “copious historical miscellany” of Herodotus, and infers from this coincidence that there was no commercial intercourse between the two nations. The conclusion seems to me quite untenable. It is true that we [54 HISTORY OF ANCIENT GEOGRAPHY. CI-IAP. V. find the Greeks and Tyrrhenians on several occasions on terms of hostility with one another, as in the case of the league formed by the Tyrrhenians and Carthaginians to prevent the Phocaeans from forming a settlement in Corsica (Herodot. i. 166); and again, in B.0. 474, when the same two powers combined to attack the Greek city of Cumm, which was saved by the intervention of Hieron of Syracuse. But such occasional hostilities at long intervals can certainly not be held to imply the absence of peaceful commercial' relations in the intervals. The piratical depredations of the Tyrrhenians-would undoubtedly be liable to give rise to frequent disputes and minor quarrels; nor can it be denied that the jealousy evinced by the commercial states of ancient times fre- quently operated as a check upon their free intercourse. But there must always have been a limit to the restraint thus imposed. And that there existed at an early period—certainly as early as the sixth century B.0.—frequent communication between the two nations, is clearly shown by the exceedingly close resemblance of their works of art—a resemblance which no one can suppose to be fortuitous or occasional. Whatever theory may be adopted to account for the existence in Etruscan sepulchres of countless works of art—especially of painted vases in almost incredible numbers— not only stamped with the unquestionable impress of Greek art, but bearing Greek inscriptions, and even the names of Greek artists, it is impossible to deny that they are a clear evidence of frequent and long continued intercourse between the two countries. Even if it be supposed that these works were actually manufac- tured in Etruria by a colony of Greek artists, this hypothesis would still require that such a colony should have continued to maintain constant intercourse with the mother-country; for not only are the vases found in Etruscan tombs in many cases abso- lutely undistinguishable from those found in Greece proper and the islands of the Egean, as well as in the Greek colonies in Campania and Sicily, but this similarity is found to prevail in works of every different age and style—the most ancient as well as the later and more perfect compositions. A very large propor- tion of those discovered in the Etruscan sepulchres certainly belong to a period of art earlier than the time of Herodotus, or even of Hecataeus, the period during which Colonel Mure would lead us to believe that there was almost no intercourse between Greece and Etruria. NOTE B. HECATZEUS. I 5 5 The Etruscans were celebrated, from a very early period, for their ‘ skill in working in metals, especially bronze, The Tyrrhenian trumpets were celebrated in the days of ZEschylus and Sophocles (Eschyl. Eumen. v. 567; Sophocl. Ajax, v. 17) ; and their cande- labra, lamps, and all kinds of ornamental works in bronze enjoyed a high reputation among the Greeks (Critias ap. Athenaeum, i. p. 28 b). Their embossed gold drinking-cups are also mentioned ‘with especial praise (Ibid.). We learn, moreover, that the opulent and luxurious republic of Sybaris maintained relations of peculiar amity with the Tyrrhe— nians (Athenaeus, xii. p. 519), which must in all probability have been founded—like those with the Milesians—upon commercial relations, and the mutual interchange of works of art and objects of luxury. ( 1'56 ) CHAPTER VI. HERODOTUS. SECTION 1.——General Views. Europe. § 1. THE great work of Herodotus constitutes almost as im- portant an epoch in the progress of geography, as in that of history.1 But in attempting to collect and arrange the vast mass of geographical information which it contains, and to derive from it a correct estimate of the geographical knowledge really possessed by its author, it is necessary to bear in mind the desultory and irregular manner in which this information is communicated to us. ]_E’[_is___h_:i_s_t_()_ry,~aswheh himself remarks, is full of “digressions and episodes ‘incidentally-y ; 2 and iiiit only does this applyjwith to geographical _ portions of the work, but the whole of these portions is intro- Vthiswpanrenthetical manner. There is nothing like an attempt to ‘begin with amgéfiéiéféiitline of a systematic cha- ‘ 1 The commentaries on the geo- graphical information embodied in the history of Herodotus are so numerous that it is unnecessary to attempt to refer to them all. The work of Major Rennell (The Geographical System of Herodotus examined and explained, 1st edit. 4to, Lond. 1800; 2nd edit. 2 vols. 8vo, 1830) is still of the greatest value, notwithstanding the imperfect character of the materials at his com- mand for the actual geography of many of the regions in question. Nie- buhr’s able essay, of which an English translation was published at Oxford in 1830, was the first that pointed out the true character of the map that Hero- dotus had present to his mind, and the necessity of adapting all inquiries to this supposed scheme, rather than to the actual geography, such as we find it in modern maps. Amass of valuable information, in illustration of his author, from modern sources and recent investigations, will be found in the notes and appendices to Mr. Rawlin- son's translation (4 vols. 8V0. Lond. 1858). As I find myself frequently compelled to differ from his conclu- sions, I take this opportunity of ex— pressing the obligations I feel for the industry and diligence with which he has brought together the materials for a fullinvestigation of many disputed points. 2 1rp00'6imas 'yap 51'? ,um 5 Aé'yos ég dpx'iis e’MQn'ro, iv. 30. See also vii. 171. Seer. l. HERODOTUS. I 5 7 racter, and to fill up the different parts from time‘ to time; but some countries which he had himself visited are described with a disproportionate amount of detail, while others are slurred over or neglected; in some instances, no doubt, because he had really no information concerning them, but in others only because no opportunity occurred of introducing them into his history. The influence of both causes may be distinctly traced; and it would be a great mistake to suppose that we are entitled to infer his ignorance of ‘any particular region or country merely from the fact that he gives us no ~ detailed information concerning it. ‘/ A single instance abundantly suffices to prove this. The great republic of Carthage was at this time almost at the height of its power and prosperity, yet it occupies but a very subordinate place in the history of Herodotus. The Cartha- ginians are indeed repeatedly mentioned incidentally, and they are even cited as authorities for specific facts, in connection with other countries ; 3 but there is no attempt at any geogra- phical account of their dominions in Africa, nor any hint of the extent of their colonial empire. Even when the author has occasion to mention the invasion'of Sicily by Hamilcar, at the‘ time of the expedition of Xerxes against Greece, which prevented the Sicilian Greeks from furnishing assistance to their fellow-countrymen, he contents himself with telling us that the Carthaginian general was at the head of an army of 300,000 men, composed not only of Carthaginians and Libyans, but of Iberians, Ligurians, Helisycians, Sardinians and Corsicans.4 This list of names—especially the mention of the Helisycians, an_o_b,scllr.e Ligurian tribe, whose name is cited also from Hecataeus,5 but disappears in later times—T suffices to show that Herodotus must have had accurate infor- " mation on the subject, and could doubtless have told us much 3 Herodot. iv. i3, 195; vii. 167. tempts to identify them with the Vol- 4 vii. 165. scians, appears to me to be utterly un- See Chap. V. p. 139. tenable. The conjecture of Niebuhr, who at- 158 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. more about the invading power, had he considered it as coming within the scope of his work. In like manner there occur none but incidental notices of the Tyrrhenians, though these are sufficient to prove that Herodotus was well acquainted with that people as a formid- able naval power.6 He, moreover, mentions details concerning the Agyllaeans in particular, which certainly imply that they at least maintained habitual intercourse with the Greeks.7 There could therefore have been no difficulty in obtaining further information concerning them, had it suited his pur- pose. N or are we entitled to assume from his silence con- cerning Rome, that the name of that city had never reached his ears. It appears to have been certainly known to his contemporary Damastes,8 as well as to Antiochus of Syracuse, who was not many years his junior. We must therefore be careful how we admit mere negative evidence, as proving the limits of his knowledge, except where the nature of the case is such as to render his silence in itself significant. § 2. There seems no doubt that the history of Herodotus, in the form transmitted to us, was not completed till after his emigration to/Tki (3.9. 443 l, or must at least have received additions and finishing touches subsequently to that period.9 But notwithstanding the increased interest that his removal to that colony would naturally excite in his mind with regard to Italy and the adjoining lands, we find very few traces of this. in his work; and there is nothing like a connected or syste- ‘‘ The contest in which they had en- readers to the Introduction to the trans- gaged, in alliance with the Cartha~ lation of his history by Mr. Rawlinson ginians, against the Greeks of Cumae, . (chap. i.), who has in my opinion suc- supported by Hieron, king of Syracuse, ' cessfully refuted the paradoxes of some was in itself calculated to impress this ' modern writers (especially Dahlmnnn, fact upon the mind of every Greek. in his Life of Herodotus), who would The victory of Hieron, celebrated by bring down the date of its publication Pindar in a well-known passage (Pyth. to so late a period as B.0. 407 , when the i. vv. 136-146), took place in B.0. 474. historian would have been in his 1 Herodot i, 166, 167, seventy-seventh year! M. Stein, in his 8 Dionys. Haliearn. i. 72. edition of Herodotus (8V0, Berlin, 1856), 9 Concerning the life of Herodotus, adopts substantially the same views as and the period of the composition of Mr. Rawlinson (Einlez'tu'ng, pp. 23, his work, I must be content to refer my > 44). SEcT. 1. HERODOTUS. I 59 matic review of the Greek colonies in Magna Greecia, similar to that which Thucydides has given of their settlements in Sicily. Probably Herodotus assumed his readers to be as familiar with the shores of Southern Italy and Sicily, which were in his days studded with Greek colonies, as they were with those of Greece itself and the opposite coasts of the ZEgean. But the omission does not the less serve to show how little he aimed at any systematic geographical description of the countries that were known to him. § 3. It is certain that Herodotus had travelled extensively, and that many of the geographical details with which he has furnished us were the result of personal observation. But it is not easy to determine the extent and limits of his travels, and modern commentators and geographers have differed considerabl in the conclusions they have arrived at on this subject.1 t may, however, be taken as certain that he visited Egypt—where he ascended theNile as far as the first Cataract -—-Babgylon, and the adjoining country, and perhaps Susa. He also\ unquestionably describes ICyrene from personal obser- vation; and the same remark applies to the northern shores of therEuxine as far as Olbia, on the Borysthenes. His observa- tions also on the Colchians and their personal characteristics} certainly seem to imply that he had himself visited their country. Towards the west we have no evidence of his having extended his personal researches beyond the south of Italy,3 where he spent the latter part of his life. But we learn fro incidental notices that he visited the island of Zacynthus and the oracle of Dodona, in Epirus ; in both cases evincing that zealous curiosity and diligent spirit of inquiry that were so characteristic of his mind. 1 See Mr. Rawlinson’s Introduction, 3 As Mr. Rawlinson observes (Hero- already cited, pp. 8—12, and ,_ Stcin’s (lotus, vol. i. p. 12), the only places that Einleit'mg, pp. 13-21, and compare he can be proved to have visited, even them with C01. Mure’s History of Greek in the south of Italy, are Thurii, Cro- Literature (vol. iv. pp. 245-248), who : tona, and Metapontum: and there is has in my opinion greatly exaggerated no evidence of his having ever been the extent of his travels. in Sicily. '-’ Herodot. ii. 104. 160 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. § 4. It is fortunate that Herodotus was led, in one of the many digressions to which we have already adverted, to give a brief general outline of his ideas concerning themagnitude and ' RQSitiOn 0f the (Jhiefoportions.-,of.-the..earthis_surface with which he was acquainted; and though it is extremely difficult to gather from his description a distinct notion of the configura- tion of these countries as he represented them to his own mind, we at least derive from it certain information concerning the extent and limits of his knowledge. After expressing his disbelief in the tales concerning the Hyperboreans, which he justly treats as a Greek fable, and not one of indigenous origin,‘ he proceeds to ridicule the pretensions of those who drew maps of the world, “ without any sense to guide them,” in which they made the whole earth round, as if drawn with a pair of compasses, with the Ocean stream flowing round it, and represented Asia and Europe as of equal size.5 He then adds that he will describe in a few words the size and configuration of the two conti- nents.6 Beginning with Asia, he tells us that there were four nations which occupied the whole breadth of the conti- nent from south to north, extending from one sea to the other. First, the Persians, who extended down to the southern or Erythreean Sea; next beyond them to the north, the Medians; then the Saspirians,7 and beyond them the Colchians, who 4 Herodot. iv. 32. 5 Though it is probable enough, as already suggested, that in these words he refers to Hecataeus, his strictures are certainly not directed only against that author: as he expressly speaks of many such representations—yaw 8% 6péwv 'yiis 1repui50vs 7poiillav'ras 1r 0 A A0 l‘IS ‘if 517, Kai oz’io‘e'va 116011 é'xov'ras e’fi'n'y'no'd- 51.61/01!‘ 04‘ ’Qrceav6v 're pe'ov'ra 7pd¢ovo1 1répz§ 'riw 'yfiz/ eoiio'av KvIcAo'repe'a dis time 'rdpuov, Kai 'riw 14041711 71')‘ Ez’lpib'n'p 'II'OL- ei'wq'wu 10ml (iv. 36). It appears, there- fore, that such maps were not uncom- mon in his time. 6 £11 imi'youn 'yap e’ycb BnAa’m'w ,ué'yafids 're élcoia'rns az’rre'wv Kai 071) 'ri's éa'n e’s 'ypaq‘riyv éxoia'r'n (iv. 36). 7 The Saspirians or Saspeires (Eda- 1reipes)_are also mentioned in the first book (0. 104), as situated between the Colchians and -Medians. They were included, with the Matienians, in the 18th satrapy of the Persian Empire (iii. 94), and‘would seem to have occu- pied the eastern part of Armenia. But the name is not found in any of the later geographers, though cited by Ste- phanus, who writes it Edvreipes : as does also Apollonius Rhodius (ii. vv. 397, 1246), perhaps only for the sake of the metro. This last author places them SECT. 1. HERODOTUS. I61 extend to the northern sea (the Euxine), into which the Phasis pours itself.8 From the broad central tract of Asia thus defined there extended-two projecting tracts,9 ,orjActae, as he terms them, thrown out ~tiiwards the west, and both of them reaching to the sea. The one of these (nearly corresponding with what is now known as Asia Minor) extended from the mouth. of the Phasis_tjoj__th_e_Gulf .of .Myriandrus, and stretched towards the westas far as the,_Hcllespont anduthemTriopiall Promontory; the other included the countries between the Erythraean Sea and the Mediterranean, west of Persia, viz. the Assyrians, Syrians, Phoenicians, and Arabians, and ended at the Arabian Gulf. But he adds that this was no real limit, for that Egypt was“ connected with the tract in question by a narrow neck of land about 1000 stadia (100 G. miles) across. And he pro- ‘ ceeds, still more strangely, to include the whole of Libya“; (meaning the continent of Africa) in this second Acte or projecting tract, although, as he justly remarks, beyond the narrow neck just described Libya spreads out again to a very great breadth.1 The eastern portion of Asia—that is to say, east of the line above drawn from the Persian Gulf to the Euxine—was bounded on the south by the Erythreean Sea, and on the north by the Caspian Sea and the river Araxes‘; which flowed from west to east. It was known and inhabited as far as India, but beyond this the country was uninhabited, and altogfiitllql',Rllknownf The same was the case with regard to the north and east of Europe, concerning which no one was able to say whether they were bounded by the sea or not.3 But even what was known of Europe greatly exceeded in size bqthnAsiaand ‘ Libya, being equal in length to. them -~both \together, and beyond all comparison of greater breadth.4 on the coast of the Euxine, but this is i 9 See Note A, p. 207. clearly opposed to the view of Hero- 1 iv. 38, 39. dotus, and the geographical authority i 2 iv. 40. 3 iii. 115. of Apollonius is Of very little value. 4 fL'FIIKEl: p.611 'yzip ‘trap’ d,u¢o're’pas 1mm’;- 8 iv. 37. E Re: i; Ez’ipdrrrn, ei‘ipeos 5% wept 056% U'Ufl" VOL. I. M I62 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. Libya (Herodotus adds) was known to be surrounded by ‘the sea; for it had been circumnavigated by the Phoenicians in the time of N echo, king of Egypt;5 and the greater part of Asia had been discovered in the time of Darius, by whose orders Scylax of Caryanda had sailed down the river Indus to the sea, and then turning westward, had followed the shores of the Erythreean Sea till he arrived at the head of the Arabian Gulf.6 But the limits of Europe were quite unknown", § 5. The general idea which Herodotus wishes to express in the above description is not difficult to seize, though it is difficult to follow it out in detail, or to represent it on a map. The main ppointof all is the line of separation between Europe and Asia, which he undoubtedly conceivedas'~ ‘running—“not from south to north, as we naturally assume it—but from, ‘west ‘to, , firm—{beginning with the river Phasis, which flowed into the Euxine,-andprolonged from t ence to the omens eagward ‘from that againlbflywthelriverAraxes 8for ‘an inde- fihiteT'distance, ending in the unknown regions to the east- ward. Europe thus extended from the Pillars of Hercules in the west, lying opposite to Africa in the first instance, from which it was separated by the Mediterranean; and subse- quently to Asia, from the Hellespont to the land of the lsse- dones and the Massagetee, who appear to have been the remotest nations known to Herodotus towards the east. This explains his expression (twice repeated) that Europe extends along by the side of the other two continents, so as to be equal to them both in length.9 Bakéew dfiin (pail/emf p.01. ell/cu, iv. 42. Herodot. p. 412). 5 Herodot. iv. 42. The last words are susceptible of two 6 iv. 44. 7 iv. 45. different meanings, but the reasoning of the whole passage leaves no doubt of the sense in which they are used by the author. The ambiguity, how- ever, was the means of misleading Major Rennell, who trusted to Beloe’s translation, and thus led him to entertain a wholly erroneous idea of the views of Herodotus (Geogr. of ' 8 The difficulties connected with the meaning of this name in the Herodotean geography will be considered hereafter. But there can be little doubt that in this passage he means the river Iaxartes, though he erroneously conceived it as flowing from. west to east (iv. 40). 9 iv. 42, 45. SECT. 1. HEBODOTUS. 16 3 On the other hand he greatly underrated the size both of Asia, even as then known to the Greeks, and of Africa? With regard to the latter especially he seems to have been led to a very erroneous estimate of its dimensions by the fact that it had been circumnavigated, though, according‘ to his own account, the voyage had occupied between two and three years.1 In common with most ancient geographers, he sup- posed its greatest length to extend from the Pillars of Hercules to the Isthmus of Suez; and its breadth from the mouth of the Nile to the southern confines of Ethiopia. But it is difficult to understand how, even on this supposition, he could have so greatly underestimated its extension towards the south. In another passage he distinctly speaks of Arabia as the most southern of all inhabited countries} and Ethiopia as the one extending farthest to the west. He therefore appears to have regarded the form of Africa as trending away abruptly towards the west, from near the entrance ‘of the Arabian Gulf; and it is in conformity with this view that he describes the Nile, above the Egyptian fron§i§_,1f,.3t..Ele- flowing from west to east.3 These erroneous notions, on points of fundamental importance, must be care- . fully borne in mind in endeavouring to represent to ourselves the map of the world, as it was conceived by Herodotus. / i It is not clear whether he regarded the African continent as projecting far to the west beyond the Pillars of Hercules. He mentions indeed the promontory of Soloeis, on the western" coast, as the extreme western limit $53165, shifts cites the name as one with which he was familiar ;4 it was doubtless well known to the Carthaginians in his day. But this is far from implying that he was really acquainted with its geogra- phical position.5 ---_____.—______—____._—_1____.__i.. # . ____ to the reality of this circumnavigation 3 ii. 31. See Chapter VIII. will be considered in a future chapter. 4 ii. 34, iv. 43. For our present purpose it is sufficient 5 The Soloeis of Hanno and Scylax that Herodotus undoubtedly believed is certainly Cape Cantin, on the western it, and founded his geographical views . coast, but the expressions of Herodotus 1 The much controverted question as f upon it. 2 Herodot. iii. 107. M2 164 HISTORY or ANCIENT GEOGRAPHY. OHAP. v1. § 6. With the extreme west and north of Europe on the other hand he was, as he himself tells us, altogether unacquainted. He was indeed familiar with the name of Tartessus, as a country of considerable extent in the south of Spain, which as late as the middle of the preceding century had been under the government of a king of its own, who had entered into friendly and commercial relations with the Phocaeans.6 The region thus designated was probably the part of Spain ad- joining the mouth of the Bsetis, or Guadalquivir, though it may also have been applied to the southernmost tract of Spain immediately within the Straits. At all events Hero- dotus was certainly acquainted with the name of Gadeira or Gades, at that time a flourishing Phoenician colony, and which lie-‘distinctly describes as situated “ without the Pillars of H€2Q11138y~11p011 the Ocean.” 7 But this was probably the limit of his knowledge towards the west. How far the ‘Euro- pean continent extended in that direction he had no know- ledge. He had indeed heard of the Cassiterides, “ the islands from “whence tin was brought,” but had no definite informa- "raid concerning them, and‘ ‘appears to have disbelieved in their existence.8 In like manner he rejected the notion (generally received in his time) of ‘a riverjjealled the Eri- if danus, flowing into the northern sea,ifrom which amber was brqught; rand he distinctly adds that, “ though he had taken have been thought to point rather to 8 iii. 115. 0676 w'qo'ovs 075a Kaam're- Cape Spartel, the first promontory after passing through the straits. It seems to me more probable that he had con- founded the two, or rather that in fact he had no distinct idea on the subject at all. He had heard, from Cartha- ginian sources (see iv. 43), of the name of the promontory of Soloeis, as the western extremity of Libya, and knew that it was beyond the Pillars of Her- cules, but that was all. 6 Herodot. i. 163. 7 iv. 8. He here speaks of the fabu- lous Geryones as na'romnuévou 'rip/ "EA- Am/es Aé'yovo'r ’Ep1'16aav Viio'ov 'riyv 1rpbs I‘naei'poro'i v'ro'irn é'gw ‘Hpaxhe'wv o'rnAéwx/ ._ e’1ri 7Q ’.Qrceavq3. pifias e’oi'zo'as, ex 7631/ b nao'o'i'repos iyui‘u (pond. 9 It would seem clear that the iden— tification so strangely made in later times of the Eridanus with the Padus, the‘great river of Northern Italy, had not yet come tobe adopted. Herodotus had only heard of it in connection with the northern sea, and the amber trade, and certainly did not connect it in any way with the Adriatic. The name Eridanus, as he justly‘ observes, was clearly Greek, and he conceives it to have been invented by some poet. It was more probably merely a Greek modification of some barbarian name; and appears to contain the same root as Seer. 1. 165 HERODOTUS. much pains to inquire,” he had been unable to meet with any one who could state from personal observation that there was sea to the north of Europe.1 The passage is a curious one, both as showing the pains that Herodotus took to obtain trustworthy information upon any subject that appeared to him of real interest; and as an instance (not uncommon in the history of science) where the more philosophic mind is practically led into error by an excess of caution, fully justifiable in itself. The popular notion, which had been followed by Hecataeus, assarnecl the existence of an Ocean on all sides of the earth, though the Greeks at this time had certainly no sufficient evidence of the fact.2 Herodotus, on the contrary, required such infor- mation as would satisfy his mind, before he gave credence to this article of general belief; and, failing this, he rejected the fact. \Ve now know that he was wrong, and that the fact is true—though in a very different sense from that supposed by Hecateeus and his contemporaries—but it cannot be denied sophical of the two. \ (that the course taken by Herodotus was the more philo- Towards the north, then, the world as known ‘to our author had no definite vlifmits. Beyond the Scythians, who occupied tmusteppes to the north of the Euxine; dwelt various other barbarian nations whose names: as well as many other particulars concerning them, are recounted to us in detail by Herodotus, from information furnished him by the Greek colonists on the Euxine. But beyond these tribes, who formed a kind of belt around Scythia proper—extending from Rhodanus and Rhenus: though it is impossible to identify it with either of these two rivers. (See Latham’s Ger- mam'a, p. 13, and the article Eridanus in Dr. Smith’s Diet. of Ancient Geo- graphy.) 1 r0570 5% 0066/1/05‘ airrd'zr'rew 'YGVOILéVOU 5ziyaaai axoiiaac 'roii'ro pe/‘te're'wv ihcws doiaamroi e’o'ri 'rd é-n-éirewa 'riis Ebpcéirns, iii. 115. It is certain, however, he adds, that both tin and amber were brought from the farthest parts of the continent. 0:35 érrxa'riis 5’ 3w 5 're Kacrai'repos infill (pom-a“ Kai 'rb 7‘7'Aem'pou. 2 This is again pointed out by Hero- dotus in another passage (iv. 8): row 5% ’.Qiceav5u Mi'yqo ,aézl Aé'youm (oi"E/\M7ves sc.)d1r’ iyAiov dra'rolte'wv dpgoiuex/ox/ 7171/ wepl 1rdo'av (iéew, E'p'ycp 5% of”: d'rro5euc- v50‘. I66 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. _of the central and western portions of Europe. the Carpathians to the Ural Mountains—nothing was known, and it was generally supposed that these regions were unin- hallitable from cold. It is an additional proof of the good sense of Herodotus that he has no mention of the Rhipaean Mountains, which, fictitious as they were, so persistently maintained their place in the geography of the Greeks down to a late period. § 7. It is more surprising to our minds to see how very im- perfect and limited was the knowledge possessed by Herodotus With Thrace indeed he was pretty well acquainted. But he gave to the country designated under that name a much wider extension than later geographers, so as to include the land of the Getae, and the whole region from Mount Haemus (the Balkan) to the Danube. Hence he speaks ‘of the Thracians as, next to the Indians, the most numerous people in the world.3 They were divided into many tribes, among which he particularly specifies the Getae, the Trausi, and the Odrysae. The Getm dwelt near the Danube, extending to the south bank of that river, and hence they were traversed and subdued by Darius on his march to the Danube, when he was on his way to attack the Scythians.4 In the lower part of its course that river formed the boundary between the Thracians and Scythians; but the country due north of the Danube was for the most part\ unknown, and was believed to be uninhabited. The only people on the other side of the rivéifvéfidsfiiame had reached the ears of Herodotus were the Sigynnaeja tribe who had a race of small horses-ponies in fact-with ‘long, shaggy hair, 3 Herodot. v. 3. 4 iv. 93, v. 3. The ethnological re- lations of the Getae have been the sub- ject of so much controversy that it is important to bear in mind that Here- dotus, the earliest author by whom they are mentioned, distinctly regarded them as a Thracian tribe, and calls them the most warlike, as well as the most just, of all the Thracians (oi 6% I‘é'raz . . . aim-flea e’BovAn’rBno‘av,®pnfxwu s’dvp'es dvdpmofi'raroz real Sucaid'ra'ror, iv. 93 . The Odrysse, who, under Sitalees, assumed so prominent a position, and became for a time the dominant power among the Thracians (Thucyd. ii. 96, 97), appear in Herodotus only in a very subordinate character. SE01‘. 1 . HERODOTIIS. 167 which were well adapted for chariots.5 According to our author’s view the country inhabited by them extended to the borders of the Eneti, or Veneti, who dwelt near the head of the Adriatic.6 From this incidental mention of the Veneti, it is clear that not only was their name familiar to Herodotus, but he was acquainted, in a general way, with their true geographical position. He elsewhere terms them an Illyrian tribe.7 § 8. As might have been expected Herodotus was evidently: familiar with the name qfliwqupt Hqaemus (the Balkan), though he gives us no geographical particulars concerning it, and only mentions it incidentally, as the source from whence several considerable tributaries flow into the Danube.8 With that great river itself he was in a certain sense well acquainted, though his knowledge of it will be found on examination to be neither extensive nor accurate. Its mouths indeed were well known to the Greeks, who had founded a colony almost at their entrance,9 and probably traded up the river for a considerable distance. Hence Herodotus had learnt the names of several of the minor streams that flow into it from both. sides, and there is no reason to doubt the accuracy of his information in this respect, though many of these names can- not now be identified. But the navigation up the river was probably cl ked by the rapids at the point now known as the Iron Gates; and the accounts that had reached the Greeks of, the upper part of its course were of the vaguest kind. After describing in detail the streams that descended from the 5 Id. v. 9. No other mention is found of these Sigynnae as an European people in any ancient author, except Apollonius Rhodius, who appears to have regarded them as dwelling on the Euxine (iv. 320). Strabo, on the con- trary, describes a people called Siginni, with their long-haired ponies, in almost the same terms as Herodotus, but places them in the neighbourhood of the Cas— pian and the Caucasus! (xi. 11, § 8, p. 520). 6 Ka'r'hlcew 6% 'rozi'rwv 'roizs ob'povs d'yxoii ’E1/e'ré3v 75511 51! 7Q ’A5pi77 (V. 3). 7 i. 196. 8 iv. 49. 9 The Greek colony of Istrus or Is- tropolis was situated on the Euxine between Tomi (the site of which is clearly fixed at Kustendje) and the mouths of the Danube, but its precise position has not been determined. It appears to have been a place of consi- derable trade, and continued to subsist down to the Roman Empire (Ammian. Marcell. ii. 8, § 43). 168 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. Illyrian Mountains through the plains occupied by the Tri- brafliaiis, Herodotus- adds: “ Two other streams also fall into ’it (the Ister) that have a northerly course, proceeding from 3the country above the Umbrians—the river‘ Carpis and the %i For the Ister flows through the whole length of urope, rising in the land of the lLltae, who, next to the gynetes, dwell the furthest towards thefiwest of allnEuropean nations. And after thus flowing through the whole of Europe, it issues forth upon the flanks of Scythia.” 1 It would be difficult to find a more curious instance of the sort of geographical confusion to which the first hearsay accounts of unexplored regions are liable. The “country above the Umbrians ” must undoubtedly mean the north of Italy, for Herodotus elsewhere2 speaks of the Tyrrhenians (Etruscans) as settling “in the land of the Umbrians;” but so little did he know of these regions that he was evidently unacquainted with the existence of the great mountain chain of the Alps, and erroneously supposed the name to be that of a river. The Cwarpisiin like manner is in all prqbability due to some confused account of the Carpathian Mountains, th$glfii1 this case the direction of the river would also be erroneous. As if to complete the confusion, Herodotus else- ‘where describes the Ister as flowing from‘ the land of the Keltee and the city of Pyr'ene,3 where we have no doubt a mention of the Pyrenees, the third great mountain-chain of central Europe, though applied in an equally distorted manner.‘ § 9. In the passage above quoted 5 Herodotus speaks of the Keltee as the most westerly people of Europe, with the excep- tion of the Kynetes or Kynesians. The same statement is repeated in the somewhat parallel passage,6 where he com- ‘ Herodot. iv. 49. 2 i. 94. I was a mountain, which he placed in 3 ii. 33. the far west of Gaul: and thus de- 4 Even Aristotle supposed the Da- scribes the Ister, in almost the same nube to have its source in “ Pyrene in words as Herodotus, as traversing the the land of the Kelts ” (Meteorol. i. 13, whole of Europe to the Euxine Sea. § 19); but he was aware that Pyrene “ iv. 49. 6 ii. 33. SE01‘. 1. HEBODOTUS. r69 pares the course of the Danube with that of the Nile. But he here adds that the Kelts are situated beyond the Pillars of Hercules, so that he seems to have considered their country as stretching out towards the west, far beyond its real position. Yet in this second passage he again speaks of the Kynesians as dwelling the farthest to the west of all European Hations, and beyond the Kelts. The name is unknown to later geo- graphers,7 but it is cited from Herodorus of Heraclea, a contemporary of Socrates, as that of a region of Spain, adjoin- ing the Ocean.8 What idea Herodotus had formed to himself of their geographical position and that of the Keltae, it is impossible to say; but it is clear that heigadneknowledge whatever of the western, or Atlantic, coasts, either. QLSpainor The name of Iberia is used by him only in reference to the eastern, or Mediterranean side of Spain ;9 and that of the Kelts occurs only in the two passages already referred to. Yet he could hardly have been ignorant of the Kelts who adjoined the sea between the Rhone and the Pyrenees, whose city of Narbo was already mentioned by Hecatzeus as an important emporium of trade.1 § 10. Another passage that affords us important assistance in the attempt to comprehend the general idea that Herodotus had formed to himself of the geography of the earth’s surface, is that in which he institutes a comparison between the two greatest rivers known to him, ufé Nile and the Ister.2 After showing that there was reason to believe that the Nile had its sources far away in the west of Africa, and that it flowed for the greater part of its course from west to east—a subject to 7 It is not found in Strabo, Pliny, or Ptolemy; but reappears in Avienus (cle Ora Maritirna, v. 566), much of whose information is derived from very early authorities. 8 Steph. Byzant. s. v. Kvv'n'rucdv. 9 Thus he speaks of the Phocmans as being the first to make long voyages, and who showed the Greeks the way to Tyrrhenia, Iberia, and Tartessus (i. 163). It was from thence that the Carthaginians drew a portion of their mercenary troops, with which they in- vaded Sicily, under Hamilcar, the son of Hanno (vii. 165). 1 Niebuhr must have overlooked this passage of Hecataeus (Fr. 19) when he says that the Celts, in the time of He- rodotus, had not yet extended to any part of the coast of the Mediterranean (Geogr. of/Herodot. p. 12, Eng. transL). 2 2,33, 34.’ I70 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. which we shall hereafter have to revert—he proceeds to say that he conceives the Nile and the Ister to have somewhat parallel courses, the one dividing Europe through the middle, and the other flowing in like manner through the midst of Africa; and that they both had their origin at about equal distances from their mouths,3 the Ister rising in the far west of Europe, and the Nile (as he conjectured) in the western regions of Libya. He then adds that their mouths were in fact opposite to one another, for that Egypt lay just about opposite to the mountainous. parts of Cilicia, from whence it was a direct journey, in a straight line, of only five days to Sinope, and that city was situated opposite to the mouth of the Ister. He here appears to be, as it were, in a rude manner, drawing a meridian line from the mouth of the Nile to that of the Danube; and infers that, as these were in this sense opposite to one another, and their sources were so likewise (according to his own views concerning the origin of the Nile), the length of both rivers would be about equal. It is scarcely necessary to point out how erroneous was his reason- ing concerning the Nile, the upper course and sources of which were, as he himself tells us, utterly unknown; and we have already seen how extremely imperfect was his informa- tion concerning the upper course of the Danube, though this he supposed to be well ascertained, for (he tells us) “the Ister flows through an inhabited region, and is therefore known to many.” 4 It is more curious that, while his infe- rence concerning the relative position of the mouths of the two rivers is approximately correct—the Canopic, or western should express in modern geographical the sense of the somewhat obscure phrase by saying that they flowed expression that the Nile n5 "Imrpqo e’x through the same number of degrees of ‘rap Tawv aé'rpwv lip/lawn. The two ; longitude: though of course no such rivers had a general analogy in all re- , idea, in a distinct form, was present to spects; they both flowed from west to ‘ the mind of Herodotus. east, they had their mouths opposite to ‘"6 néu dif'lo'q'pos pe’et ,uév yap 84’ 01’- one another, and so were their sources. mevnévns, 1rpbs mam'bv 'ywa’mce'rai, 34. The sense is very much the same as we , 3 This appears to me to be certainly SECT. 1. HERODOTUS. 171 mouth of the Nile, difi‘ering in fact only a few miles in longi- tude from those of the Danube—the data upon which it is founded are glaringly erroneous. Cilicia Trachea, as the mountainous part of Cilicia was called, lies considerably to the east of the whole Delta of Egypt, and Sinope still more so; so that that city is in fact more than five degrees of longitude to the eastward of the Canopic mouth of the Nile. But on the other hand, it is in fact by about as much further east than the mouths of the Danube, to which Herodotus considered it as directly opposite. Thus, in this case, it accidentally happens that the two errors committed by him just about neutralize one another. Had he not mentioned the steps of the process, we might have wondered at the accu- racy of the result, and given him credit for an amount of geographical knowledge which he certainly did not possess. There can be little doubt that Herodotus was led to insti- tute this parallel between the two rivers by something of that feeling of symmetry so congenial to the Greek mind. But we must not press the comparison too far. It is hardly to be inferred from the passage in question that the historian con- sidered the Ister as making a great bend analogous to that of the Nile where it entered Egypt. Yet it appears certain from other passages that he did consider it as turning south- wards in order to reach the sea, and he distinctly tells us that where it entered the Euxine its mouth was turned towards the south-east.5 This would in fact be the direction of one of the mouths that formed the delta of the river, and 5 5’Irrrpos M54502‘ e’s ab'rhv (Thu 60i- Aao'aow Sc.) 'n'pbs edpov duel/.011 'rb 0"r6,ua 're'rpapuél/os, iV. 99. This passage is, in my opinion, mistranslated by Mr. Rawlinson, who renders it “ the Ister falling into the sea at this point, with its mouth facing the east.” But the Efipos live/1.03 of Herodotus is certainly meant to designate the south-east, in accordance with the usage which we find established in the time of Ari stotlc. Only a few lines further on, he dis- tinctly uses the expression 1rpbs ampu- a’rrm/ due/tow in the sense of “towards the east ;” and he elsewhere repeatedly uses the same term for .the east wind (iv. 152, vii. 188). M. Kiepert, in his map annexed to Stein’s edition of Herodotus, has correctly represented the Danube, in accordance with the views of the author; as has also Dr. G. Miiller in Smith's Historical Atlas of Ancient Geography. 172 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. this perhaps gave rise to the misconception that it was that of the main stream. But it will be found impossible, as we shall hereafter see, to understand the view taken by Hero- dotus of the geography of Scythia, without bearing in mind this fundamental error concerning the course of the Danube. Snc'rron 2.—Seythia. § 1. The very scanty and imperfect knowledge which Hero- dotus displays of so large a portion of Europe contrasts strangely, to the modern reader, with the large amount of information which he has furnished us in regard to the countries north of the Euxine, and the vast tract of country that was commonly included by the Greeks under the vague and general appella- tion of Scythia. Several circumstances had indeed combined from a very early period to direct the especial attention of the Greeks to the countries in question. The first of these was the great migratory movement of the people called by the Greeks-Cim'l merians, who, according to the tradition universally received, ‘and adopted by Herodotus, had originally occupied the tract of country north of the Euxine, between the Ister and the Tanais ;- but having been expelled from thence by the Scythians, had poured themselves in a vast horde down upon Asia Minor, where they had carried devastation and terror to the gates of the Greek cities of Ioniah/This invasion is un- doubtedly entitled to be received as an historical fact ; 6 nor is there any reason to doubt the traditionary belief that the people in question came from the regions north of the Euxine, in which the Scythians were found in the time of Herodotus. Such a displacement of one of these nomad nations by another, 6 The invasion of Asia by the Cim- xiii. p. 627). It was doubtless from merians is alluded to by the earliest these early authorities that Callimachus Greek elegiac poets, Archilochus and derived the picture he drew of them as Callinus, who were contemporary with “ dwelling in waggons ” and “ milkers it (Callinus, Fr. 2, 3, ed. Bergk; Strabo, . of mares ” (H. ad Dian. v. 252—260). “A . U Ger us ‘8 8 1‘ - 9 'w 6 ‘8 Q _ \ a 5 'o ?; 05¢ fl v ~>~ _. t x P’ ) ~ -, ‘ q a o e a? x \ ‘_ . 'I/ZflnW/V m; *,=.‘-.Y1;I}?,x\.. a b L. ( ~ .__ - ~ ~__ ' . _ "/fiimMtnsk/Q Q"“"""'e§i¢> A v _——:_—:."__:_:_:1'____ ___._4‘_______ "'Lil‘yli’i'mz-{glg "'Yf'lvmniyl; ‘ ~:—_—_—~_'—::———-——*——— _ “-,'.'.' /, ( will‘, ~, _ . ,, —————'-. ~mtT—mw ' QJ’MQL: us”: —————~—— _—'__"i ~_:—'-'; m\.k\\.\“\\“\\_\_ \\\'\\\§"".1'| '\ ‘~17 523-1- 5 -' ' '_"_“ ‘ZékmY‘N‘ . wit-M»- \ ‘-=\\\\‘¥\\\‘ ...__.._-- =5: : —~~——'_—> "P" _‘ \ ‘é ‘ = l ‘ 0 60 A 1 a} I]. s —- ' ______*-— Phasis R‘ xYians Cas pagy rus O LIB‘YA THERIIODES (Region, of” Wild Beast“ /) “mm” THE WORLD according to AE THIOPIAN HER-ODOTUS ____l:‘_ _~A~ -,_ _ I I Stam‘arfzils‘ 5pc LondonrJohnMurray ' ~ - SECT. 2. HERODOTUS Z SOYTHIA. I is entirely in accordance with what we know to have taken place at subsequent periods; and is of great interest as the first authentic record of those great movements of nations that have taken place in all ages on the frontiers of Europe and Asia.7 /' The gradual extension of the Greek colonies along the northern shores of the Euxine,“andwthemsettlement of such powerful cities and commercial centres as Olbiai and Panti- capacum, would naturally add to the general interest felt by the Greeks in these countries; and the expedition of Darius into Scythia, before the close of the sixth century B.0., must i have given a fresh stimulus to their curiosity as ‘to the vast regions that lay beyond the Danube? The grossly exaggerated rumours that reached their ears with regard to the extent and circumstances of that expedition would only tend to increase this feeling. It was doubtless the same cause that led Hero- dotus himself to visit the remote Greek settlements in this quarter, and to collect there by personal inquiry and observa-_? tion the valuable mass of information which he has embodied} SLin the fourth book of his history.8‘/ § 2. There are indeed few parts of that great work that display in a more striking manner at once the merits and defects of the writer. It is difficult to determine the extent of his personal knowledge of the regions that lay to the north of the Euxine, but it is certain that he spent some time at Olbia on the Borysthenes,~where he evidently occupied himsel'f'dil‘i- . gaitlgf'i‘fi'emé'cfing information from the traders and, others that were accustomed to penetrate into the interior. “He men- tions having himself seen a remarkable monument at a place called Exampzeus, or “the Sacred Ways,” four days’ voyage up the river Hypanis,9 but we have no proof that his personal 7 See Note B, p. 208. i 1855). 8 In regard to this portion of the l 9 Herodot. iv. 81. He appears also to geography of Herodotus, every student ! have been on shore at the mouth of the of that author must consult the valu- Tyras, where there was a Greek colony able work of M. Neumann (Die Hel- of the name, as he mentions two things lenen 't'm Skythenla-nde, 8V0, Berlin, to be seen there in a manner that cor- I74 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. observation extended farther inland. The lively picture that he has drawn of the mode of life and customs of the ngmadic Scythian tribes wo.uld-readily_,_l_eadx us to suppose that he had himself seen them in their native wilds; but it is difiicult to reconcile this with the geographical inaccuracies which we shall hereafter have to point out; and it seems more probable that his personal acquaintance with them was limited to those he might have seen at the fairs and markets of Olbia or other Greek towns, while he had gathered fuller particulars from careful inquiry. In proportion indeed as he recedes from the shores of the Euxine, his information becomes more vague and untrustworthy; for his geographical data had to be derived from the statements of traders who had travelled only with a view to their own commercial objects, and who had doubtless never seen a map, or attempted to form in their own minds any definite geographical idea of the countries they had visited. Still, the amount and extent of his information concerning the various tribes and nations of the interior is in every way remarkable; and leads us to the unquestionable inference that the Greeks of his day had carried their commercial relations, either by land or water (probably the former), to an extent that remained unsurpassed until a much later period. Even in the days of Augustus, Strabo was able to add little or nothing to the extent of our knowledge of the countries in question, while Pliny and Mela mix up the statements of Herodotus with those of later writers, referring to a wholly altered state of things, in a manner which throws the whole subject into inextricable confusion. - In another respect Herodotus stands conspicuously dis- tinguished from his predecessors. All previous writers—so tainly seems to point to personal ob- Thracian Bosphorus, where he refers servation (iv. 11, 82). But there is no to the brazen bowl set up by Pausanias other point on the coast of the Euxine in a manner that clearly shows that he where he can be proved to have touched had himself seen it (iv. 81). on his way to Olbia, after leaving the SECT. 2. HERODOTUS: SCYTHIA. I75 far as we can judge from the fragments that remain to us-- had concurred in making the Scythian deserts the abode of mythical races, and the scene of poetical fables. Herodotus, on the contrary, is careful to distinguish what he received on credible testimony from all such fabulous accounts ; and though he does not refuse to mention the Arimaspians, he expresses his doubts of their existence, and distinctly tells us that the fact rested only on the authority of the Issedones, from whom the Scythians had learnt the tale, which had passed through them to the Greeks.1 ()f the Hyperboreans on the contrary, as we have already seen, he altogether disbelieved the existence.2 Nor is any mention found in his work of the Rhipzean Mountains, and his silence may be considered in this instance conclusive, as he could not have failed to notice them in connection with the great rivers of Scythia, had he believed in their existence. § 3. Before attempting to examine the account given by Hero- dotus of the Scythians and the adjoining tribes, it is necessary to advert briefly to his ideas on the subject of the Euxine itself, and its subordinate or tributary sea, the Palus Maeotis. His misconceptions concerning the extent and configuration of these two great basins of water will be found to have an important bearing on the geographical arrangement of the numerous tribes that adjoined their shores; and it is impossible to represent to ourselves the idea that he had formed'of the Scythian territories, and the succession of the different tribes,“ without clearly comprehending the wide divergence of his notions on this subject from the reality. Herodotus had himself navigated the Euxine, andmitsmyvaters were in his days frequently traversed by Greek traders in all 1 iii. 116, iv. 27. He speaks with equal caution of the strange races re- ported by the Argippeans to dwell be- (ad Dionys. Perieg. v. 31) cites Hero- dotus as affirming the very fact that he combats: viz. the existence of the yond them to the north (iv. 25). 2 See above, p. 160. It is an important lesson to us, of the caution necessary in making use of the statements quoted by late grammarians from earlier writers, that Eustathius Hyperboreans, beyond the Arimaspians and the griflins, extending to the sea : the account quoted indeed by Herodo- tus from Aristeas (iv. 13), but of which he repeatedly expresses his disbelief (Ibid. 16, 32). I 76 HISTORY or ANCIENT GEOGRAPHY. can). v1. directions. But we must not hastily conclude that he had therefore correct ideas as to its form and magnitude, or that he was acquainted with its true geography, in the modern sense of the term. He had indeed arrived at definite con- clusions on the subject, which prove the pains he had taken to inform himself; while their erroneous character shows the difficulty that existed in his time in arriving at a correct result. He tells us that the Euxine, which he terms “the most wonderful of all seas” is eleven thousand one hundred stadia in length, and its breadth, at the widest part, is three thousand three hundred stadia. The greatest length he con- siders to be from the mouth of the Pontus (i. e. the Thracian Bosphorus) to the Phasis; and its greatest breadth to be between Themiscyra at the mouth of the river Thermodon, and Sindica (the peninsula of Taman) at the entrance of the Cimmerian Bosporus. The last supposition involves a con- siderable error. The broadest part of the Black Sea is in reality between Heraclea on the coast of Bithynia, and the mouth of the estuary of the Borysthenes, where its breadth is not less than 325 geographical miles, or 3250 stadia—very nearly agreeing with the estimate of Herodotus—while the actual distance between the points selected by him does not exceed 235 G. miles or 2350 stadia. The length on the other ‘hand is enormously exaggerated; the distance from the mouth of the Thracian Bosphorus to that of the Phasis being in a direct course not more than about 5500 stades, or just about half that stated by Herodotus. ' § 4. The source of his error is not far to seek; for he has himself told us in this instance the process of computation by which he arrived at his result. “ In a long day (he tells us) a ship usually accomplishes about 70,000 fathoms, and about 60,000 . fathoms in the night.”3 N ow from the Phasis to the mouth 3 The mention of fathoms in this cal-‘ l converts these fathoms into stadia—the culation seems to show that this was common measure of land distances. the customary nautical mode of mea- It would be very curious to know surement. It is Herodotus himself who , what was the mode employed by the SECT. 2. . HERODOTUS: SCYTHIA. ' I 77 of the Bosphorus was a voyage of nine days and eight nights; and from Sindica to Themiscyra was a voyage of three days and two nights. Hence he computed the distances by a simple process of multiplication; reckoning, as the Greeks always did, 100 fathoms, or 600 feet, to the stade.4 Here it is evident that the error arose simply from an over-estimate of the average rate of sailing. It was no doubt possible that a Greek merchant vessel should, under favourable. circumstances, accomplish as much as 700 stadia (70 G. miles) in a day, and 600 in the night; but it was a great mistake to take this as the distance performed on the average. The ordinary estimate of later geographers is 500 stadia a day, and the same for a night.5 But even this would make the estimate of Herodotus greatly exceed the truth, and it is evident that we must allow for considerable loss of time, in consequence of the windings of the coast, bafiiing winds, caused by projecting headlands, and other causes of delay, of which Herodotus took no account. On the other hand his estimate of the width is comparatively so little in excess, that it shows clearly that the Greek sailors of those days were in the habit of taking advantage of the prevalence of north winds, and stretching directly across from the mouth of the Cimmerian Bosporus to the Greek settlements on the southern shores of the Euxine. § 5. But if his ideas of the dimensions and form of the Euxine, or Pontus, as he generally calls it, were thus disfigured by grave errors, his conception of the PalnsMaeotis—“ the mother of the Euxine,” as it was commonly termed by the Greeks—- was still more erroneous. “ The Pontus (he tells us, after de- scribing its dimensions in the above passage) has also a lake belonging to it, not very much inferior to itself ‘in size ; the waters of which flow into it. It is called Meeotis, and the mother of the Pontus.” This is indeed a startling mis- statement. At the present day the Sea of Azovcovers an shipmasters of the Euxine in order to no clue. ‘ See Note C, p. 209. calculate the distance run in a day, 5 Scylax, Pe'rz'plus, § 69. but to this we have unfortunately VOL. I. N I78 HISTORY or ANCIENT GEOGRAPHY. crap. v1. area of about 14,000 square miles, which is little more thap one-twelfth of that of the Black Sea. Hence many geographers and commentators on Herodotus (including even his most recent translator) have thought it necessary to suppose that some great physical changes have taken place since his time, and that the Palus Meeotis was really much more extensive in ancient times than at the present day.6 It is certain indeed that the Sea of Azov, which is throughout its extent very shallow, is gradually filling up, and becoming perceptibly shallower; while the continual deposit of silt and mud by the river Don is necessarily pushing forward its delta into the sea, and a similar deposit is taking place at the mouth of the Kuban also, on the southern shore of the sea.7 It is true also that there are geological reasons which have led Pallas and other writers to speculate on the probability that the whole tract between the Caspian and the Sea of Azov was submerged, and the two seas consequently united, at no very remote period, geologically speaking. But it would be very hasty to assume from this that any material change has taken place within the historical period, or since ‘the time of Herodotus, and it is to be remarked that Scylax (or the author of the Periplus that bears his name), writing little more than a century after Herodotus, estimates the Maeotis at only half the size of the Euxine.8 This is, of course, still a great exaggera- tion ; but the idea of its great size seems to have continued fixed in the minds of the Greeks for ages afterwards, and even Ptolemy still represented it as enormously larger than it really is.9 6 See Mr. Rawlinson’s note to his translation of Herodotus, vol. iii. p. 7 All these causes were in full ope~ ration in the time of Polybius, who enters into some interesting specu- lations on the subject: anticipating the time when not only the Palus Maeotis, but even the Euxine itself, might be filled up by these ‘deposits (iv. 40). These views are perfectly sound, as a ‘ matter of geological theory; but the rate of progress has been very much slower than he supposed, and the amount of actual change that has taken place within 2000 years is compara- tively trifling. (See Chapter XVII.) 8 Scylax, Peripl. § 69. _ 9 Ptolemy supposed the Palus Meeotis to extend through nearly six degrees of latitude from S. to N., with a breadth more than equal, if we include the SECT. 2. HERODOTUS: SOYTHIA. I 79 , The truth appears to be, that the Palus Mzeotis was generally regarded by the Greeks as a lake, not as an inland sea; con- sidered as such, it was ‘enormously larger than any other lake with which they were acquainted; and the idea thus formed of its vast size led readily to an exaggerated estimate of its real dimensions. It seems much more natural to believe that Herodotus was misled by erroneous information than to have recourse to the supposition of physical changes having taken _ place with unexampled rapidity. § 6. N o allusion is found in Herodotus to the fancied resem- blance of the general form of the Euxine to a Scythian bow, so familiar to the later Greek geographers. This comparison indeed implies a much more accurate conception of its general configuration than was possessed by the historian, and especially of the true size and position of the Tauric peninsula, concern- ing which he entertained strangely erroneous notions, for one who had himself navigated the Euxine, of which it forms so prominent a: feature. He indeed correctly describes the Tauric ~ territory as a mountainous district projecting into the Pontus, _ and forming the advanced point of Scythia between the seas that bounded it on the south and east ; but in order to convey to his readers an idea of its form, he compares it successively with the projecting portion of Attica, and with that of Iapygia in Italy, beyond the line from Brundusium to Tarentum.1 Both comparisons show clearly that he conceived it only as an Act‘e, or projecting tract of land, and that he was unaware of its being a peninsula in the strictest sense of the term, joined to the mainland only by a narrow neck or isthmus : otherwise the Peloponnese must have offered itself as a much apter com- parison, and one familiar to all Greeks. The two peninsulas are indeed very nearly about the same size, and the Isthmus of Perekop, which unites the Crimea to the continent, is little broader than that of Corinth.2 inlet, which he calls Byce Leone, and rently‘ a very imperfect notion. (See which must represent the Putrid Sea Chapter XXVII.) ‘ Herodot. iv. 99. of Strabo, of which Ptolemy had appa- 2 The resemblance between the two N2 1840 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. Here again it has been suggested that some physical change may have taken place, and that the shallow inlet now called the Putrid Sea, which alone gives this strictly peninsular form to the Crimea, may not have existed in the time of Herodotus. It must be admitted that a very slight alteration of levels would effect this, and would transform what is now a shallow inland sea into a mere salt-marsh, or even into dry land. But even then the description of Herodotus would be far from characteristic, and it seems much safer in this instance also to acquiesce in the conclusion that he was simply misled by imperfect information.3 ' § 7. But however imperfect were the geographical notions of the Greek settlers as to the extent and configuration of the Euxine, and its tributary the Maeotis, they had long been familiar with the great rivers which flow into those seas, and constitute the leading characteristic of Eastern Europe. The Ister or Danube, the Tyras (now the Dniester), the Hypanis (Bug), the Borysthenes, or Dnieper, and the Tanai's or Don, are not only all mentioned by Herodotus, and enumerated in their correct order, but they are described more or less fully. Of the Ister (as we have already seen) he speaks in considerable detail,4 and seems to have had pretty accurate information concerning the lower part of its course, where it formed, according to his view, the frontier between Scythia and Thrace.5 But as he makes no mention of the‘ cataracts or rapids at the Iron Gates, which have in all ages opposed so serious an- obstacle to the navigation of the river, it is clear is well pointed out by Strabo (vii. 4, § 5) i) 5% 'pe'yoikn xeppdvmros 'rfi IleAo- wowficcp 1rpoo'e’ouce Kai. 'rb aXii/m Ital 'rb né'yefios. 3 The improbability of any such change having occurred within the his- torical period is greatly augmented by the circumstance that Strabo describes the Tauric peninsula, and the Putrid Sea in particular, with great accuracy, and precisely in accordance with their present condition (Strabo, vii. 4). But if any great physical change had taken place in the interval between Hero- dotus and Strabo, in the immediate neighbourhood of the flourishing Greek settlements on’ the Bosporus, which were during this period at the very height of their prosperity, some tra- dition of it would surely have been pre- served, and have become known to the later geographer. 4 Herodot. iv. 48-50. 5 iv. 99. SE01‘. 2. HERODOTUS: SCYTHIA. 181 that he had no full or complete knowledge of its course, as high as that point (about 450 miles from its mouth), and we have already seen how extremely vague and imperfect were his notions concerning the upper part of its course, and the affluents it there received. Even in the lower portion, though he enumerates the tributaries that fall into it on the one side or the other, as if he had accurate knowledge on the subject, it is impossible to identify with any certainty the streams really meant, with the exception of the Pprata, or Pruth, which, as it joins the ‘Danube but a short distance above its delta, and has a course nearly parallel with that of the Tyras, would naturally be well known to the Greek colonists of the Euxine.6 Farther inland indeed he mentions the Maris, as flowing from the land of the Agathyrsi, which may in all probability be identified with the Marosch, the principal river of Tran- sylvania:7 but the other names enumerated by him, are not mentioned by any later writer, and can for the most part only be identified by arbitrary selection among the numerous streams that pour their waters into the Danube on the one bank or the other. § 8. Herodotus expressly calls the Ister the greatest of all known rivers,8 though he elsewhere remarks that it owes its “ ' predominance to its numerous tributaries: for that, taking the main streams separately, the Nile was far superior to it in the volume of its waters.9 Besides its magnitude, another circum- stance that particularly attracted his attention in this mighty stream was its equable flow, which presented no difference in 6 Herodotus himself tells us (iv. 48) that it was called Pora-ta by the Scy- thians, but Puretos by the Greeks, a statement that clearly shows the name to have been one‘ familiar to Greek ears. 1 It is true that the Marosch does not fall directly into the Danube, but into the Theiss, which is itself a tribu- tary of the Danube. But this is a point of little importance. Even at a much later period geographers do not seem to have appreciated the superior importance of the Theiss. Strabo speaks of the Marisns as flowing into the Danube, and serving as the chan- nel by which the Romans sent up their supplies for the Dacian war (vii'. 3, § 13) ; while the name of the Tibiscus, or Theiss, is first found in Ptolemy. 8 "Icr'rp0$ ,uéu e’cbz/ aé'yw'ros 1r0'raac'6v ' mix/Tau 'réiv 1314.625 i’fiaeu (iv. 4:8). 9 Ibid. 50. 182 HISTORY or ANCIENT GEOGRAPHY. crap. VI. winter or summer: affording in this respect a marked‘ con- trast, not only to the Nile, but to all those rivers with which Herodotus was likely to be most familiar, either in European Greece or Asia Minor. This peculiarity he justly attributes / I to the melting of the accumulated snows by the summer heat compensating for the increased evaporation at that season.l But he was not aware that this great accumulation of snow was owing to the chains of high mountains that supply the main tributaries of the Danube, and he ascribed it to the ordinary severity of the winter, in the lands from which they flowed ;— the climate of which he supposed, naturally enough, to re- semble that of Scythia. § 9. As has been already pointed out, he describes the Ister as flowing into the Euxine with its mouth facing the south- east: the next of the great rivers, the T raslwflowed from north to south, having its sources in a large lake on the outer confines of Scythia, where it adjoined the land of the N euri.2 The Hypanis, which came next, also rose in a large lake, which was thence called “ the mother of the Hypanisi‘f" its waters were at first quite fresh, but in the lower part of its‘ course they were rendered bitter or salt by the admixture of those of a source of extremely salt water, at the distance“ of four days’ voyage from the sea.3 It is singular that Herodotus describes the Hypanis, though he calls it a river with which few could compare in size,‘ as if its whole course was only nine days’ voyage from its sources to the sea, and though these are obviously intended to be reckoned descending the stream, the statement is never- theless difiieult to understand.5 § 10. The Borysthenes, which he justly accounted the largest ' Ibid. 2 iv. 51. 3 Ibid. 52. miles in length, and not less than 300 ‘‘ 'rbvu'f'rravw édv'ra 1rora,u.bv,é1l imi'yozo'r miles in a direct liné from its source to pé'yav (iv. 52). It is strange that Mr. its mouth. It is probable that there is Rawlinson should translate these words some confusion in our existing text; as “ a large stream among those of the and that Herodotus did not mean the second order.” They are correctly five days’ voyage mentioned in the first rendered-by Valckenaer (ad loc.) “ fiu- , instance to comprise the whole distance men in paucis magnum." from the lake to the salt fountain. ~" The Bug is in reality about 480 i - SECT. 2. HERODOTUS: SCYTHIA. 18 3 of the Scythian rivers, after the Danube, was, he adds, the most productive of all rivers, not only in Scythia, but in the world, with the single exception of the Nile. The vast quantity and variety of fish that it produced, some of them of the largest size; the richness and extent of the pastures on its banks ; the fertility of the soil for cultivation; and the sweet- ness of its clear waters,—at the same time that salt was pro- duced in abundance at its mouth—gave it the palm over all its competitors.6 Nor are these praises exaggerated. The Greek colony founded near its mouth, and from thence gene- rally known among the Greeks as Borysthenes, but called by its inhabitants Olbia or Olbiopolis (“the prosperous city”) owed its wealth and prosperity mainly to its position at the entrance of this great stream, which opened out the access to the richest and most fertile provinces of Southern Russia. Yet even of the Borysthenes itself his geographical know- ledge was very imperfect. He himself tells us that no one was acquainted with its source, but that it was known for a distance of forty days’ voyage, as far as a place called Gerrhus, and that its course was from north to south.7 He therefore evidently considered it as flowing parallel with/the Hypanis, and had no conception of the vast bend by which the Dnieper sweeps round from below Kiev by Ekaterinoslav to Kherson. It is more remarkable that he seems to be un- acquainted with the cataracts or rapids, which interrupt the navigation of the river during this part of its course, for a distance of more than forty miles, and must in all ages have ' opposed a barrier to communication with the regions beyond. But he himself tells us that it flowed through the land of the / Georgi, or agricultural Scythians, for the lowest ten days’ \voyage, and above that its course lay through uninhabited 6 Herodot. iv. 53. This character of ' 05709 5% mil/ml’ év'fi xpflwsévmros- _ I p.e'yzi1\a. Kai. n'oAMi Kill. Kap‘rroi); gbe'pwv the Bolysthenf's seems to have become 70:); ¢vopévovs uojuuig~ 're 10?}; Bocmfinaaw. traditional. Scymnus Chius, who pro- vv. 813-815, ed. Muller. bably QoRled_ du'ectly from Ephorus’ The huge fish are doubtless sturgeon, says of It m 111% meme“ . which still abound in the Dnieper. 7 See Note D, p. 211. 184 CHAP. VI. HISTORY OF ANCIENT GEOGRAPHY. regions ;8 so that little would be known concerning it. His erroneous view of its direction would however tend materially to vitiate all his ideas of the geography of this part of Scythia, and must be carefully borne in mind in attempting to arrange and explain his other statements on the subject. § 11. The last of the great rivers of Scythia which he mentions is the Tingis, with which he was well acquainted, by name at least: this\he describes as flowing in the first place (like the Hypanis and Tyras) from a great lake, and discharging itself into a still greater, the Maeotis, which formed the boundary between the Scythians and Sarmatians. He adds also that it received a tributary river called the Hyrgis.9 In point of fact the Don does rise in a lake, while the Bug and the Dniester do not, but it is one of such very small dimensions, as not even to figure on any ordinary map of Russia,1 and it is wholly inconceivable that the informants of the historian had sufficiently accurate knowledge of these remote regions of the interior to be aware of this minute fact, while their general notions were so vague and incoherent. It is much more pro- bable that in all three cases the lake was invented, or assumed by the persons from whom Herodotus derived his information, as the readiest means of accounting for the origin of a great river. At the same time the extensive marshes in the interior of Russia would favour the idea, and may perhaps have really contained accumulations of stagnant water of greater extent than they do now. In one respect indeed the knowledge obtained by Herodotus was superior to that possessed by many later writers. Both Hippocrates and Aristotle believed in the existence of a great chain of mountains “in the extreme north, beyond the limits of Scythia,” in which all the greatest of these rivers took their rise, and to which they gave the name of the Rhipaean Mountains.2 Herodotus on the contrary 8 Herodot. iv. 53. 9 The Hyrgis is generally identified with the Donetz, but merely on the ground that that river is the most con- siderable amuent of the Don. I 1 It is called Lake Ivan Ozero, and is situated in about 34° N. latitude, but is described as a mere pool. . 2 Aristot. Meteorol. i. 13. Hippocrates even accepts the popular notion that it SECT. 2. HERODOTUS: SCYTHIA. ' 185 (as has been already mentioned), not only makes mention of the existence ogfmagywspphmrange, but his account of the great rivers just described, altogether excludes the supposition that they,dprivedwtheirlwaters from such a-source.. § 12. On the whole, the knowledge possessed by Herodotus of the five principal rivers of Scythia is very much what might have been expected. Their mouths and the lower portion of their courses would be well known to the Greek settlers on the northern shores of the Euxine, while vague and often erroneous rumours would be all that reached them concerning the sources of the same rivers, or the part of their course during which they flowed through wild and thinly peopled regions. In every instance indeed these great streams took their rise beyond the limits of Scythia proper, as defined by Herodotus, among the barbarous nations that are described as surrounding the Scythians, with whom the Greeks themselves had little, if any, intercourse. § 13. But if the account given by Herodotus of the five great rivers in question is tolerably clear and distinct, and their identification admits of no reasonable doubt, the case is alto- gether otherwise with regard to three other streams, all of which he distinctly places between the Borysthenes and the Tan "s, and to which he gives ‘the name of Panticapes, Hypa- cyris, and Gerrhus. Of these it may fairly be said that they have defied all the efforts of successive geographers to identify them with any known rivers, or to propose any plausible solu- tion of the difficulty. The Panticapes is described as rising in a lake, having a_course from north to south, therefore parallel with the Borysthenes, into which river it ultimately fell, apparently not far from its mouth. The intermediate space (which was a distance of three days’ journey from west to east 3) was occupied by the Husbandmen (Georgi) or agri- cultural Scythians, while the river in the lower part of its \ was from thence the north wind blew. reform, 36w 6 Bope'ns m/éa (De Aeris, rcée'raz yap (i7 Elwfiuc'h Xaip'n) fm’ az’rr‘fio'z Aquis, 850. 0. 95). 'rfim iipx'rowz, Kai 'ro'icn ob’pem, 'ro'irn Pl- 3 IV. 18. 186 HISTORY or ANCIENT GEOGRAPHY. can». VI. course traversed the land which he calls Hylma, or the Forest Region.4 Next comes the Hypacyris, rising also in a lake, traversing the country of the Nomad Scythians, leaving the Hylaea on the right, and falling into the Gulf of Carcine, opposite the town of that name.5 There is no river that at the present day at all corresponds with either of these; in fact, there is no stream of any magnitude between the Dnieper and the Don, except the Donetz, which is a tributary of the latter river, and does not approach the Palus Meeotis. The account is further perplexed by the statement that the Hypacyris, the most easterly of the two rivers, discharged its waters into the Carcinitic Gulf, which is unquestionably the Gulf of Perekop, west of the isthmus of that name. Still more enigmatical is the description of the third river, the Gerrhus, which, according to Herodotus, was a branch of the Borysthenes, separating from that river at the place called Gerrhus, which (as we have already seen) was the farthest point to which the Borysthenes was navigable, and was distant forty days’ voyage from the sea. In the lower part of its course the river Gerrhus separated the Nomad Scythians from the Royal tribe, and was in one part not less than seventeen days’ journey distant from the Borysthenes. Yet it ulti- mately, instead of falling into the Palus Meeotis, joined the Hypacyris The whole account is utterly unintelligible, and is not only at variance with the actual geography of these regions, but may fairly be said to involve physical impossibilities.6 It can only be explained on the supposition that Herodotus, who apparently never himself crossed the Borysthenes, was misled by the confused accounts of different travellers, of which, from his own imperfect ideas of physical geography, he was unable to see the contradiction.‘ 4 Her. iv. 54. It is remarkable that Ephorus, the next writer from whom we have any details concerning the land of the Scythians, describes in like manner a river Panticapes, E. of the Borysthenes, and forming the boundary between the Georgi, or agricultural Scythians and the nomad tribes beyond (Ephorus, Fr. 78, ed. C. Miiller; Scymn. Ch. vv. 843- 852). 5 iv. 55. 6 See Note E, p. 212. —___-______ SECT. 2. HERODOTUS : SCYTHIA. 1 87 § 14. It is almost equally difficult to identify the rivers alluded to in another passage, where he tells us that there were four great rivers, which had their sources in the land of the Thys- sagetee, beyond the desert that bounded the Budini on the north, and after flowing through the territory of the Meeetee, fell into the Palus Mmotis.7 These he enumerates in the following order: the Lycus, the Oarus, the Tanais, and the Syrgis. Of these the ‘Tanai's is of course well known, and the Syrgis is probably the same that he elsewhere calls the Hyrgis, though he there describes it as a tributary of the Tanais, not as falling into the Palus Maeotis. The Carus has been supposed by some modern writers to be the Volga, though that river in fact flows into the Caspian Sea, but there is really no clue to its identification.8 Of the Lycus nothing whatever is known. Even of the Tanai's itself it may be observed, that though it was certainly identical with the Don, it may be questioned whether the Greeks were sufficiently acquainted with the upper part of its course to distinguish the main stream from its tri- butary the Donetz, which is itself a large river, and has the more direct course from north to south, while the Don itself makes so great a bend to the cast, that it might readily be mistaken for a different river.9 § 15. \Vith regard indeed to all these streams it must be borne in mind that the geographical statements of Herodotus as to their sources, their course and their outflow, could be derived only from the reports of travellers, who had crossed them in their commercial journeys with caravans into the interior. The Greeks certainly had navigated the Borysthenes, the Hypanis and the Danube for a considerable distance from their mouths: but it is uncertain whether they had done so in the case of the Tanais, at least as early as the time of Herodotus}l and the 7 iv. 123. 1 The Greek colony of Tana'is, at 8 See Note F, p. 213. the mouth of the river of that name, 9 Rennell’s Geography of Herodotus, was certainly not founded till long p. 57. after the time of Herodotus. 188 HISTORY or ANCIENT GEOGRAPHY. can». VI. traffic from the Borysthenes eastward was in all probability carried on almost wholly (as it has been in modern times) by caravans of waggons across the steppes. A traveller accom- panying such a caravan would readily observe and remember the number of considerable streams that he crossed, and might report correctly their general direction, and their names, at least those by which they were known at the point where he crossed them; for it must be remembered that it is not uncom- mon for a river to bear different appellations in different parts of its course—but he would have to rely upon vague hearsay as to the points from which they came or to which they ulti- mately tended. Purely geographical questions of this sort have little interest for a semi-barbarous people, or even for a half-‘educated traveller, and the statement that all the four rivers last referred to ran into the Palus Maeotis may very pro- bably have been a mere conjecture hastily adopted by those who, like Herodotus himself, believed that sea to be nearly as large as the Euxine. § 16. Far more interesting and valuable than these professed geographical data, are the notices that Herodgtushasmreseryed ~— tonisefwthe—varionstribes-andsnatigmnsfit'hnat Erhabminted thevwiglg traetsmonutheflnorthfiof. the.-E11xine_.an_d,ih the adjpipjng, dis- trictsanilwAfiia... This was precisely the kind of information that he was likely to obtain most correctly from the class of travellers to whom alone he could have recourse. Traders, travelling solely for commercial purposes, would be likely to trouble themselves little about purely geographical questions, but their attention would necessarily be directed to the manners and customs of the tribes that they visited, as well as to the natural productions of their countries, and the uses to which they were applied. Nor could they fail to notice the diversity or identity of race among neighbouring populations, as attested by the use of different languages, requiring in consequence a succession of different interpreters. In one respect Herodotus is favourably distinguished from almost all his successors. With many ancient geographers the SE01‘. 2. HERODOTUSZ SGYTHIA. 189 term Scythians was applied in as wide and vague a sense as that of Tartars has been in modern times. Just as under the latter appellation have been included tribes of Mongolian, of Turkish, and even of Finnish extraction, so did the later writers on ancient geography frequently extend the name of Scythians to all the nomad nations that inhabited the northern regions of Europe and Asia. Herodotus, on the contrary, uses the term in a strictly ethnographical sense.2 With him the Scythians are a distinct people, differing from those that adjoined them on the east, north, and west,3 and Scythia is a definite area, of the extent and configuration of which he springs* to have formed to himself a distinct idea, though it is very difficult for us to represent to ourselves his conception “of it. § 17. One point, however, is certain. He tells us distinctly that Scythia was bounded on two sides, theflsdiith and east, by the sea.4 Hence it is clear that he regarded the Palus Meeotis ——as was done by all subsequent geographers, including even Ptolemy—as extending from south to north, and thus forming the eastern boundary of Scythia. Thew Tanai's, which pro- longed the frontier between Europe arid Asia and separated the Scythians from the Sarmatians, was also supposed to flow from north to south. These two sides he considered to be about equal, and supposed Scythia on the whole to be (of course speaking approximately) of a square form, extending inland about the same distance as the length of its sea-front, which he reckoned at twenty days’ journey, or about 400 G. 2 This remark, however, must be found elsewhere of this native appella- confined to the portion of his work (the tion; nor does it tend to throw any fourth book) in which he is treating light upon the ethnic affinities or de- specially of the Scythians. In another scent of the people designated by Hero- passage (vii. 64) he applies the name dotus under the name of the Scythians. of Scythians to the Sacee or Asiatic Few questions in ethnology have been Scythians, in the same manner as is more disputed than this. See Note H, usual with later geographers. p. 215. 3 He tells us that they called them- ‘ é’a'n yelp 'rfis Zlcuflmfis as. 8150 ,ue'pea selves Scoloti (Zmiko'ror), and that it 7651/ ot’z'pwv és doihao'o'av cpépov'ra, 'r'hv 're was the Greeks who gave them the 1rpbs ,ueaapfipinv Ital 7971/ 1rpbs as (iv. name of Scythians (iv. 6). No trace is 99). 190 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. miles.5 It is evident that this idea of the conformation of the country is so widely different from its real position and figure, that it would be a mere waste of time to attempt to discuss it in minute detail, or attempt to reconcile it with the natural boundaries. The important result is, t he con- sidered the country inhabited by the Scythian (properly so called) to extend only about 40_0___G;__miles inland,‘ whether measured from the Euxine or the Palus Maeotisfii - They there- fore may be considered as occupying the whole of Southern Russia, including the Ukraine, Podolia, and the country of the Don Cossacks, together with Bessarabia and Moldavia to the banks of the Danube.“ But their limits towards the i terior cannot be defined with any approach to accuracy.7 § 18. The Scythian people as conceived by the historian, was divided into several tribe , characterized by some difference in their modes of life and habits, as the Georgi or agricultural Scythians, the Royal, and the Nomad Scythians: to each of which he assigns a definite geographical position.“ But n11- fortunately he has described these positions and their re- spective limits, with reference to the rivers between the Borysthenes and the Tanai's, in a manner which involves them in the same hopeless confusion that attends the determination 5 o’s'o'n div 'rfis Ercvfimi'is dis éozio'ns 76- l 6 He reckons (iv. 101) from the Ister 'rpa'ya’wov, 7651/ 360 aepe’wv Ica'rmcdv'rwv e’s 6dAao'o'av, 1roiv'ry 100:1 1'6 1'6 e’s 'riw ,ueo'd- yaiav ¢épou real 1'?) 1rapa 1'1‘711 QciAao'o'ax/ (iv. 101). It seems diificult to understand how, in the face of these two passages, Mr. Rawlinson can say: “ The truth seems to be that Herodotus regarded Scythia as having only one of its sides washed by the sea ” (Herodotus, vol. iii. p. 209), when he distinctly states, twice over, that it was bounded by the sea on two sides. His whole conception of the geographical views entertained by He- rodotus on the subject appears to me fundamentally erroneous: to which his mistranslation of the statement con- cerning the mouth of the Danube ma- terially contributes. to the Borysthenes ten days’ journey, and ten more from the Borysthenes to the Palus Meeotis, calculating, as he himself tells us, 200 stadia to the day's journey. This would give for the southern, or sea, front, 20 days’ journey, equal to 4000 stadia (400 G. miles): and he adds that the distance from the sea to the Melanchleeni, who adjoined the Scythians on the north, was also 20 days' journey. Thus the extent of the boundary lines drawn towards the interior ('rd. r’ipfiia 'ra es 'riiv jueo'd'yarav (pawn) was the same as that of the boundaries on the other side (Ta ém- Icdpo’la), i.e. the two sides washed by the Euxine and the Palus Mseotis. 7 See Note G, p. 214. SEGT. 2. HERODOTUS Z SCYTHIA. 191 of those rivers. All that can be affirmed with certainty is that he placed in the western part of Scythia a tribe whom he characterizes as the Ploughmen -(Aroteres) because they cul- lllilatiehdthe. land in order to-raise corn for eaportmithoutnsing >it for their own subsistence”: while the Gepjgi (or Agricul- Lturists), east of the Borysthenes,V subsisted “6h the produce of their own_tillage, like all other nations. Beyond them towards the east were the Nomad Scythians, inhabiting an open steppe count'r'y ; and again beyond them, extending to the Palus Maeotis and the Tanais, was the tribe of the Royal Scythians, who looked upon all the others as their slaves or vassals. The whole tract occupied by these successive tribes was an open treeless plain, with the exception of a district (apparently of small extent) near the mouth of the Borysthenes, which was called in consequence Hylaea, or “ the Forest country.”8 § 19. Beyond the limitsithus assigned by Herodotus to the Scythians that people, as he tells us, was encircled by a series of different nations, extending from west to east, in the following order: the égathyrsi, the N euri, the Androphagi, the Melan- chl‘aeni, the Geloni, the Budini and thehSauromatae. All these tribes had their own separate rulers and were in the opinion of the historian distinct from the Scythians, though in some instances resembling them in their manners, and occasionally presenting a similarity of language. The Tauri also, who inhabited the peninsula that derived its name from them, were regarded by him as a distinct people from the Scythians.9 The first of these surrounding tribes was the ath 'rsi: a people who are distinguished by Herodotus as the most refined 8 The limits and position of this tract cannot be determined, for the reason already stated, that they are inseparably connected with the enig- matical rivers Panticapes and Hypa- cyras. But it appears to have been situated to the east of the Borysthenes and adjoining the sea. No extensive forest tract exists in this part of Russia at the present day: but the tradition still remained in the last century of some portions having been formerly covered with forests. (See Rennell’s Geography of Herodotus, p. 63, 4to ed.) Portions of the valley of the Borys_ thenes, where the river spreads into a variety of channels, are indeed, even at present, overgrown with trees, and the same thing is the case with the other rivers, though in a less degree. 9 iv. 99. 192 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. among them, and who were remarkable also for the abundance of their gold ornaments. They may be considered on plan- sible, though hardly conclusive grounds, as occupying the region now called Transylvania, the gold mines of which may probably have been worked from the earliest ages.1 Next to them came the w who resembled the Scythians in man- ners: but were said to have the peculiar power of transforming themselves for a few days every year into wolves.2 This story, of which Herodotus frankly expresses his disbelief, is curious as the earliest allusion to the superstition of the “ were-wolf,” so generally prevalent among the northern nations of Europe. Another circumstance mentioned by Herodotus that the N euri/ about a generation before the expedition of Darius, had been compelled to quit their homes for a time, on account of the multitude of sermnts, has at first sight a very fabulous air; but is in fact by no means improbable. A modern traveller,8 who visited the German colonies in the south of Russia, found them still full of recollections of the dificulties with which they had to cope, when they first settled there about 30 years before, from the multitude of serpents with which the whole country swarmed; and which only gradually gave way before the increase of tillage and population. The N euri are sup- posed by Schafarik, but on very slender grounds, to have been a Slavonian tribe : they ap arently occupied the country near the sources of the Dniestef,P the modern Volhynia. § 20. Beyond the N euri were the ndropha i, who, as their name imports, were cannibals, in which respect they stood alone among all the nations in this part of the world. Herodotus tells us that their manners were in all respects the most rude and savage with which he was acquainted: and that they spoke 1 In other respects (says Herodotus, question, a species of viper (Colaber iv. 1044) their customs are like those' of trabal'is), are said to grow sometimes to the Thracians. It is not improbable a very large size. But some of the that they were in fact a race of Thra- stories related to Kohl bear the stamp eian origin. 2 iv. 105. of exaggeration, and remind one of _3 Kohl, Reisen in Siid-Russland, similar legends in the early ages of vol. ii. pp. 153-156. The serpents in Greece. SECT. 2. HERODOTUS: SCYTHIA. 193 a peculiar language, distinct from the Scythian.4 All these particulars tend to support the conjecture of Neumann that they were a Finnish race: several of the tribes of that family being known to have retained the habit of cannibalism even in the middle ages.5 They were separated from the country of the Scythians by a desert tract of considerable extent, and seem to have ‘been situated nearly due north of___the...Greek settlemerflitskonnthe Borysthenes but howflfar inland we have no 5655113 of~determi-ning. Next to them—proceeding still from west to east—-came the Milanchlaani, or “Black-cloaks,” of whom we learn nothing, except that they were distinguished from the Scythians, whom they resembled in their other customs, by the constant use of the blackydréss, from which they derived their name.6 But though their manners were like those of the Scythians, Hero- dotus expressly tells us that they were a distinct and non- Scythian race. § 21. The next nation to the Melanchlaeni were the Budini, a people concerning whom we have more interesting informa- tion. They were (Herodotus tells us) a larggi and powerful nation, and were all of them distinguished by light. blue eyes and red hair.7 They were nomads like their neighbours on bbthfii‘desfbut their country was no longer the mere open 4 Herodot. iv. 106. frequent occurrence (see Plut. Cat. 5 Neumann, Die Hellenen im Sky- i Maj. i.), and exactly corresponds to the thenlande, p. 212. 1 “ coerulei oculi, rntilze comae ” of Taci- “ iv. 107. The Melanchlseni were i tus in speaking of the Germans (Ger- already mentioned by Hecatzeus. They i mania). But it by no means follows are noticed at a later period by Dion ' that we are therefore to suppose the Ohrysostomus (Orat. xxxvi. p. 78) who Budini to be of Germanic race, as has says that the Olbiopolites had derived been suggested by Manner-t, and par~ from them the fashion of wearing short tially adopted by Rawlinson. The black cloaks, which prevailed among Russians are spoken of by an Arabic them in his day. ' author as having “red hair and blue 7 There appears to me no doubt that eyes ;” and the Thracians are also de- this is the sense of the words of Hero- scribed as having the same character- dotus. Bovdfuoi 5% 501105‘ ébu ,ué'ya Kai istie. According to Humboldt (Asie noAAbv 77\av!c6v 're 1r3w ia'xvpés éo'rr Kai Centrale, vol. 1. p. 393), tribes with red meta (iv. 108). The combination of beards and blue eyes are mentioned in wept; in the sense of “red-haired” the Chinese annals as living in the with 'yAavlcos or 7Aavn6npwros is one of interior of Central Asia or Mongolia. VOL. I. O 194 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. steppe or plain: it contained extensive forests, as well as lakes and marshes which abounded with beavers and otters. In the midst of these dwelt a tribe called the Geloni, who were often confounded with them, so that the name of Geloni was fre- quently applied by the Greeks to the whole nation of the Budini 8——but incorrectly, as Herodotus assures us. According to him, the Geloni differed from the Budini in personal characters and in language, as well as in the habits of life. They were not nomads, but settled agriculturists, having gardens or orchards, and raising corn for their own consump- tion: they even possessed a city of large extent, surrounded by walls, built of timber, and containing houses and temples, also of wood. These temples, according to the informants of Herodotus, were dedicated to Greek divinities and adorned with altars and statues in the Greek fashion. In fact, he tells us, the Geloni were originally Greeks, who had quitted the trading-places on the coast and settled among the Budini ; and their language was a mixture of Greek and Scythian.9 It is very difficult to give credence to this last statement. It would be an unprecedented thing for Greek settlers to have established themselves at such a distance in the interior, with- out keeping up any communication or intercourse with their countrymen on the Euxine ; and the explanation suggested by Heeren that it was merely a Greek factory established thus far inland, for the purposes of trade, is wholly at variance with Herodotus, who evidently meant to describe these Geloni as a barbarian tribe, though more cultivated and civilized than, their neighbours; and retaining traces of their Greek origin in their religion and language. Very little value can be attached to the former ground of identification; and if the Geloni were a Slavonian tribe (as has been suggested with some plausibility by Schafarik)1 there would really be sufficient 8 Herodot. iv. 109. Both names are l 9 iv. 108. found in the later geographers, but with l 1 Schafarik, Slam'sche Alterthiimer, no additional particulars, and appa- vol. i. pp. 184-190. rently only derived from Herodotus. . SECT. 2. HERODOTUSI SCYTHIA. 19 5 resemblance between the deities of the two systems to lead readily to the belief. The wooden city, with its walls of wood, resembles very much the'accounts we find of the old Slavonian cities: that of Saratov, on the Volga, is said to have retained its ancient wooden walls and towers down to a late period? In the neighbourhood of that city, within the province to which it gives name, and between the Don and the Volga we may fix with reasonable assurance the position of the Geloni,» and the surrounding nomad tribes of the Budini. This is an important point for the determination of the geographical position of other tribes, to be hereafter mentioned. § 22. East of the Tanais, but south of the Budini, dwelt the Sauromatae, a nomad race inhabiting a region of open steppes, wholly destitute of trees, like the greater part of Scythia proper.3 They occupied a tract fifteen days’ journey in length, from the mouth of the Tana'is towards the north. Most modern writers are agreed in identifying the Sauromatae of Herodotus with the Sarmatee or Sarmatians of later authors, who at a subsequent period crossed the Tanais, drove out the Scythians from the steppes north of the Euxine, and ultimately extended themselves into the plains of Hungary and Poland, where, under the name of Slavs, they still form the bulk of the popu- lation. The legend related by Herodotus which ascribes their origin to a casual mixture of certain Scythians with a body of Amazons} is doubtless a mere myth, invented to account for the masculine and warlike habits of the women among them: and no dependence can be placed upon it, as evidence of any real connection of race between the Scythians and Sarmatians. But the case is otherwise with his statement of the resemblance between the two languages :5 this is a point upon which the Greek traders were likely to be able to judge, and is certainly a circumstance of importance in attempting to determine the ethnographical character of the Scythians themselves.6 2 Gtibel, Reise z'm S'z'id-Russland, 3 Herodot. iv. 21. cited by Neumann, p. 91. See also 4 iv. 110—116. Schafarik, l. c. p. 191. 5 lb. 117. 6 See Note H, p. 215. 02 196 HISTORY or ANCIENT GEOGRAPHY. can. v1. Such was, according to Herodotus, the series of nations or tribes by which the Scythians were immediately surrounded. In most directions these tribes formed the limit of his know- ledge and of the supposed habitable world. We hear nothing of any people beyond the Agathyrsi to the west: and he expressly tells us that the country to the north of the Neuri was, so far as he could learn, uninhabited. The Androphagi were themselves separated from the northern limit of Scythia by a desert tract of considerable extent; and beyond them again was an absolute desert, where no people were known to dwell. North of the Melanchlaeni again nothing was known but uninhabited deserts. It is evident therefore that in these directions neither the Greek traders nor their Scythian in- formants had penetrated any farther.7 §23. But it was otherwise towards the north-east. Beyond the Budini, who, as we have seen, were situated in the neighbour- hood of Saratov, between the Don and the Volga, there was also a desert tract, but only of seven days’ journey in extent; and beyond this, in a somewhat easterly direction, dwelt the Thyssagetae, a numerous and peculiar people.8 They, as well as the Iurcae,9 who were contiguous to them in the same 7 iv. 17, 18, 20. l etymological support for this : and all ‘5 E’Gz/os makev Ical 154011. Herodot. iv. I such inferences from mere names, the 22. He doubtless means by this ex- original forms of which we have no pression to intimate his own conviction means of discovering, appear to me in that the Thyssagetae were a distinct the highest degree hazardous and un- people from their neighbours, Scythians certain. and others. But of course this does not exclude the probability of ethno- graphical relations between them. The name has been regarded, naturally enough, as indicating a connection with the Massagetae, who were not far dis- tant towards the south-east; and both have been supposed to be related to the Getae of the Greeks, the supposed ancestors of the Goths (Donaldson, Varronianus, p. 41, 2nd edit). Professor Rawlinson even goes so far as to assume boldly that the Thyssagetae were the “lesser Goths" as distinguished from the Massagetee or “greater Goths” (vol. iii. p. 19, note). But there is no See some judicious remarks of Niebuhr (Researches, p. 82). Hum- boldt, however (Kosmos, vol. ii. p.176), adopts the Gothic hypothesis with re- gard to the Massagetae without hesi- tation. 9 The name of these Iurcae has been transformed by Pliny (vi. 7, § 19) and Mela (i. c. 19,§ 116), both of whom are obviously . copying Herodotus, into Turcae, and they have in consequence frequently been supposed to be iden- tical with the Turks. But there is no reason to suppose the name of Turks to be anything like so ancient as this; and the coincidence is probably more accident, having apparently originated $3013 2_ HERODOTUSZ SOYTHIA. regions, were hunters, and subsisted by the chase: their whole country being overspread with forests. Beyond them again still bearing towards the east, dwelt a Scythian tribe, who were believed to have formerly emigrated from the land of the Royal Scythians, on the Euxine, and established themselves in these remote regions. i Thus far, we are told, the country was level and fertile: but after this it became stony and rugged. After traversing a considerable extent of this rugged tract there occurred a people called the Argippeei, dwelling at the foot of a range of lofty mountains, who resembled the Scythians in dress, but spoke a peculiar language. They had fiat noses and projecting chins, or jaw-bones, and were always bald from their birth upwards.1 Their principal subsistence was derived from a kind of fruit, which they dried in solid masses, after straining off the juice, which they drank mixed with milk. This is precisely the same process used by the Galmucks of the present day in preparing the fruit of the bird- cherry (Prunus Padus), which with them also forms an im- portant article of diet. The description of the physical characters of the Argippeei also resembles that of the Kalmucks, and other Mongolian tribes, as these would present themselves to an unscientific observer. The baldness indeed is not really common to the whole race, but peculiar to the sacerdotal caste: and must have been erroneously transferred to the whole people. It was doubtless the same confusion that gave rise to the notion preserved by Herodotus that the Argippeei were a sacred race, who never made use of arms and had no occasion for them, as no one ever attacked them : but the neighbouring tribes referred their disputes to them as arbiters.2 The mountains, at the foot of which the Argippeeans were settled, can hardly be any other than the Ural :3 and if we only in a false reading of Herodotus. l Minor. But the similarity of name On the other hand, it is certainly tempt- ' may well be in this case also merely ing to connect the name of the Iurcee casual. with that of the Yuruks, a wandering 1 iv. 23. 2 Ibid. race of herdsmen akin to the Turco- 3 They are indeed supposed by Hee- mans, who roam over the plains of Asia ren (Asiatic Nations, vol. ii. p. 272) to 198 HISTORY or ANCIENT GEOGRAPHY. CHAP. v1. place them on the western or south-western slopes of that chain, there will remain abundance of room for the different tribes enumerated by Herodotus between them and the Budini. It is indeed impossible to assign to each of these tribes its precise geographical limits or position, but we trace distinctly, as we follow the course indicated by Heroadotus, the transition from the open steppes to the forest-covered regions,—-which were certainly far more extensive in ancient times than they are at the present dayf—and from thence to the rugged hilly tract immediately adjoining the mountains. § 24. The Argippaeans were the last nation towards the north, with which the Greek settlers on the Euxine or their Scythian neighbours, had any direct communication. They were cut off on the north by an impassable barrier of lofty mountains; beyond which no one had ever penetrated : but the Argippaeans reported them to be inhabited by men with feet like goats ;— and beyond these again were a people who slept for six months in the year.5 In these traditions—naturally treated by Hero- dotus himself as mere fables—it is not difficult to discern a germ of truth. The one is no more than a natural hyperbole to express the habits of an active race of mountaineers; the other is obviously founded on the well-known facts of the pro- longed winter night of the Arctic Regions :6 though, like most similar traditions, it was exaggerated far beyond the truth. be the Altai, and the same view is adopted by Humboldt (Asie Centrale, vol. i. pp. 389-407). But, notwith- standing these high authorities, it appears to me far more probable that the Ural Mountains are these really meant. The course followed by Hero- dotus is clearly towards the NE. (ac- cording to his own conception it was nearly N. but verging somewhat to- wards the E.), and this direction must inevitably lead to the chain of the Ural. These mountains could not have been overlooked or ignored by the in- formants of Herodotus, and would con- stitute a natural barrier to all further communication in this direction. It is true that the chain of the Ural in reality extends from S. to N. and not from W. to E. as Herodotus apparently conceived the mountains in question to do, but there are few points upon which the information of ancient writers was so often inaccurate or erroneous, as this of the direction of mountain ranges. 4 See this point fully examined and discussed by Neumann, pp. 85—98. 5 iv. 25. 6 It is, however, justly observed by a recent traveller (cited by Mr. Rawlin- son) that the statement as reported to Herodotus was not that there was a night of six months’ duration, which of course is only true at the pole itself, SECT. 2. HERonorUs : SCYTHIA. 199 But while the country north of the Argippeeans was thus unknown to the Greeks, it was otherwise with the tract that lay to the east of them. Here dwelt the Issedones, a people who are represented as just and humane in other respects, but having the custom of eating the bodies of their deceased fathers, whose skulls they afterwards gilt and preserved as objects of reverence.7 The name of the lssedones was certaiw familiar to the Greeks long before the time of Herodotus. It was mentioned by Hecataeus of Miletus, as well as by Aristeas, who professed to have visited them himself.8 Whether or not this was true, it may be taken as showing with certainty that they were already known to the Greeks on the Euxine in his time. They are placed by Herodotus‘, as we have just seen, east of the Argippaei: in another passage he tells us that they were situated “opposite to ” the Massagetee, whom he regarded as occupying an extensive tract to the east of the Caspian Sea.9 The Issedones would in this case be due north of them,—-a meaning which would suit very well with the passage in question—and may be placed in the broad steppes to the south-east of the Ural Mountains—now inhabited by the hordes of the Kirghiz. § 25. The regions north of the Issedones were equally un- known with those to the north of the Argippaeans. “But it was ' here that they placed the fabulous Arimaspians, ahidpthe equally fabulous Griffins. Herodotus is distinct in his assertion that this story was told by the Issedones, from whom the Scythians but that the inhabitants slept for that period, which may be understood as arising from their keeping closely within doors through the long and dreary winter nights. A similar tradition as to the “land of darkness ” in the far north was cur- rent among Oriental nations (see a note by Sir H. Rawlinson on Herodotus, iv. 25), and is mentioned by Ibn Batuta, who was informed that it was distant forty days’ journey from the city of Bul- gar, which he visited. (Travels of Ibn Batuta, translated by Lee, p. 78.) How easily such a statement might arise from exaggeration of a true fact is shown by Pliny in his account of Thule, where, after stating that there was no night there at the summer solstice, and no day at the winter solstice, he adds, “ Hoe quidam senis mensibus continuis fieri arbitrantur” (iv. e. 16, § 104). 7 iv. 26. 8 iv. 13. See above, Chap. IV. 9 i. 201. The conclusions to be drawn from this passage will be dis- cussed in the next chapter. 200 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. learnt it, and that it passed from them to the Greeks.1 If any reliance can be placed on this statement, it would appear pro- bable that the fable was connected with the fact of the abun- dance of gold on the flanks of the Ural and Altai mountains, the streams of which may have been worked by people in a very primitive state of civilization.2 Herodotus himself, while rejecting as altogether fabulous the story of the grifi’ins and the one-eyed Ar_i_m’a_s,pians, admits as a certain fact that by far the greatest quantity of gold came from the north of Europe (under which appellation he includes the tracts in question), “though how produced (he adds) I know not.”3 The abun- dance of gold among the Scythians on the Euxine, and its extensive use by them, is sufliciently attested by the contents of their tombs, which have been opened in modern times. Whatever therefore may have been the origin of the fable, the fact that large quantities of gold were brought from the interior of Northern Asia to the coasts of the Euxine is one that admits of no doubt. To this fact we are probably indebted for the great extension of the geographical knowledge of the Greeks in this direction. The extent and accuracy of th information collected by Herodotus concerning the region which lay to the north-east of Scythia, as compared with those which bounded it on the Qlorth-west and north, is very remarkable, and points, as was long ago remarked by Heeren,4 to the existence of a well- established caravan route in that direction. But the commo- dities that can be carried long distances by caravans must necessarily be light, and readily portable. Gold was exactly a production of this kind, adapted for ready transport; and the gold-bearing regions of Northern Asia would naturally exercise a strong power of attraction upon the Greek settlers on the Euxine, as well as on the more civilized Scythian 1 iv. 27. He adds that the name of 2 See the remarks of Humboldt on Arimaspians was Scythian: for that in this subject (Asz'e Centrale, v01. i. pp. the Scythian language “arima” meant ] 330-408). 3 iii. 116. one, and “ spou,” an eye. , 4 Asiatic Nations, vol. ii. p 285. SFCT. 2. HERODOTUS: SCYTHIA. 201 tribes, who adjoined that sea. The only other production of these northern regions that was likely to be sought after by visitors from the south was their furs, which would be an indispensable article to the Greek colonists in Scythia, however little they were used to require them at home. § 26. It is a remarkable instance how little such sources of information as were open to Herodotus could be relied on'for correct geographical knowledge, that, while he obtained these interesting and valuable particulars concerning the various nations and tribes that were successively met with in pro- ceeding towards the north-east, no mention is found of the great river Volga, which must have been necessarily crossed on the way, and which was in fact a larger stream than either the Tanai's or the Borysthenes. The supposition that the Oarus, mentioned by him in another passage among the rivers flowing from the land of the Tliyssagetae, was really the Volga, would not mend the matter, as there is no indication that he regarded that river as of any greater importance than the other streams with which he associated it.5 §27. On the other hand, the knowledge which Herodotus had acquired of these countries would naturally lead him to the inference, even if he possessed no specific information on the subject, that the Caspian Sea must be an inland sea, sur: rounded on all sides by land. This, indeed, results of neces- sity from the data with which he himself furnishes us. For he distinctly places the Massagetee on the eastern side of the Caspian : 6 these adjoined the Issedones on the north, and the Issedones again were connected through the Argippeei and an unbroken chain of nations towards the west, with the Budini and the Scythians. But apart from this train of reasoning, it is probable that Herodotus had derived from some other 5 It seems not improbable that the land of the Budini, and might be sup- Volga was in fact confounded with the posed by him to be only two branches Tana'is: the two streams approach near of one great stream. But in any case one another in the part of their course the omission is a remarkable one. where they would naturally be crossed 6 i. 201. by a traveller proceeding through the /202 HISTORY or ANCIENT GEOGRAPHY. CHAP. VI. .> \w“ source more definite information concerning the Caspian Sea, as he not only states distinctly,7 as a fact, that it is a separate sea, not communicating with any other, but he gives an estimate of its length and breadth, which implies that it was habitually navigated.8 In this respect it is remarkable that Herodotus was in advance of almost all his successors, who, from Eratosthenes to Pliny, agreed in describing the Caspian as an inlet from the outer Ocean. It is not till the time of Ptolemy that we find the correct view again prevailing. SEOT. 3.——Eapedition of Darius into Scythia. § 1. The expedition of Darius against the Scythians is undoubt- edly entitled to be regarded as an historical fact, however difficult it may be to admit many of its details as historically accurate. But we havehere only to consider it from a geographical point of view; and in this respect we shall find that the narrative of it, as related by Herodotus, accords extremely well with the geographical account of Scythia and the adjoining lands given by the same author. So close, indeed, is the coincidence that, when we bear in mind how much vagueness and uncertainty still hung about the regions in question, it is impossible to avoid the inference that the two have been adapted to each other. Either the narrative of the campaign has been fitted into the geographical views obtained from other sources, or those views have been themselves derived from the informa- tion obtained during the expedition. The latter alternative is that adopted by Major Rennell, who supposes that Hero- 7 i. 203. 8 Herodotus gives the length of the Caspian as fifteen days’ voyage for a rowing vessel (er’peaip Xpewuévcp, Ib.), and the greatest breadth as eight days. These proportions are very nearly cor- rect, and I agree with Mr. Rawlinson that there is no reason to suppose (as has been suggested by some critics, in- cluding Niebuhr) that he misconceived the position of the Caspian, and reckoned its greatest length from E. to W. instead of from N. to S. This was certainly done by later geographers, but from whatever source Herodotus de- rived his information concerning this sea he certainly appears to have had a more accurate idea of its extent and configuration than any of his successors before Ptolemy. SECT. 3. HERODOTUSI SCYTHIA. 203 dotus “ drew his materials for the inland part of the geography [of Scythia], scanty as they may be, from the history of this expedition.” 9 But the historical details of the narrative are such as in themselves to present insurmountable difficulties to our receiving it as an accurate account of the operations of the Persian army; and the particulars previously given by Herodotus of the different nations that were successively visited by the invader, resemble much more the impressions that might have been gathered from the Scythians themselves, than such as would be derived from a rapid hostile incursion, traversing the lands in a hasty manner, and almost without seeing an enemy. § 2. The substance of the narrative of Herodotus is briefly this.1 Darius having determined to take vengeance upon the Scythians for their inroad into Media (about 130 years before), assembled a great army, amounting to 700,000 men, with which he crossed the Bosphorus by a bridge of boats, and advanced through the land of the Thracians and the Getae to the Danube, where another bridge of boats had been already prepared by the Ionian and other Asiatic Greeks, who fur- nished the greatest part of his fleet. From thence he plunged at once into the wilds of Scythia, leaving the Ionians to guard the bridge, and thus, in case of necessity, secure his retreat. . But they were only ordered to remain at this post for sixty days. According to a preconcerted scheme, the Scythians offered no opposition to the advance of the Persians, but con- tinually retreated before them, laying waste the country as they went, and directing their line of march eastward towards the Tanais. This course they continued till they came to that river, which they crossed, and the Persians after them, still continuing the pursuit. In this manner both armies traversed the land of the Sauromatae, and entered that of the Budini. Here the Persians found the wooden fortress, already men- tioned, which they burnt to the ground, and continued to 9 Rennell’s Geography of Herodotus, p. 107, ‘Ito ed. 1 Herodot. iv. 120-142. 204 HISTORY OF ANCIENT GEOGRAPHY. OHAP. VI. pursue .the Scythians through the land of the Budini to the confines of the desert beyond. Here Darius at length made a halt, on the river Oarus, where he erected eight large forts, at an interval of about 60 stadia apart, the remains of which were said to be still visible in the days of Herodotus.2 From hence he turned westward, and having fallen in with other bodies of the Scythians, which retreated before him in like manner, was thus led on, first into the land of the Melanchlaeni, afterwards into that of the Androphagi, and then of the N euri. The Scythians intentionally led the Persians by this great circuit through the territories of those different nations, because they had previously refused to make common cause with the Scythians and unite in repelling the Persian arms. When they now found themselves threatened by this double invasion, they broke up in terror and confusion, and fled into the deserts towards the north. The Agathyrsi alone took up arms in their own defence, and, posting themselves on their frontier, forbade the entrance of the Scythians, who thereupon turned aside and re-entered their own land. Various attempts at negotiation now took place, but without effect ; and at length Darius, finding his troops continually harassed by the Scythians, without any opportunity of striking a decisive blow, determined to retreat to the Danube, which he re-crossed in safety. The Scythians had indeed made two attempts to persuade the Ionians to destroy the bridge; and the term of sixty days had already expired, but the Ionian leaders judged it for their own advantage to secure the retreat of Darius, and consequently maintained the bridge in such a state that it was restored without difficulty. § 3. It is scarcely necessary to point out the difficulties and objections that present themselves at every step, if we attempt to consider this relation as an authentic historical narrative, 2 iv. 124. 'ré‘w é’n ml e’s e’pé Ta e’pei- remains of earthworks really existed on ma 013a 571/. Of course this was mere the banks of the Oarus, these would be hearsay, though the expression would naturally connected by the native tra- certainly lead one to suppose that He- dition with the name of Darius. rodotus had himself ‘seen them. If any SE01‘. 3. HERODOTUSZ SCYTHIA. 205 like that of the campaigns of Alexander the Great; or the expedition of the younger Cyrus. As Mr. Grote remarks, Herodotus conducts the immense host of Darius from the Danube to the Tanais “ as it were through fairy land,” heedless of the distance, of the great intervening rivers (of which he himself elsewhere speaks as one of the great marvels of Scythia), and of the difficulty of obtaining supplies for a vast army, in a country almost devoid of natural resources, and in which whatever was to be found had been studiously destroyed?‘ The distance from the Danube to the Tanai's was, according to Herodotus’s own conception, forty days’ journey for an ordi- nary traveller ; 4 it could hardly, therefore, have been less than sixty days’ march for an army; and yet Darius is represented as traversing the whole of this distance, apparently without a halt, and without a mention of any obstacle; then proceeding through the country of the Sauromatse and the Budini to the river Oarus; halting there long enough to erect important works of fortification (the object of which is utterly unintelli- gible), and then making a vast circuit through the nations to the north of the Scythians, till he returned once more into their country, apparently at no great distance from the point at which he started. It was not till after the Scythians had led him this wild-goose chase through the desert regions of the north, that we are told they despatched messengers to the Danube to try the fidelity of the Ionians, but failed in the first instance because the appointed sixty days had not yet elapsed ! But such a march as he is represented as having made could hardly have been accomplished, even supposing it feasible at all, in less than an hundred and fifty days, or five months.5 3 Grote’s History of Greece, vol. iv. p. 355. 4 He computes the sea front of Scy- thia from the Danube to the Maeotis at 20 days’ journey (iv. 101), and the side perpendicular to it, parallel with the Meeotis (according to his conception) was of equal length. Of course if Darius marched direct to the Tana'is he would take the diagonal of the square, and thus cut off a portion of the supposed distance : but this it was impossible for him to do in reality. 5 This is Major Rennell’s calculation (vol. i. p. 150), who therefore assumes that there must have been some error, when we are told that the 60 days had not elapsed. But the question of 206 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. Now the duration of the expedition is precisely the point upon which Herodotus was most likely to be well informed, for this was the circumstance that must necessarily impress itself the most strongly upon the minds of the Ionians and other Greeks who kept watch at the bridge. His account of their pro- ceedings has every appearance of being derived from good information; while that of the operations of the Persian army could only be derived either from the vague reports of those who returned after wandering in the unknown regions of Scythia, or (more probably) from the exaggerated accounts current among the Scythians themselves, and picked up from them by the Greek colonists on the Euxine. We may safely pronounce that narrative, as transmitted to us by Herodotus, to be impossible and unintelligible ; 6 but this does not affect the geographical account of the regions in question, which is both intelligible, and consistent with the statements he has previously made concerning the order and relative positions of the different nations through whose territories Darius was supposed to have been led. time is an essential element in the story, as told by Herodotus. Mr. Rawlinson indeed says (vol. iii. p. 118, note) that we do not know the whole time employed on the expedition. But the narrative certainly excludes the supposition that any long interval elapsed between the first visit of the Scythians to the bridge (when the 60 days had not yet elapsed) and the second; and the whole of the great circuitous march, by which Darius re- turned into Scythia, is distinctly repre- sented as having taken place before the first message was sent. 6 See Note I, p. 217. SKETCH OF SCYTHIA ACCORDING TO HERODOTUS IV THYSSAGETAE IYRCAE ‘I SAUROMATAE ? 0 9. $0 (Q i Q - 0». $ "\7 a8 at NEURI SI 5'— 4.0005tdia Y‘ --------- -- ‘i . $ 0 4 i h a b S C Y T H LN t he a f .. g S? q] \ Q. o ‘i -v.- 1,. m a Q tel: :n 5a, m a p} t; 6’ ,5 Q ‘1% [in 9v.- t. .s s ‘ c1 s 0 Q3 2 . 015a,} HYLAEA ___.._ ~ I I‘ 1 Y :5!“ 1mm? .1 (n'qnfEa'mh I London : John Mu rray NOTE A. HERODOTUS2 SCYTHIA. 207 NOTE A, p. 161. THE ACTZE OF HERODOTUS. WE have unfortunately no English word which conveys the exact meaning of the Greek ’Alm‘7, a clear comprehension of which is essential to the understanding of this passage, as well as of some others in the Greek geographers. It does not correspond either to a mere promontory (though used as such in poetical Greek), or to a peninsula strictly so called, but is something intermediate between the two. The term peninsula is indeed at the present day often vaguely applied to large tracts of land connected with the con- tinent by a broad belt, not a mere isthmus. Thus we speak of the Spanish peninsula, the Scandinavian peninsula, &c.; and as thus employed, the word approaches closely to the meaning of the Greek ’Al<~n‘7. But the latter term was applied also—and perhaps more properly—to a mere advanced tract of land, projecting somewhat in the manner of a wedge from the mainland, and ending in a promontory. The two most characteristic instances of this meaning of the word are the dK'r'i] of Attica—from which, according to some writers, that country derived its name—and the Argolic Acte, the tract of land projecting between the Saronic Gulf and the Gulf of Argos, and including the cities of Epidaurus, Troezen, and Her- mione. On a large scale the country which we new term Asia Minor answers exactly to the first sense of the term ; more especially as Herodotus supposed the neck connecting it with the rest of Asia to be much narrower than it really is. Hence there is no difiiculty in understanding this, his first Acte. But it is more difficult to comprehend its application in the second instance. Herodotus, however, appears to have conceived Arabia, together with Syria and Assyria, as forming one great dim), of which, strange as it appears, he regarded Egypt and Libya (or the whole continent of Africa) as a mere appendage, attached to it by the Isthmus of Suez. This is much as if the whole of Greece (including Thessaly and Epirus) were considered as one great Acts, of which the Pelopon- nesus was only a subordinate part, attached to it by the Isthmus of 208 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. Corinth. But the disproportionate size of Africa, even acogrding to the conception of Herodotus, renders such a view still more extraordinary. NOTE B, p. 173. THE CIMMERIANS. It would be a very interesting ethnographical point to determine who these Cimmerians were ; and it is obviously very tempting to identify them with the Cimbri and Kymry that appear at a later period in the west and north of Europe. But such inferences from mere resemblance of name are very apt to be misleading, and our materials are too scanty to enable us to attain to any satisfactory conclusion on this subject. While the general facts of the Cim- merian invasion of Asia, and their ultimate expulsion from their abodes on the north of the Euxine, may be regarded as well attested, it must be admitted that the details furnished by Hero-' dotus appear in several respects to be untrustworthy. The main body of the Cimmerians, according to his account, fleeing from the Scythians, held their route along the eastern shore of the Euxine, between the Caucasus and the sea, a rugged and difficult tract, almost impassable for an army, and still more so for a migratory tribe like that of the Cimmerians. The pursuing Scythians, on the contrary (he tells us), missed their way, and kept the Caucasus on their right hand, so that they fell into Media, and thus became by accident the destroyers of the Median monarchy. Admitting the destruction of the Median monarchy by a Scythian invasion to be historically true, it seems impossible to believe the story thus told to connect it with the Cimmerian invasion of Lower Asia, though, as Mr. Grote observes (History of Greece, vol. iii. p. 332), it is not improbable that both movements were connected with some sudden development of the Scythian power or propulsion by other tribes behind them. It is remarkable that the existence of such a simultaneous move- ment is affirmed in all the traditions connected with the Cimmerian invasion. Thus Aristeas of Proconnesus, in whose time (about B.0. 550) there were Cimmerians still remaining to the north of the Euxine, stated that the nations of the interior were all pressing upon one another, each urging the other onwards from north to south; the Issedones in the far north were being gradually driven NOTES B, C. HERODOTUSI SCYTHIA. . 209 out by the Arimaspians, and in consequence were in their turn pressing on the Scythians, and these again on the Cimmerians, who, being bounded by the sea on the south, could not transmit the pressure in their turn, and were compelled to leave the country (Herodot. iv. 13). Herodotus himself, indeed, gives a different account of the origin of the movement. According to this, the first shock came, not from the north, but from the east, where the Massagetae, an Asiatic nomad tribe, pressed so severely upon the Scythians,- who at that time dwelt wholly in Asia,——that the latter gave way before them, and, crossing the river Araxes, invaded the land of the Cimmerians, who felt themselves unable to resist so formidable a host, and abandoned their country without a contest (Id. iv. 11). But whatever may have been the cause of the movement, the emigration of the Cimmerians appears to have been complete in the time of Herodotus, who no longer found any Cimmerians settled to the north of the Euxine, though, as he himself points out, there still remained manifest traces of their previous occupation of the country in the term “ Cimmerian,” attached by tradition to earthworks and other relics of the past, as well as in the names of localities, especially that of the Cimmerian Bosporus, as the strait between the Crimea and the mainland of Asia was uni- versally called, to distinguish it from the Thracian strait of the same name (Id. iv. 12). NOTE C, p. 177. .THE GREEK STADIUM. As this is the first occasion on which I have had occasion to refer to the Greek mode of computation by stadia or stades, I may take this opportunity of stating that throughout the present work I shall uniformly assume that the Greeks employed but one measure under that designation, which was, as stated in the text, a hundred fathoms, or 600 Greek feet (Herodotus, ii. 149). This has been proved, in my opinion, beyond a doubt, by Col. Leake in his paper On the Stacie as a Linear Measure, first published in 1839 in the‘ Journal of the Geographical Society, vol. ix., and republished in his treatise On some disputed Questions of Ancient Geography, 8vo. VOL. I. P ZIO HISTORY OF ANCIENT GEOGRAPHY. OH'AP. VI. Lond. 1857.7 But, in fact, the circumstance that neither Strabo nor any other ancient geographer adverts to the existence of any doubt on the subject, or to the use of any other than the ordinary Greek stadium, is conclusive in regard to the question. The incon- gruities and inaccuracies of different ancient writers in regard to distances have led many modern inquirers to the idea that they must have employed diiferent scales of measurement, and therefore. stadia of different lengths. Any one who has looked into the writings of D’Anville or Gosselin will be familiar with the con- fusion and diificulties arising from this source; and even Major Rennell fell into the same error, and thought it necessary to suppose the existence of an itinerary stade, distinct from the true one,8 rather than admit the fact, confirmed by general experience both in ancient and modern times, that distances which are only estimated, and not measured, are almost always overrated. At the present day the controversy may be considered as settled. Dr. G. Miiller, in his valuable edition of the Geographi Grzeci Minores, takes for granted that in all cases a stadium of 600 Greek feet is meant; and M. Vivien de Saint-Martin, in his most recent work (Histoire de la Geographic et des Décoavertes Géographigues, 8vo, Paris, 1873)., admits that the Olympic stadium of 600 feet was the only one in general use among the Greeks, and is therefore always to be understood when distances are given in stadia. Another source of error was introduced at a later period by the conflicting results of the attempts of mathematical geographers to determine the circumference of the earth. Eratosthenes was thus led to the conclusion that a degree of the circumference was equal to 700 stades; while Posidonius, who was unfortunately followed by Ptolemy, calculated only 500 stades to the degree. But here, as we shall see more fully hereafter, it was the computation of the degree that was erroneous, not that there was any difference in the scale of measurement. A stade of 600 Greek feet was in reality very nearly the 600th part of a degree ; ten stades are consequently just about equal to a nautical or geographical mile of 60 to a degree ' It had indeed been already clearly STADIUM in Dr. Smith’s Dictionary of established by Ukert (in his Geographfe Greek and Roman Antiquities, p. 893. der Grieehen and Homer, vol. i. pt. ii. pp. 8 See his Dissertation on the Itinerary 51—72), but his work was unknown to Stade of the Greeks, inserted in his Geo- Leake. A good summary of his argu- graphy of Herodotus, Ohap. II. ments will be found in the article _ NOTE D. HERODOTUS I SCYTHIA. 2 I I --a circumstance that materially facilitates the reduction of the measures given by ancient geographers into such as are familiar to the modern reader. NOTE 1), p. 183. NAVIGATION OF THE BORYSTHENES. This statement has given rise to much discussion. Several of the editors of Herodotus have suggested that the number should be fourteen, and this is strenuously maintained by Neumann (Die Hellenen im Shythenlande, p. 205). But it appears to me certain that Herodotus wrote forty, as,‘ independently of the agreement of all existing MSS. of our author, the statement is repeated by Scymnus Chins (v. 816), as well as by Pomponius Mela, both of whom clearly derived it, directly or indirectly, from Herodotus. We are not entitled to alter the reading of an ancient author merely because it apparently involves an error in fact. Nor is there any reason to assume that fourteen days would be a more correct statement. Herodotus is clearly speaking of the voyage up the stream (,ae'xpi ,ue'v vvv Felfifiov Xo'ipov he 'rdv Teo'o'epoiKor/Ta iylaepe'wv ‘ix-Mics 30-11’, ytvo'io'Ke'rat pe’wv (ivro Bopéw dvéuov, iv. 53), and the duration of this we have no means of measuring; but there is certainly no improbability in its taking double the time, allowing for the windings of the river, that would be occupied by a direct journey to the same point by land. There is therefore no real discrepancy, as has been assumed, between this statement and the assertion that the Scythian terri- tory extended inland only twenty days’ journey. A more serious difficulty arises from the fact alluded to in the text, that the navigation of the Dnieper is obstructed, at a distance of about 260 miles from its month, by a succession of cataracts or rapids, “ which limit the passage to the time of high water during the spring, and even then attended with some difliculty, and only of a fortnight or three weeks’ duration ” (Clarke’s Travels, vol. ii., Appendix, p. 465, 8vo. ed.). As, however, the river is ascended by barges in the spring through its whole course, it is easy to suppose that the Scythians may have had a knowledge of it above the cataracts, whatever may have been the difficulty of its navi- gation. It is curious—and inexplicable—that Strabo (vii. c. 3, § 17, p. 306) speaks of the Borysthenes as navigable only for 600 stadia P 2 2I2 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VI. from its mouth, though below the cataracts there can never have been any obstacle to the navigation, which is indeed particularly easy. (See Clarke’s Travels, l.c.) NOTE E, p. 186. RIVERS OF SCYTHIA. Professor Rawlinson justly observes that “ there is the utmost uncertainty with respect to all identifications east of the isthmus of - Perekop,” the Tanai's alone forming the exception. It is simply impossible to accept the statements of Herodotus as they stand, and even the favourite resource of modern commentators—to suppose that great physical changes may have taken place in the countries in question—will do little to remove the difficulty. The main point of all is the Gerrhus, which is clearly described as branching off from the Borysthenes, at the place of the same name, which was the farthest point of that river that was known to the Greeks (iv. 56). An ancient geographer would see no impro- bability in this, as such bifurcations were supposed to exist in other cases, as that of the Ister, which was believed by geographers much more advanced than Herodotus to send off an arm to the head of the Adriatic, while the main stream flowed into the Euxine. But there is in fact no such case known in physical geography. The supposition of Professor Malden (cited by Sir R. Murchison, Russia and the Ural Mountains, vol. i. p. 57 4) that the Borysthenes may have originally formed a delta, and reached the sea by two separate months, is not destitute of plausibility; but it would offer a very inadequate solution of the difficulty. For the formation of such a delta is only possible at a short distance from the mouth of a great river, or where (as in the case of the Euphrates and Ganges) it flows for a long distance through marshy and alluvial lands. But the region where Herodotus places the river in question is the steppe country, which is throughout considerably elevated above the beds of the streams that traverse it. The formation of a gigantic delta in such a country may be safely pronounced to be physically impossible. And whatever may be. thought of the distance assigned by Herodotus from Gerrhus to the sea (see Note D), the bifurcation must have taken place, according to his view, NOTE F. HERODOTUS-i SCYTHIA. I 3 a long way up the course of the Borysthenes, as he supposes the two rivers Panticapes and Hypacyris, both of them considerable streams (wo'raptoi. oiivoptao'rot', 0. 47 and 58), and taking their rise in tWO separate lakes, to have their sources and their whole course between the other two rivers, or rather arms of the same river! And he tells us, moreover, that the country of the agricultural Scythians (the Georgi), which was situated between the Panticapes and the Borysthenes, was three days’ journey in width, and extended up the Borysthenes for eleven days’ voyage (iv. 18), while that of the nomad Scythians extended fourteen days’ journey between the Panticapes and the Gerrhus (lb. 19). Even if We abandon the attempt to explain or reconcile these strange statements concerning the course and connection of the three rivers, the difficulty remains that nothing corresponding to them can be found between the prescribed limits. No streams of any considerable importance are found between the Dnieper and the Don (except the Donetz, a tributary of the latter) ; and those which exist all flow into the Sea of Azov, not into the Euxine. Nor do the statements of later writers throw any light on the subject. Both Pliny and Mela mention the name of the Panticapes, but their accounts of these Scythian rivers are a mass of confusion, and Dionysius, who describes it as flowing from the Rhipeean Mountains (Periegesis, v. 315), is in direct contradiction with Herodotus. NOTE F, P. 187. THE RIVER OARUS. ‘ The identification of the Carus of Herodotus with the Volga was adopted by Major Rennell (Geogr. of Herodotus, p. 90, 4to. ed.), and has been accepted by most recent commentators. But it certainly rests on no adequate authority. If, indeed, the details of the expedition of Darius against the Scythians could be regarded as trustworthy, the supposition that the river Carus, which was the limit of his progress towards the cast, was the same with the Volga, would not be devoid of plausibility; but, as we shall hereafter see, those details are so clearly unworthy of credit, that no dependence can be placed upon this argument, and there is really no other. The supposed resemblance of the name Carus to the Rha of Ptolemy, which is certainly the Volga, is so slight as to have no weight at 214 ANCIENT HISTORY OF GEOGRAPHY. CHAP. VI. all. On the other hand, in the passage under discussion in the text he enumerates the rivers in an order which would seem to place the Oarus to the west of the Tanais; while in the account of the expedition he clearly represents Darius as crossing the Tanai's, and advancing eastward to the Oarus. But he there does not mention the Lycus at all, though on this supposition Darius must have crossed it before coming to the Tanai's, and we should thus have a fourth river to account for between the Tanais and the Borysthenes, without anything really corresponding to it. This discrepancy in regard to the rivers would tend to show that Herodotus derived his account of the expedition of Darius from a different quarter from that which furnished him with the account previously given of Scythia and the adjoining countries, notwith- standing the close agreement already pointed out in the arrange- ment of the surrounding nations. NOTE G, p. 190. LIMITS OF SOYTHIA. The limits here assigned are materially less than those adopted by Mr. Rawlinson, who appears to me to extend the Scythia of Herodotus much too far to the north. He seems to have been in part led to this conclusion by assuming that it comprised the “ two great basins of the Don and Dnieper,” as well as the “ two minor basins ” of the Dniester and Boug (Herodotus, vol. iii. p. 210). But we know from Herodotus himself that both the two great rivers had their sources far beyond the Scythian territory; the one rising in the land of the Thyssagetae, separated by a broad desert from that of the Budini, who themselves lay to the north-east of Scythia proper; while the sources of the Borysthenes were unknown to him; but they were clearly situated beyond the farthest limits of Scythia, with the whole of which he considered himself as well acquainted. On the other hand, M. N eumann, whose general views on the subject appear to me to be sound and judicious, carries them, I think, to an extreme, and is disposed to restrict the Scythians of Herodotus within narrower limits than is reasonable or necessary. Whatever value we may attach to his statement of their extending 4000 stadia, or 20 days’ journey, inland, it is certain that he NOTE H. HERODOTUS 2 SCYTHIA. 2 I 5 regarded the Scythians as extending nearly, if not quite, to the sources of the Hypanis and the Tyras, as well as up the Borysthenes as far as Gerrhus. The hypothesis of M. Neumann, who would bring down Gerrhus below the cataracts of the Borysthenes, within a few days’ journey of the sea, appears to me utterly at variance with the conception of Herodotus, who distinctly tells us (iv. 71) that the Gerrhi were the remetest tribe that was subject to the royal Scythians. - NOTE H, p. 195. ETHN-OGRAPHICAL RELATIONS OF THE SCYTHIANS' OF’ HERODOTUS. There are few questions in ethnography that have been the subject of more discussion and controversy in modern times than the origin and ethnic-a1 affinities of the people described by Hero- dotus under the name of Scythians. The prevailing opinion is that they were I a Mongolian race, like the Kalmucks in modern days; and this view has been adopted by Niebuhr (Kleine Schriften, p. 362; Vortr'age iiber alte Geschichte, vol. i. p.179); by Schafarik (Slavische. Alterthiimer, vol. i. p.279), and by N eumann (Die Hellenen im Skythenlande, pp. 1498“, 199); as well as by our own historians, Thirlwall and Grote. On the other hand, several eminent philo- , logers have contended that they were a people of Aryan or Indo- European race. Dr. Donaldson (Varronianus, 2nd edit. pp. 40-45) attempts to prove that they were a Slavonian race, like their neigh- bours the Sauromatse; and Jacob Grimm (Gesch. der Deutschen Sprache, vol. i. p. 219) maintains that there is sufiicient evidence to assign them to the Indo-European family, without venturing to determine the particular branch to which they belong. The same hypothesis is adopted by Alexander Humboldt (Kosmos, vol. i. p. 491); Professor Rawlinson (Herodotus, vol. iii. pp. 192-205); and by Zeuss (Die Deutschen and die Nachbarstc'tmme, pp. 285-299). It may well be doubted whether we possess the means of arriving at a satisfactory solution of the question. On the one hand the elaborate account of the manners and customs, as well as the reli- gious rites, of the Scythians, transmitted to. us by Herodotus, presents so many points in common with those of existing Mon- golian races, that it appears at first sight to be decisive of the 216’ HISTORY or ANCIENT GEOGRAPHY. 01m. VI. subject : and the description of their physical characters by Hippo- crates, who wrote but little after Herodotus, and undoubtedly refers to the same people (De Acre, Aqua, et Locz's, c. 6, p. 558), tends to confirm this conclusion. But it must be admitted that there is so strong a general resemblance in the habits and manners of all the nomad races that have inhabited in successive ages the vast plains of Asia and Eastern Europe as to detract materially from the force of this argument. (See the judicious remarks on this subject of Humboldt, Kosmos, vol. i. p. 492.) On the other hand the linguistic grounds, which are principally appealed to by the advocates of the other theory, are scarcely sufficient to carry con- viction to a mind not predisposed in their favour. The few words of the Scythian language quoted by Herodotus—who was himself wholly unacquainted with it—would naturally be liable to much distortion, and the same thing would apply to their proper names, _ which we possess only in the form into which they were altered by the Greeks. Philological conclusions based upon such slender materials are very far from possessing the conclusive authority which they justly claim when they rest upon a sufiicient knowledge of the language. The Scythians appear to have continued to occupy the regions north of the Euxine for some centuries after the time of Herodotus, but they gradually gave way before the advancing tide of the Sarmatian or Slavonian races from the east. Whether they were driven westward, or were gradually absorbed by the successive waves of nomad population that swept over their country, we‘have no means of determining; but at a later period they disappear both from history and geography, and Pliny’s statement that in his time the (European) Scythians had become merged in the Sarmatians and Germans is probably well founded. (Scytharum nomen usquequaque transiit in Sarmatas atque Germanos, H.N. iv. 12, § 81.) It is remarkable that the Alani, who were found in the fourth century after the Christian era in possession of the same tract, and whose manners, as described by Ammianus Marcellinus (xxxi. 2), present much resemblance to those of the Scythians of Herodotus, disappear in like manner from history, and their ethnical relations are almost equally uncertain. . NOTE I. HERODOTUS : SCYTHIA. 2 I 7 NOTE I, p. 206. EXPEDITION OF DARIUS. This is the conclusion of Mr. Grote (Hist. of Greece, vol. iv. p. 354-360), as well as of Dr. Thirlwall (Hist. of Greece, 0. xiv. pp. 200-202), of Niebuhr (Vortriige iiber alte Geschichte, vol. i. pp. 189— 192), and, though in a modified form, of Heeren also (Asiatic Nations, vol. ii. p. 255). M. Duncker also, in his recent work (Gesch. des Alterthums, vol. ii. pp. 855-872), after subjecting the whole narrative to an elaborate discussion, rejects it as altogether unworthy of credit. Mr. Rawlinson has endeavoured, but, as it appears to me, with little success, to combat their arguments and diminish the improbabilities of the case. See his note on the subject in his Herodotus, vol. iii. p. 115. In addition to the inherent difficulties of the story, it may be re- marked that Ctesias, writing obviously from different materials, says simply that Darius marched for fifteen days into the Scythian territory, and then retreated to the Danube, which he recrossed, with the loss of a tenth part of his army (Ctesias, Persica, c. 17). Strabo’s account is that the Persian army never even reached the Dniester, but suffered heavy loss from want of water in the desert between that river and the Danube, and that Darius, discovering his error, returned to the Danube (Strab. vii. 3, p. 305). This is treated by N iebuhr (Kleine Schriften, p. 372) as a mere suggestion, suited to the probabilities of the case; but on whatever authority it may rest, it at least shows that the narrative of Herodotus did not obtain credence in ancient times, any more than with modern critics. It would be a mere waste of time to discuss the various theories of modern writers, who have attempted to explain away the narrative of Herodotus, while admitting its untenable character as it stands. To all such attempts we may reply in the words of Niebuhr: “As Herodotus tells us what is impossible, we know nothing at all historically respecting the expedition.” (218) CHAPTER VII. 0 GEOGRAPHY OF HERODOTUS. ASIA. SECTION 1.——General Views. § 1. WE have already briefly adverted to the extent and limits of the knowledge possessed by Herodotus of the continent of Asia. In general terms it \may be said that the portion of that vast continent with which he was acquainted did not amount to one- third of the whole.“ But within these limits his information, though of course very imperfect, was more definite, and on the whole more accurate, than might have been expected. The reason is obvious enough. The fact is. that l__1_i§_l_-\ together with the Caspian Sea, the northern limit of Asia.6 § 6. The Massagetac were apparently the only people whom Herodotus knew even by name as dwelling in this part of Asia ; he tells us that they were situated opposite to the Issedones,7 whom, as we have already seen,8 he placed in Europe, appa- rently to the south-east of the Ural mountains. Some accounts, _ he adds, represented them as a Scythian tribe, and he himself says that their manners and dress were similar to those of the Scythians. This was indeed probably the case with almost all the nomad races that have at successive periods occupied the vast steppes in this part of Asia; and this similarity of habits renders it almost impossible to judge of the true ethnical relations of any of these tribes mentioned by ancient writers. Herodotus himself appears to have regarded the Massagetm as a distinct people from the Scythians, whom, according to the tradition which he adopts as the most plausible, they had them- selves driven out of this part of Asia.9 But, as if on purpose 5 It is hardly necessary to remark that the Iaxartes does not in fact flow into the Caspian Sea at all, but into the Sea of Aral. But as the existence of this last inland sea was unknown to all ancient geographers before the time of Ptolemy, and even after the Oxus and Iaxartes were well known as sepa- rate streams, both were regarded as falling into the Caspian—it is hardly worth while to notice this additional source of confusion. 6 iv. 40. 7 1re'p'nv 1'05 ’Apoi§ew ‘Irma/1.01’), dim-{011 3% ’Io'0'r;36vwv dvfipé‘w, i. 201. The exact sense of this last expression is not easy to determine: but he probably means that they were to the south of them, facing them on the north. It is in this sense that he employs the word din-{air in the passage already examined, where he compares the Nile and the Danube. 8 See Chapter VI. p. 199. 9 Herodot. iv. 11. The ethnical re- lations of the Massagetae are as uncer— tain as those of most of the other nomad nations of Asia and Northern Europe. Dr. Donaldson considers them to be a Gothic race, arguing that the termination of the name is identical with that of the Getaa; and the same view is taken by Mr. Rawlinson, who even attempts to explain Massagetae as “ the greater Goths,” in contradis- tinction to the Thyssagetae, or “lesser Goths” (Herodotus, note to iv. 22). But all such conclusions from mere names are worth very little. Niebuhr, on the other hand, Heeren (Asiatic Nations, vol. ii. p. 279), and Schafarik (Sla'vz'sche Alterthiimer, vol. i. p. 279), concur in regarding them as a Mon- golian race. Humboldt was of opinion that they belonged to the Indo-Euro- pean family, without attempting to assign them to any particular branch of it. In the absence of all trace of their language the point must always remain uncertain. SE01‘. 1. HERODOTUS: ASIA. 225 to increase the confusion, he adds that the Scythians were driven across the Araxes, and thereupon occupied the land of the Cimmerians. Here it has been supposed that the river meant by the Araxes was the Volga,1 which would indeed render the geography in this particiilit? case (comparatively) intel- ligible. But the fact really seems to be that the ideas of Hero- dotus—and doubtless of his informants too—upon the whole subject were so utterly vague, that it is impossible to extract anything with clearness out of the confusion. It must be remembered,——as some excuse for our historian—or rather as showing how extremely uncertain were the ideas of his country- men on these geographical questions, that when the army of Alexander found itself actually on the banks of the Iaxartes. it was generally believed that that river was the same with the Tanais? The most remarkable circumstance mentioned by Hero- dotus concerning the Massagetae undoubtedly is the great abundance of gold which they possessed, and with which they lavishly ornamented both themselves and the trappings of their horses.3 Copper was also found in their country in great quantities, and was used for their arms and armour, while iron was unknown, as well as silver. They are described as ex- 1 It is probable, as has been repeat- error. Aristotle also confused the edly suggested, that the name Aras or Ras, out of which the Greeks made Araxes, was in fact applied to all the great streams of this part of Asia: it is probably the same root as we find in Rha, the name applied by Ptolemy to the Volga. According to Sir H. Raw- linson, in “primitive Scythian,” Aras signifies “ great.” 2 Arrian, Anabasis, iii. 30, § 7. Arrian himself seems disposed to accept as a fact that the Tana'is was the name of the river, but adds that it must be a difl‘erent Tana'i's from the one that flows into the Palus Maeotis. But there can be no doubt that the Macedonians in the first instance took it for the Tanais, and writers of the succeeding period, as Strabo has justly pointed out (xi. p. 509), intentionally confirmed the VOL. I. Araxes with the Iaxartes, and regarded the Tana'i's as a branch of it (Meteoro- logica, i. 13,§ 16). It may be added that the views of Herodotus concerning the course of the Araxes are not more widely erroneous than those entertained in the last cen- tury with regard to the great rivers of Central Africa, and the supposed rela- tions of the Niger with the Nile. ‘‘ i. 215. Both gold and copper are found in large quantities in the Ural Mountains, but it is difficult to suppose the Massagetse of Herodotus to have extended so far to the west. It is pro- bable, however, that some of the ranges of the Altai (still little known) may also contain these metals in equal abundance. Q 226 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VII. tending far to the east; but of their limits in this direction Herodotus had doubtless no knowledge; he seems however to have regarded the extreme east of Asia, beyond the Massagetae, as well as beyond the Indians, as a vast tract of uninhabited deserts. § 7. The account given of India and its inhabitants by the historian is in many respects interesting and instructive. He appears to have heard vague reports of its great extent and population, as well as of the vast wealth of its inhabitants The tells us indeed that the Indians were by far ‘the most numerous people with which he was acquainted ;4 and that the tribute lvhich they paid to Darius much exceeded in amount that of any other province of his vast empire.5 But it appears certain that the Persian kings never extended their dominion beyond the Punjaub and the valley of the Indus, nor is it likely that they possessed even the lower part of that river, though its course had been fully explored by order of Darius. It may indeed be questioned whether they ever possessed any real sovereignty to the east of the Indus, which certainly formed the limit of the Persian dominions in this direction in the time of Alexander. But it is not improbable that Darius may for a time have levied tribute from the neighbouring princes beyond this limit; the amount of the tribute derived from India being greater than can well be supposed to have been drawn-from the provinces west of the Indus~ only: and this is a point upon which Herodotus was more likely to have 0 tained correct information, than as to geographical details. Of the extensive and fertile regions of Hindostan proper he had un- questionahlyno knowledge: to him, as to all his Greek con- tempo/r/aries, India was the land of the Indus, which he regarded as flowing “towards the east, and the rising sun.”6 The Indians themselves dwelt the farthest towards the east ‘6? any 4 iii. 94. 11166» ~% wafiflés 're woAAqv" speaks of the Thracians as the greatest wAeTo'rziu elo'n rota/raw 'robv ime'is 15pm! and most numerous people next to the azlflpeé'lrwu, Kai ¢6pou drra'yiveov 'n'pbs 'miu- Indians. 'ras‘ 'robs dimer/s. Again, in v. 3, he 5 Note A, p. 255. 6 iv. 40. SE01‘. 1. HERODOTUS : ASIA. 227 people with which he was acquainted? beyond them in this direction there was nothing but sandy deserts, uninhabited and 1mknown.7 \Vhat lay beyond these deserts no one (he tells us) was able to say: the extreme east of Asia, like the north and north-west of Europe, was wholly unknown, and there was no certainty whether’ it was bounded by the sea or not.8 // § 8. From the mouth of the Indus to the Arabian Gulf the voyage of Scylax of Caryanda (which has been already referred to) was regarded by Herodotus as establishing beyond a doubt that there was continuous sea. Unfortunately the historian has given us no particulars concerning this interesting voyage, and he seems to have possessed no detailed information with regard to it, notwithstanding the fact of its being commanded by a countryman of his own.9 All that he tells us is, that Darius, wishing to know where the Indus had its outlet into the sea, sent out some ships, on board of which were Scylax of Caryanda, and other persons in whom the king had the greatest confidence. These set out from the city of Caspatyrus in the Pactyan land, and sailed “ down the river towards the east and the rising sun ” till they came to the sea; then turning to the west they sailed along by sea till they ultimately arrived, in the thirtieth month of their voyage, at the head of the Arabian Gulf, from whence the Egyptian king Necho had sent out his exploring expedition.1 Herodotus, moreover, adds that, after this, Darius, having reduced the Indians to subjec— tion, “ made use of this sea.” It would therefore appear that in this instance at least the voyage of discovery did not remain a wholly isolated occurrence. Yet we have seen how very imperfect was our author’s knowledge of the Erythreean Sea; and we shall hereafter see that when Alexander sent out the expedition of Nearchus to explore the course of the Indus, all memory of this voyage of Scylax would seem to have dis- appeared. Hergodotus had heard of the existence of crocodiles in the = iii. 9s, iv. in. 8 iv. 44. 9 Note B, p. 250. 1 iv. 44. Q 2 228 HISTORY or ANCIENT GEOGRAPHY. can». VII. Indhs, which was, in his opinion, the only river, except the Nile, in which they were found.2 He had also heard that the Indians were clothed in eglttmnsvhich he describes as a kind of wool, the spontaneous produce of certain trees, but of a finer and better quality than that of sheep.3 They had also a large kind of reeds (doubtless meaning bamboos), of which they made their bows. But these are the only natural productions of the country (except gold) to which he alludes. His account of the manners and customs of the inhabitants contains some particulars that were probably never correct, certainly not as applied to the Indians properly so-called; while he notices few, if any, of those peculiarities, which have distinguished the true Hindoo races in all ages, and which attracted the atten- tion of the Greeks as soon as they came in contact with them. But the Indian tribes which Herodotus directly describes, and whom he represents as living in a statggf revolting barbarism,‘i appear to have been situated on the borders, rather than within the bounds, of India proper, and may probably have been some remnants of the earlier races, who gradually disappeared before ‘the advancing civilization of the Hindoos.5 § 9. The only city which he mentions by name within the con- fines of India, according to his acceptation of the term, is Cas- patyrus, placed by him in the Pactvan land, from whence, as we have seen, Scylax and his companions set out on their exploring voyage, but there are no means of identifying either the city or the district with any certainty.6 All that we know is that the Pactyans were situated to the north of the other Indians, apparently bordering on the Bactrians, whom they resembled in their habits of life, and in their warlike disposition. It is in connection with this tribe of Indians 9 iv. 44. 3 iii. 106. is still said to prevail among the abo- ‘ iii. 98-101. riginal races who inhabit the upper ~" One of the most revolting of these valley of the Nerbudda, among the customs—that of killing and eating the recesses of the Vindhya mountains. aged and feeble members of their (Duncker, Gesch. des Alterthums, vol. ii. families—~ascribed by Herodotus to a p. 268.) tribe whom he names Padaeans (iii. 99), 6 Note C, p. 256. SECT. l. HERODOTUSZ ASIA. 229 that Herodotus tells us the strange story (copied by many succeeding writers) of the manner in which they procured the gold, with which they paid the Persian tribute. Gold indeed, he says, was produced in vast abundance in India, some of it washed down by the streams, and some dug out of the earth, but the greater part of it was procured in the following manner.7 Beyond the confines of the Pactyan land lay an extensive sandy desert in which there dwelt a kind of ants, not so large as dogs, but larger than foxes, which burrowed in the sandy soil, and threw updjargwefihea‘psgiike ordinary ant-hills, at the mouth of their burrows.” This sand abounded in gold; and the Indians, traversing the desert upon very fleet camels, filled their sacks with sand from these heaps, and then retreated in all haste, pursued by the ants with such speed that, if they did not succeed in obtaining a considerable start, they had no hope of escaping from them. / ' § 10. The locality from whence this gold was derived, is very obscurely indicated, though Herodotus appears to have con- ceived it as situated to the ‘north or north-east of the Indian tribes to which he refers.8 Heeren identifies it without hesita- tion with the desert of Cobi, on the north side of the moun- tains of Little Thibet, but there is really no foundation for this assumption. Herodotus never mentions the existence of ,- any mountains in this part of Asia at all,9 and hence it is" evident that he had no real notion of its physical geography’. The vague idea, that all to the east of the Indians was a sandy desert, probably arose in the first instance from the real fact of "h 7 Herodot. iii. 105, 106. trionalium Indoruin” (H. N. xi. 30, 8 Megasthenes (ap. Strab xv. p. 706), in repeating the same story, places the scene of it among the Dcrdze, “ a people among the mountains towards the east of India,” and describes the locality as a high table-land (dpo-n'é'diov) of about 3000 stadia in circumference. Pliny, who probably derived the story from Megasthcnes, writes the name Dardm, and places them “in regione Septem- § 111). They are in all probability the same people who are still known as Durds or Dards, and inhabit the lofty mountain tracts on the borders of Ka- firistan and Thibet. 9 In this respect, as we shall see, Ctesias was far in advance of him, as he was aware that many of the tribes in this part of India inhabited rugged, mountainous districts. 230 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VII. the occurrence of a broad desert tract to the east of the fertile lands of the Indus, and would be confirmed by vague reports that similar deserts were found also to the east of Bactria and the adjoining countries. Almost all the mountain chains in this part of Asia appear to be highly auriferous, and there is no reason to doubt that very large quantities of gold may have been derived by the Indian tribes from the plains at their foot, where it would naturally be found in the shape of gold-dust, or gold-sand. That the Indian tribute was actually paid in this form into the treasury of the Persian king was a fact concerning which Herodotus would have been likely to have obtained correct information. The story of the ants on the other hand is one of which it is very diflicult to divine the origin ; and of the various explana- tions that have been proposed, none can be said to be altogether satisfactory.1_ It appears certain however that the fable is a genuine Indian tradition,2 though embellished by the infor— mants of the historian; it is not only repeated by later Greek writers on India, such as Megasthenes and N earchus, but re- appears in the middle ages, and is found in the Arabian geographers.3 One of the most curious points in the matter is that the animal, whatever it was, to which this strange misnomer was applied, was certainly a real creature, as not only does Herodotus tell us that specimens of them were preserved in the menageries of the Persian king,4 but N earchus—one of the most careful and accurate writers of his time—asserts that he had himself seen their skins.5 And so persistent are such fables, when once propagated, that even in the sixteenth cen- tury one of these “ Indian ants ” was said to have been sent as a present by the Shah of Persia to Sultan Soliman6 at Con- stantinople. 1 Note D, p. 257. (ant-gold)” 2 See apassage from the Mahabha- 3 See the passages collected by rate, cited by Professor Wilson (Ariana Larcher and B'ahr in their notes upon p. 135), where mention is found of “that Herodotus. 4 iii. 102. gold which is dug up by Pippilikas 5 Ncarchus ap.Arrian. Indica,c. 15. (ants), and is therefore called Pippi’lika 8 It is described by Busbequius SECT. 2. HERODOTUS: ASIA. ' 231 § 11. \Vith this exception it is remarkable that Herodotus tells us none of the fables which were current in later times about India, and which are found in such abundance in Ctesias and other subsequent writers. It is however not improbable that this was due rather to the scantiness of his information than to its authenticity. It is curious that he never even mentions the elephants of India, which naturally figure in so prominent a manner in the writings of all subsequent authors; nor does he notice its ivory or precious stones; even where he cites India in confirmation of his view that the extreme regions of the earth possessed the most valuable productions, he dwells only upon its abundance of gold and cotton; adding however, in a general way, that all the animals there,——-both quadrupeds and birds—grew to a larger size than elsewhere.7 SECTION 2.~—Persz'an Empire: Satrapz'es. § 1. Of the countries that formed part of the Persian Empire in his time, Herodotus had undoubtedly ‘a general knowledge. He knew not only the provinces into which that empire was divided, but was acquainted at least with the names of the principal tribes and nations by which they were inhabited; and has preserved to us a catalogue of them, which is a docu- ment of the highest interest. Unfortunately he has not told us from what source he derived the account of the satrapies which he has inserted in his third book; but there can be no doubt that it was based upon authentic information, and was in all probability derived originally from some official record.8 But we must not hastily conclude that he possessed anything (Epist. iv. p. 343 ed. Elzevir), but only _ 8 The discovery and interpretation from hemsay, as “formica lndica, me- f 111 modern days of the hats of subject diocris canis magnitudine, mordax ad- . tribes and races that are found on the modum et saeva.” What the animal l-‘ersian monuments, have supplied us may really have been it.is impossible with materials of the greatestvalue for to Conjecturm comparison with those furnished by ' iii. 106. Herodotus. 232 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VII. like a complete geographical knowledge of the countries in question. We have already seen how extremely vague were his ideas concerning the Erythraean Sea and the lands that adjoined it. It must be remembered also that he regarded the Euxine as far exceeding in length its‘ real dimensions,9 and as he had an approximately correct idea of the true position of its western extremity (near Byzantium), he necessarily extended it at the other end a great deal too far to the east. One effect of this error was to bring it nearly to the same meridian with that of the Persian Gulf; and there can be little doubt that in the passage already cited, where he describesifour nations—the PersiansfMedesf Saspirian‘s/and Colchians’: as filling up the whole breadth of Asia from the Erythraean Sea to the Euxine, he considered them as lying nearly on the same line from north to south, so that the Colchians on the northern sea, would be (to use the phrase which he himself elsewhere employs) oppo- site to the Persians on the south; or as we should say, on the same meridian with them.1 The Caspian Sea would be in consequence placed several degrees too far to the eastward, and the whole of the adjoining regions thrown into confusion, but as we have no details concerning these countries—Herodotus himself having evidently no clear idea of their position—we cannot trace any farther the results of this error. § 2. Even in regard to the great peninsula of Anatolia orAsia Minor—as it was called by later geograp ers iii—with which“ Herodotus was comparatively familiar, as its coasts were sur- rounded by agirdle of Greek colonies, many of which the historian had himself visited,_——his geographical knowledge, in the strict sense of the term, was very inaccurate. While 9 See preceding Chapter. 1 It is in this sense (as we have seen) that he himself speaks of the mouths of the Nile as lying opposite to those of the Danube. 2 It may be as well to mention here (once for all) that the name of Asia Minor, so familiar to the student of ancient geography, was not in use either among Greek or Roman writers until a very late period. Orosius, who wrote in the fifth century after the Christian era, is the first extant writer who employs the term in its modern sense. SECT. 2. HERODOTUS: ASIA. 2 3 3 the excessive length assigned by him to the Euxine necessarily led him to prolong the northern shores of the peninsula to nearly the same extent, he considered its breadth to be much less than it really is. For he tells us that from the moun- tainous district of Cilicia, the direct route across to Sinope on the Euxine, was not more than five days’ journey “for an active man.”3 But the real breadth of the peninsula is in this part more than five degrees of latitude, thus giving about 60 (Sh/miles in a direct line for each day’s journey—a distance far exceeding what can be supposed to be intended by that expression :4 and even the narrowest neck or isthmus (between the gulf of Scanderoon and Samsun, the ancient Amisus) is Very nearly four degrees and a half. “ § 3. Farther to the east again helvas not only familiar with the two great streams of the Euphrates and the Tigris, and knew that theyboth flowed from the mountains of Armenia: a Tint he correctly describes several of the principal afHuents (if theglatter stream; as well as the Choaspes, on which the Persian capital of Susa was situated. He had indeed special means of information in regard to this part of Asia, for there can be no doubt that Herodotus had himself visited Babylon, and per- haps even Susa itself.5 It is at least highly probable that his 3 i. 72, ii. 34. linson’s note on the passage in his He- 4 Scylax (about acentury later) gives rodotus, vol. i. p. 210.) It seems more the same estimate of five days’journey probable that Herodotus relied upon (Periplus, §102). without adding the erroneous information. Eratosthenes, important qualification of Herodotus at a later period, assigned 3000 stadia “ for an active man” ef/{a’wrp Gwfipi). It as the breadth of the isthmus, which has been suggested that this expression (at 600 stadia to the degree) would be points to the case of acourier; and that pretty nearly correct (ap. Strab. ii. 1, these may have accomplished in ancient p. 68). Yet Pliny gives it as only 200 times as long journeys as the Persian Roman miles; reducing it again even couriers at the present day, who not below the estimate of Herodotus; while unfrequently travel on font at the rate Ptolemy on the other hand extends it of 50 miles a day. But there is nothing to 6% degrees of latitude, considerably to indicate that Herodotus referred to beyond the real breadth. any such exceptional rate of travelling; 5 I cannot at all concur with Mr. and the mountainous character of the Rawlinson and Colonel Mure in re- country to be traversed is such as to garding the passage in vi. 119 concern- add greatly to the difficulty of the ing the Eretrians who were established journey. (See Rennell’s Geography of at Ardericca as conclusive with regard Herodotus, p. 190, 4to edit. and Raw- to Herodotus having visited Susa. The 234 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VII. detailed account of the royal road from Sardis to Susa was derived in great part from personal observation. But of Upper Asia, as it was commonly called by later Greek writers 6—the great table-land of Iran, which occupies the “whole extent from the mountains that bound the valley of the Tigris to those on the west of the Indus, Herodotus cannot be said to have had any correct geographical knowledge what- ever, though he knew the names and some_other- particulars concerning the nations which occupied it. The great moun- tain chains, which in fact determine the physical geography of all this part of Asia, naturally occupy a prominent part in the description of later geographers, and—though their real direction and relations were very imperfectly understood, yet we at least find that Eratosthenes and Strabo were fully alive to their importance. But no trace of such an idea is found in Herodotus. With the exception of the chain of Mount Cau- casus, which (as we have seen) had at an early period attracted the attention of the Greek voyagers to the Euxine, and which is described by the historian as the loftiest of all mountains with which he was acquainted 7—he does not notice any of the mountain ranges of Asia“. Even in the description of the route from Sardis to Susa—:already adverted to—he briefly notices the existence of a narrow pass or gate (Pylae) at a statement that they continued there down to his own time (oi Ical ne’xpi e’néo eTXoz/ 'r'hz/ Xo'apnzl 'raz'r'rnv, qbvAciocroz/Tes 'rip/ apxaznv 'yaéooav) is surely one that he might easily learn at second- hand; and the very same phrase is used by him in speaking of the Bar- caeans, who were transferred to a village in Bactria (ii'zrep 51': real es é/aé iiv oircev- [new] £11 '13? 7f)‘ 'rfi Barc'rpifi, iv. 204), where we certainly cannot suppose Herodotus to have visited them in person, At the same time if it be admitted that he actually went to Babylon—and this is agreed on all hands—there is certainly no improbability in his having continued his journey to the court of the Great King. But the absence of any description or dctailtd notice of Susa itself appears to me conclusive against the idea of his having been there. Few subjects could be more in- teresting to the Greeks than the court and capital of the Persian king, and Herodotus would have had abundant opportunities of introducinga description of them into his history, had he been able to do so from personal observation. 6 The expression i7 z‘ivw ’A0117 is used by Herodotus himself in one passage (1. 95, ’Ao'o'vpiwv dpxov'rwu 'rfis til/w ’Ao'i7]s), but apparently in a wider sense than that in which it was employed by later geographers. The Assyrian Empire extended to the Halys, and probably this was the limit which the historian had in his mind. 7 i. ‘2.03. SEcT. 2. HERODOTUS : AQIA. 235 particular point of the Cilician mountains,8 as one of the diffi- culties which an army would have to encounter, but there is no other allusion to the great chain of Mount Taurus as one of the leading features of this part of Asia. \Vhile the systematic geographers of later times attached even an exaggerated im- portance to this mountain range, the very name of the Taurus is not found in Herodotus, though we can hardly doubt that he was familiar with it in its original and proper signification _as applied to the mountains of Lycia and Cilicia. As a general rule it may be observed that there are few subjects upon which the ideas of an unscientific observer are apt to be so vague and uncertain as that of the conformation and character of mountain chains. Isolated peaks, like ZEtna or Ararat, attract at once his attention and wonder: but it is only the scientific traveller or geographer who will group into their natural order and connection the more complicated ramifications of extensive mountain ranges. At the same time it is but fair to Herodotus to observe that he has nowhere professed to give us any regular geographical‘sketch of Asia, with the exception of the brief outline already discussed, and he may not unreasonably have regarded the ethnography of the Persian Empire (to use the modern phrase) as having a more immediate bearing upon his historical subject than its physical geography. § Al. The most important passage in Herodotus for the geo- grajjiy of the Persian Empire—that in which he describes its division into satrapies by Darius 9—-;is. indeed primarily of an ethnographical character, though combined with information of a purely statistical nature, which has everyappearance of being derived—though perhaps not directly—from an official record. But interesting as this enumeration of the provingfifi. and Jeep-- nations subject to the Great King is in itself, it derives great 8 This, as we shall see, was not the ' ander, but one farther east, on the fron- celebrated pass commonly known as tiers of Cappadocia and Mclitene. the Cilician Gates, which was traversed " iii. 89-97. by the younger Cyrus and by Alex- 236 HISTORY or ANCIENT GEOGRAPHY. CHAI’. v11. _ “satrapies or provinces. additional value from a comparison with the catalogue of the different tribes that sent their contingents to the vast army of Xerxes :1 a statement evidently proceeding from a different and independent source, and remarkable at once for its general agreement, and for its occasional discrepancies with the list of the satrapies. We have now also the opportunity of comparing the two with the less copious and detailed—but undoubtedly authentic—lists of the different nations subject'to the Persian king, found in the inscriptions of Darius himself at Behistun and Persepolis.2 In the following brief review of ‘the nations in question, the order adopted by Herodotus in the enumera- tion of the satrapies will be followed, as the most convenient for reference, notwithstanding its occasional deviations from geographical sequence. § 5. Beginning with Asignor (as was natural for the historian to do), we find that great country divided into four I The first of these comprised the Ionians, Eolians, Cariang: Lyciafis, and Pamphylians: that is to say the inhabitants of the coast regions from the borders of the Troad to those of Cilicia (including all the Greek colonies with which the coasts were studded), together with the inland tribes of the Milyans, whom Herodotus regarded as a distinct people from the Lycians, and the Magnesians, whom he here separates from the other Ionians, with whom they were usually associated.3 '/ The second satrapy contained the two well-known nations 1 vii. 61,88. Xerxes. Probably these wild moun- 2 These lists are given by Mr. Raw- linson in a note to his Herodotus (iii. 94, note 6). The original inscriptions will be found in Sir H. Rawlinson's Memoir on Persian Cuneiform Inscrip- tions (published by the Royal Asiatic Society in 1846), at pp. xxvii. 280 and 294. 3 The omission of the Pisidians is re markable, as they could hardly have been included among the Painpliylians. Nor do they figure in the army of taineers were not reduced to any real subjection, but continued to enjoy a state of virtual independence, as was the case with the lsaurians down to a much later period. In the time of the younger Cyrus the Pisidians appear as a race of lawless freebooters, whom that prince proposed utterly to extirpate as the only means of effectually protecting the neighbouring nations against their incursions. (Xenoph. Anal). I. i. § 11, 2, § 1.) SEGT. 2. HE RODOTUSI ASIA. 237 of the M sians and Lydians, together with three obscure tribes —the_falsbnians, Gabalians, and Hygennians. Of these the two former names appear again in the catalogue of the troops of Xerxes, where they are associated with the Milyans :4 and are apparently regarded by our author as the same people, originally of Mseonian race: the Hygennians are otherwise unknown. In the third satrapy were comprised the Hellespontians on the right bank of the Straits (including probably the whole district subsequently known as the Tread), the Phrygians, the Asiatic Thracians (that is, as he elsewhere explains,5 the Bithynians), the Paphlagonians, the Mariandynians, and the Syrians, by which term he here means the Cappadocians, who were generally known to the Greeks of his day by the name of Syrians or WVhite Syrians.6 This powerful people, whose native name of Gappadocians—under which they figure in the inscriptions of Darius—was well known to the historian, occu- pied in his time the whole of the interior of Asia Minor, from the river Halys on the west to the mountains of Armenia on the east, and from the Cilicians on the south to the shores of the Euxine. The Mariandynians (who here occur somewhat out of their place) though a comparatively insignificant tribe, were familiar to the Greeks from their occupying the shores of the Pontus in the neighbourhood of the flourishing Greek colony of Heraclea. The Chalybes or Chalybians, whose name is here omitted, but who are mentioned among the nations subdued by Croesus,7 occupied a district on the shores of the Euxine east of the Halys, which must also have been included within the limits of the third satrapy. 4 vii. 77. The names here appear the name of one of the nations described under the slightly altered forms, Caba- has undoubtedly fallen out of our MSS. lians and Lysinians. The discrepance The same conjecture was previously of our existing manuscripts in regard made by Wesseling (ad 100.), but all to such obscure names is not to be won- such restitutions must be extremely un- dered at. certain; and the very rude style of 5 vii. 75. 6 i. 72. equipment of the people in question 7 i. 28. Mr. Rawlinson suggests the appears unsuitable to a race so cele- insertion of their name in the catalogue brated as the Ohalybes for their skill of the army of Xerxes (vii. 76), where as workers in metal. 238 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VII. § 6. The whole of the fourth province was occupied, accord- ing to Herodotus, by a single people-—~the Cilicians, to whom he assigns a far greater extentbftérritorygthan the later geo- graphers, comprising not only the maritime province subse- quently known by their name, but the whole chain of Mount Taurus, and the districts on each side of it, as far as the con- fines of Armenia and the Matienians. They were evidently a numerous and powerful people; and are distinguished as furnishing, in addition to an annual tribute of five hundred talents, three hundred and sixty white horses every year for the service of the Persian monarch. The fifth provinge consistednof. Syria.(in.the modern accep- tation of the term) including Phoenicia and Palestine, with the island of Cyprus, and the sea-coastas far as the frontiers of Egypt, though the intervening strip of desert between these and the last outposts of Syria seems to have been always held by the Arabian tribes of the interior, who, though on friendly terms with the Persians, were never subject to the dominion of the Great King. _ The sixth district consisted oféigggpt, together with the Cyrenai'ca and the adjoining Libyan tribes. The Ethiopian tribes above Egypt, like the Arabians, were not subject to regular tribute, but presented an annual contribution in another form, consisting of gold-dust, ebony and ivory. § 7. Thus far the satrapies are enumerated by Herodotus in their natural geographical sequence; but he now—for what reason we know not—departs suddenly from that order, and enumerates as the seventh district one of the remotest provinces of the empire towards the east. The four nations of which it was composed are indeed all imperfectly known to us: and their geographical position has been a subject of doubt. But the Gandarians at least may be placed with reasonable certainty on the frontiers of India, in the modern Afghanistan,8 and the 8 We have seen that the Gandarians Caspapyrus in their territory (Hecataaus, were already known, by name at least, Fr. 178, 179). There can be no doubt to Hccataeus, who places the city of that they are the same people who SE01‘. 2. IIERODOTUSZ ASIA. 239 other three tribes probably occupied parts of the same moun- tain-tract intervening between Bactria, Aria, and India. The Gandarians and the Dadicas are again mentioned among the tribes that sent their contingents to the army of Xerxes ; they were armed and equipped in the same manner as the Bactrians, but had a separate commander of their own. The Sattagydians and Aparytae are not elsewhere mentioned, nor are the names found in any other ancient author: but the former have been identified with some plausibility with a people called in the Persian inscriptions Thatagush,——who there figure on the borders of Arachosia and India. It is to be observed that the name of the Arachosians does not appear in Herodotus; though it is‘found in the inscriptions of Darius, as well as in all the later Greek geographers: it is therefore not impro- bable that the district subsequently known as Arachosia formed a part of his seventh province, for which there would thus be ample room between Aria, Drangiana, and the frontiers of India. §8. The eighth province consisted only of Swarm‘ the land of the Cissians ; corresponding nearly to the modern Khu- zistan. It naturally derived its chief importance from the circumstance of Susa itself having become from the time of Darius the capital of the Persian monarchy, and for the same reason the name of the Cissians, which is hardly found in later geographers,9 was familiar to the Greeks as early as the time of Hilschylusf figure as the Gandharas in the legends and traditions of the I-Iindoos, and who appear to have been situated to the south of the Paropamisus or Hindoo Koosh (see Wilson’s Ariana, pp. 125, 131). The name is thought by many to be preserved in the modern Kan- dahar. The name of the Gandarii is not mentioned at the time of Alexander’s expedition; but that of Gandaritis is applied by Strabo to a district west of the Indus in the valley of the Cabul river (xv. p. 697); and he also tells us that some gave the name of Gandaris to the territory beyond the I-Iydraotes, which was subject to the second Porus (Ib. p. 699). There can be little doubt however that the Gandarians of Hero- dotus and Hecataeus were a people to the west of the Indus. 9 Strabo only mentions it in passing as a name by which the Susians were also known (xv. p. 728). The modem name of Khuzistan is in some measure intermediate between the two. ' ZEschyl. Pe'rsze, v. 27. 240 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VII. The ninth satrapy comprised all Babylonia and the rest of Assyria—a phrase of which it is not very easy to define the meaning. Herodotus generally makes no distinction between the Assyrians and Babylonians, and in his first book distinctly describes Babylon as the capital of Assyria after the fall of Nineveh. But we not only know that the two monarchies were in reality quite distinct; but the names of Assyria and Babylonia were retained in later times as characterizing separate districts; the former appellation, as thus employed, being confined to the great province east of the Tigris, of which Nineveh was the capital. Yet Herodotus, as we shall presently see, in describing the royal road from Sardis to Susa, appears to assign the whole tract on the left bank of the Tigris either to Armenia or Matiene, thus leaving no room on that side for Assyria at all. On the other hand he never uses the term Mesopotamia, which indeed could obviously never have been an ethnic appellation, and was probably only a term invented, or adopted, at a later period by Greek geographers.2 It is therefore probable that the region which Herodotus meant to comprise in the ninth satrapy included all Mesopotamia as well as Babylonia, but did not extend to the east of the Tigris. § 9. The tenth government contained the important and well- known province of Medim'of which .Ecbatana was the capital,3 together with two other tribes -the Paricanians and the Ortho- corybantes—whose names are wholly unknown to us, though that of the Paricanians may perhaps be an altered form for Paraettecanians or Paraetacenians, who are elsewhere mentioned by Herodotus as a Median tribe. ' The eleventh district comprised four nations or tribes—none 2 The name of Mesopotamia probably Semitic nations. Arrian expressly came into use among the Greeks after the conquests of Alexander. It is 1n fact merely a translation of the name Aram-Nahara'im, “ Syria of the two rivers,” by which the district was known to the Jews, and doubtless to the other , tells us (Anal). vii. 7, § 3 ) that it was a native appellation. Among extant Greek authors it is first found in Poly- bius (v. 44, 48). 3 See Note E, p. 258. SEGT. 2. HEROKQOTUSZ ASIA. 24l of them apparently of much importance—the Caspians, the Pausicee, the Pantimathi, and the Daritae. The first of these was doubtless the same people that are found in the later geographers settled on the south-western shores of the sea to which they gave their name; they occupied the tract from the mouth of the river Cyrus or Kur to the SW. angle of the Caspian Sea, known in modern times as Ghilan. The other three tribes, whose names are utterly unknown, may probably be placed on the southern shores of the same basin, in the modern province of Mazanderan, or the lofty mountains that bound it to the south. The greater part of this tract was included by later geographers in Hyrcania: but though the Hyrcanians figure in the array of the army of Xerxes, their name is not mentioned in the list of the satrapies. § 10. The twelfth satrapy included the Bactrians, “as far as the Pilgli ”—-a tribe who have been supposed, but upon very slight grounds, to have been located upon the Iaxartes: this view, however, is difficult to reconcile with the fact that the Sogdians, of whose locality there is no doubt, and who inter- vened between the Bactrians and the Iaxartes, belonged to a different government. The twelfth satrapy was in fact pro- bably confined to the Bactrians proper, who were doubtless in the time of Herodotus still a powerful and flourishing people. The thirteenth province contained the Armenians, with the Pactyans, and other adjacent tribes extending to the shores of the Euxine. It must be observed that the Armenia of Herodotus was much less extensive than the country known. by that name in later times, as he obviously assigned to the Saspeires or Saspirians, a considerable portion of the region known in later times as Eastern Armenia. The name of the Pactyans is‘not found elsewhere.4 4 It is hardly necessary to observe that they were certainly distinct from the Pactyans, who adj oined the Indus, and in whose territory the city of Cas- patyrus was situated. (See above, p. 227). But this case of the unquestion- able occurrence of two distinct tribes of VOL. I. the same name in distant provinces, may act as a warning against the too hasty identification of different tribes, in other cases, merely on account of the similarity of name. The Pactyans who appear in the army of Xerxes (Herodot. vii. 67) were R 242 HISTORY OF ANCIENTQGEOGRAPHY. CHAP. VII. ‘ § 11. The fourteenth satrapy was an extensive one, and com- prised some tribes known to be numerous and powerful, but whose limits are not easilypdefined. It included the Sagar- tians, Sarangeans, Thamhlflaeans, Utians, and Mycians: “ to- gether with the islands in the Erythraean Sea.” This last addition leads us to suppose that the southern limit of the province in question extended to the shores of the Indian Ocean :5 and it is a plausible conjecture (though it must be admitted to be nothing more) that the Mycians occupied the district still called Mekran, on the southern border of the great table-land of Persia.6 The Sarangeans may be confidently iden- tified with the Zarangians or Drangians of later writers, who occupied thedistrict now known as Seistan. On the other hand the Utians, —who appear in the army of Xerxes associated with the Mycians and Paricanians (evidently the people of that name assigned to the seventeenth satrapy), are otherwise unknown,7 and can only be conjecturally placed in the southern portion of the empire, where they probably adjoined the Mycians. The Sagartians,—who were clearly a nomad race, and appear in the army of Xerxes as furnishing 8000 horsemen armed only with lassoes,—probably inhabited, or rather roamed over, the central districts adjoining the Great Desert of Iran. The Thamanaeans are wholly unknown: but as Herodotus elsewhere8 associates them with the Sarangians, Parthians, Hyrcanians, and Chorasmians, they evidently inhabited some part of the same great plateau. _, ' § 12. The fifteenth sati‘dapy was composed of two nations only clearly the eastern people of the name, from the borders of India. 5 The islands in question can hardly be any other than those in the Persian Gulf, which (as already mentioned) Herodotus does not distinguish from the Erythraean Sea or Indian Ocean. also from Hecataeus (ap. Steph. Byz. s. v. Mvlcoi), and the few words e’rc Mmccibu sis ’Apoi§r)v 1ro'ra,as ,uelue'rpn'rat 1'7 6'30; '1‘) ,Bao'tAn'i'n 'rofm wapao'dyyyo't ; had the distances been really “ measured and marked ” from station to station, it could scarcely have occurred to him to express any doubt upon the subject. ' .(262) CHAPTER VIII. GEOGRAPHY OF HERQDOTUS: AFRICA. SECTION 1.——General Knowledge of the Continent. § 1. THE amount of knowledge possessed by Herodotus con- cerning the continent of Africa appears at first sight to present a striking contrast with his ignorance‘of the remoter regions of Europe. But the contrast is more apparent than real ; and will be found to arise rather from our own familiarity with the one, and our comparatively imperfect knowledge of the other, than from any real superiority in the information possessed by Herodotus. His-knowledge of the southern shores of the Medi- terranean was in fact much of the same character with that of his acquaintance with the opposite coasts. There seems no doubt that he had himself visited (Qyrenefi which was at that time an important centre of Greek life~ and civilization, and he had there obtained a pretty correct general account of the coasts of Africa and the tribes that inhabited them asfar as the Lesser Syrtis and the confines of the Garthaginian territory. Beyond this to the west he either possessed no information, or did not think fit to communicate it to his readers. He was indeed, as has been already stated, familiar in a general way with the greatness and power of Carthage and the extent of 1 The only passage from which this can be directly inferred is in Book II. c. 181, where he mentions a statue con- secrated by Ladice, the Greek wife of Amasis, which was still existing in his time, outside the gates of Cyrene, in a manner that clearly implies that he had seen it. But the whole account of Cyrene and the adjoining regions in the fourth book bears throughout the stamp of having been derived from personal inquiries on the spot. Indeed it is probably to this visit to Cyrene that we are indebted for the whole of this valu- able and curious episode (iv. o. 105— 125), which is very remotely connected with the general subject of his work; SECT. 1. IIERODOTUS : AFRICA. 26 3 her dominions; and he more than once cites the information which he had received from Carthaginians—probably mer- chants whom he had met at Cyrene or elsewhere.2 But it may be taken as certain that he had not visited Carthage itself; and much of the information which he had thus picked up was of a vague and desultory character. With regard to the interior of the continent, what knowledge he possessed was derived from two sources— partly from the information he was able to obtainwin Egyptf'and partly from g ,, the accounts that he gathered from his countrymen at Cyrene. But the natural peculiarities of Africa must in all ages have presented an almost insuperable barrier to intercourse with the interior; and these difiiculties must have been vastly increased in ancient times by the want of camels, which do not appear to‘ have been introduced into Africa until a much later period.3 Hence the great desert, which extends almost without inter- ruption from the confines of Egypt and Nubia to the Atlantic Ocean, naturally formed the limit of the knowledge possessed by the Greeks towards the south; and with one remarkable exception—to which we shall hereafter have occasion to revert—- Herodotus appears to have had no conception that beyond this desert tract, there existed habitable and even populous regions. The course of the Nile was then, as it remains at the present day, the only natural highway from the Mediterranean to the remoter regions of Africa. § 2. But with the upper part of that river Herodotus was very ) imperfectly acquainted. As far as the confines of Egypt it was of course well known ; and the historian had himself ascended it as far as Elephantine, just ‘below the FirstCata- ,1 racts.4 But he had also obtained what he believed to be precise and authentic information concerning its course for a consider- able distance higher up, which he has imparted to us in a very curious passage. “Above Elephantine (he tells us) the ground 2 Herodot. iv. 43, 195, 196, vii. 167. 'rd'ir'r'ns ekda'w, 7b 5’ rim-b 'rod’rov ducofi fidn 3 See Note 'A, p. 299. io'rope'wv, 29. See Note B, p. 299. ‘ ne'xpr p211 ’E7\e¢av'rtvns miMos at’:- 264 HISTORY OF ANCIENT GEOGRAPHY. CHAP, VIII. has a considerable slope, so that it is necessary for the boats which ascend the river to be fastened with tow-ropes on each side, and thus dragged up the stream. If the ropes should { break, the boat is carried away at once by the violence of the current. This kind of navigation continues for four days’ voyage, during which the river winds very much, like the Mmander. The whole distance which it is requisite to traverse in this manner is twelve schoeni. After that you come to a level plain, in which the Nile encloses an island named TachompSO- ‘ Of this one half is occupied by the Ethiopians, who inhabit all the country above it: the other half by the Egyptians. Ad- joining this is a large lake, the shores of which are inhabited by nomad Ethiopian tribes: the Nile flows i/nto it, and you re- enter the stream after traversing the lake. But here you are compelled to land, to travel along the banks of the river for forty days: the navigation being rendered impossible by the number of sharp rocks and ledges, which occur in this part of the stream. After traversing this tractmin forty days, you em-rf bark in another vessel and sail up the river for twelve days," at the end of which you come to a great city called Merci-5,) which is considered as the metropolis of all the Ethiopians.n5 §3. The situation of Meroé is now well k110W11;——tl1011gh it is only in recent times that it has been explored and ascer- tained 6—-and it may be admitted that the account of Hero- dotus thus far presents a tolerable approximation to the truth, considering that it of course rests only upon hearsay informa— tion. The ascent of the rapids above Elephantine, commonly known as the First Cataract, is well described, though its duration is greatly overrated—the actual passage only re- quiringabout five hours ;7 the island of Tachompso may pro- bably be identified with that of Derarja short distance above 5 Herodot. ii. 29. detail by Caillaud in 1820 (Voyage a 6 As late as the time of D’Anville Meme’, au Fleu've Blane, (20., Paris, the position of Meroe could only be de- 1826), and have been since repeatedly termined approximately from Ptolemy described by Hoskins, Riippcll, and and the Arabian geographers. Its Lepsius. _ ruins were first visited and described in 7 See Note C, p. 301. Sne'r. 1. HERODOTUS: AFRICA. 265 Dakkeh ‘(the Pselcis of ancient geographers); and though there is no such lahe as that said to occur immediately above, > the error may have readily arisen from the term being used merely to designate an open rgaeh _or__u_nencumbered expanse of _ the river. The rocks and obstructions to the navigation higher up are evidently those which really occur between the second and __f9_1,ll1§h cataract, on account of which almost .. all, trayellerswhohave visited Meroé have taken the route across the desert which here cuts off a great bend of the river.8 But in the time of Herodotus it appears to have been customary to follow the banks of the river itself, and the esti- mate of 40 days’ journey may on this supposition have been not far from the truth§ but it is not clear at what point the navigation was resumed. The existence of a civilized city and people at Meroé; worshipping the same deities as the Egyptians, is first men-; tioned by Herodotus, but was undoubtedly a fact well known to the Egyptians. “7e learn from the monuments still existing on the site that the earlier kings of Egypt had not only carried their arms into these remote regions, but had left there per- manent records of their dominion; and we can hardly therefore doubt that the Egyptians in the time of Herodotus continued to maintain commercial intercourse with the inhabitants of the upper valley of the Nile as far as Meroé. But there is no reason to presume that such trade was carried higher up the river. Hence we find the information of Herodotus above this point of a totally different character from that below it. He had apparently no knowledge of the great afiiuents or tributaries of the Nile, which bear so important a part in the geography of its upper regions, nor does he anywhere allude to “ the island ” of Meroé, the designation by which that tract was generally known to later geographers. The only people of whom he had heard as situated beyond Meroé were a race 8 Mr. Hoskins, however, on his return journey from Mero'e', followed the banks of the river throughout (Travels in Ethiopia, 4to. Lond. 1835, chap. xii.—xviii.). 266 HISTORY OF ANCIENT GEOGRAPHY. GHAP. VIII. whom he terms the Automoli or Deserters, but whose native name (he adds) was Asimach. These were originally, according to the Egyptian tradition, a body of Egyptian warriors, who had migrated into Ethiopia, in the reign of Psammetichus (about B.0. 650), to the number of two hundred and forty thousand men, and had been settled by the king of Ethiopia in the extreme south of his dominions.9 Admitting the number to be a gross exaggeration, there is nothing impro- bable in the tradition itself, and the comparatively late period to which it was referred is undoubtedly in favour of its authen- ticity. But when the informants of Herodotus placed this Egyptian colony at a distance beyond Meroé, equal to that from Meroé to Elephantine, it is obvious that there must be either great exaggeration, or some misconception, to which we have no clue.1 The total distance from Elephantine to Mero'e' is estimated by Herodotus at 60 days’, or two months’ journey ;2 and as he reckoned the land of the Automoli the same distance beyond Meroé, he was thus led to assign to this people a position four months’ journey from the Egyptian frontier. “ Thus far then (he tells us) the course of the Nile is known beyond the confines of Egypt.” “ It flows (he adds) from the west and the I setting sun.”3 " -§ 4:. These last words are so strikingly at variance with the real state of the case, as rendered familiar to us by modern maps, that most of the editors and commentators of Herodotus have shrunk from accepting them in their obvious sense, as applying to the whole course of the Nile beyond the limits of Egypt, and supposed Herodotus to speak only of the course of the river at the farthest point to which it was known,4 assuming, as a matter of course, that he knew its direction from Meroé, and even from the land of the Automoli to be from south to north. But there is nothing in the text of 9 Herodot. ii. 80. 1 Note D, p. 302. 2 Nole E, p. 302. 3 Herodot. ii. 31. 4 Note F, p. 303. Seer. 1. HERODOTUS : AFRICA. 267 Herodotus to show this; and the passage already cited in a former chapter in which he compares the course of the Nile with that of the Danube seems clearly to prove that he con- sidered the general direction of the former river, as well as the latter, to be from west to east. It is not improbable that this. erroneous idea arose in part from the great bend or elbow‘ which "the Nile does actually make between Meroé and the Egyptian frontier, and which may easily have given, rise to a misconception of its‘ true direction.5 Had Herodotus placed the land of the Automoli so far to the south as would have been required on the contrary supposition, he would have had to assign to the continent of Africa a breadth’ far exceeding that which it occupied in his system, according to which, as we have already seen, it was not to be compared in breadth (i. e. from north to south) with the opposite continent of Europe. § 5. The same view will be found to be entirely in accordance with the account given by Herodotus of the expedition of Cambyses against the Macrobian Ethiopians, which is utterly unintelligible on any other supposition. These Macrobians, as he tells us, dwelt “ on the sea to the south of Africa,”6 and he elsewhere distinctly speaks of them as living “ at the extremity of the world.”7 But when Cambyses sets out to invade them, instead of ascending the valley of the Nile to Meroe, and thus penetrating as far south as he could, with the advantage of the river at hand, he plunges at once into the desert, apparently almost directly after leaving Thebes, and continues his march “ through the sands ” till compelled to return by want of pro- visions. It seems clear from this account that Herodotus did not consider these Ethiopians as dwelling to the south of those of Meroé beyond the Automoli, but in a difi‘ererit direction, so 5 For more than four degrees of lati- l Herodotus, and the historian may well tude above Syene the Nile flows nearly have misconceived it. from S.W. to N .E. The great bend or 6 oilcmuévous Arfizins e’vrl 'rfi uo'rfn 6a- loop which it forms between this part Nio'o'n (iii. 17). ' of its course and Mcroii was, doubtless, 7 e’s Ta é’o-Xa'ra T775 755;‘ {MAM (mm- imperfectly known to the informants of retention (iii. 25). i 268 HISTORY OF A‘NCIENT GEOGRAPHY. CHAP. VIII. that the king had to leave the valley of the Nile, and strike across the desert, in order to reach the southern sea, while the Nile trended away towards the south-west and west. Later writers, being unable to find any such people as those described by Herodotus, confounded these Macrobian Ethiopians with those of Meroé, but this is certainly inconsistent with the views of the historian. Whatever may have been the origin of the fable, it is certain that, according to his conception, the Macrobians dwelt on the shores of the southern ocean, and had no connection either with Meroé or the Nile. §6. But though the Nile was thus known, according to Herodotus, for a distance of four months’ voyage or journey above the frontiers of Egypt, its sources, as he tells us, were altogether unknown; no one that he had “mg; with, whether Gr9_e:_k,_‘Egyptian. or-Libyan, .bein-gabl‘emtdq ‘15511 him whence the great river came.8 It is hardly necessary to add that all suc- ceeding geographers, down to our own times, have had to repeat the same tale.9 But Herodotus, like many of his suc- cessors, had formed a theory of his own upon the subject, and it was one which certainmlhdesdldredithtoghissagacity, while it incidentally affords a curious gleam of information concern- ing the interior of Africa. Starting from the belief, already referred to, that the Nile, in the upper part ‘of its course, flowed from west to east, he connected this with a tale which he had heard from some Greeks of Cyrene, who had themselves learnt it from Etearchus, the king of the Ammonians, during a visit to the celebrated oracle of Ammon.1 Etearchus himself derived it from the casual visit of some N asamonians, a Libyan tribe who dwelt around the bight of the Greater SyFtis. It is important, in estimating the authenticity of the narrative, to bear in mind the successive steps by which it passed to the knowledge of Herodotus. a Note G, p_ 304. lakes south of the equator, though so 9 Ptolemy may indeed be justly re- long discredited in modern times, has gardcd as an exception: for his view been new proved to be substantially of the Nile as taking its rise in two correct. 1 Herodot. ii. 32. Seer. 1. HERODOTUS : AFRICA. 269 § 7. According to the statement of these N asamonians, five young men of their nation, enterprising youths of the highest rank, had set out with the express purpose of exploring the deserts to the south of Libya, to see what they could discover. After passing through the inhabited region and the tract to the south of it, frequented by wild beasts, they entered upon the actual desert and journeyed through it for many days in a westerly direction, until at length they came to a grove of trees, with fruit on them, of which they began to help them- selves, when there came up a number of men of small stature, who seized them and led them away prisoners. They were thus conducted through very extensive marshes, after passing which they came to a city the inhabitants of which were of a black colour, and of the same small stature with their captors. .Their language was wholly unintelligible to the N asamonians, and they were all of them skilled in magic. A large river flowed by their city, having its course from west to east; and in it were seen crocodiles. The young men returned in safety to their homes; and from this report Herodotus concludes that the river they had seen, flowing from west to east, with croco- diles in it, could be no other than the N ile.2 We now know with certainty that this hypothesis is wholly untenable; but it must be remembered that it continued to be a favourite theory with modern geographers down to a recent period; and long after the immense interval was known which separated the Joliba “of” Western Africa from the upper waters of the Nile, the two were still supposed to be connected, if not continuous. There seems no reason whatever for rejecting the narrative as fabulous, and it is perfectly credible that young men accustomed to the climate, and well provided with food and water (as we are told they were), may have made a journey across the desert which would be impracticable for more civilized travellers. But the extreme vagueness of the state- '-’ Herodot. ii. 32, 33. 270 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII‘. ment renders it almost impossible to judge of the point really attained. We are not told either the point from which they set out, or entered on the deserts, or the number of days occupied in the journey, either before or after their capture. The only definite statement is that after entering on the desert they travelled towards the west; and it is precisely this statement which the nature of the case compels us to reject, or at least to admit only with considerable modification. The N aiakmonians (we are told) traversed in the first instance the two zones or regions which were well known to the inhabitants of Northern Africa, but to which Herodotus assigns more definite limits than they really possess ;3 and their express object being to penetrate farther than any one had done before, they would naturally take advantage of those inhabited districts which were known to them, as far as they extended. Now it appears certain from another passage in Herodotus (which will be more fully considered hereafter) that the fertile district of Egzgan, the land of the Garamantes,—was known ~ to, and frequented by, the Libyans, in his time; and of this, therefore, as being within reach of the N asamonians, and extending far to the south into the desert, the travellers would naturally avail themselves. But if they here plunged into the deserts, and directed their course clue west, they had nothing but the vast deserts of the Sahara before them, and could never have come upon any considerable river. If they struck to the south-west, on the contrary, they would exactly hit upon the celebrated N iger—the Joliba or Quorra of modern travellers; while, if we suppose them to have travelled due south, they would have reached at a somewhat shorter distance the leourasmgfiiislléblfi-lixenflowing from. .westlneamgnd enterigg‘timeJ-alifi_hl_l'flj_iA I \l'i; is difficult to decide between these two theories: both alike require that we should depart from the direction assigned by Herodotus, but we have abundant instances of the extreme 3 See p. 274. SECT. 1. HERODOTUSZ AFRICA. 271 vagueness of ancient writers with regard to the points of the compass; and it must be added that in this case the statement in some measure disproves itself : for, had the explorers really held a course from east to west, they would, according to the view of Herodotus himself, have been moving parallel to the Nile, and could therefore never have hit upon that river. The really important parts of the story undoubtedly are, that the travellers afte_r__traye_r'_s_ing_a great extent of desert, came to A ,_____. . inhflrblffiflnlélld .Witlllil‘ees, marshes, a great river flowing from- thewwest , and...c.ontaining crocodiles, and a city occupied by. a race of black people- The combination of these circumstances seems to render it certain that they had really crossed the great desert tract and reached the N egroland of Central Africa. It is far more diflicult to determine the particular region which they visited. But on the whole the probabilities in favour of their having actually reached the Niger may be held to predominate, notwithstanding its greater remoteness from the point whence they probably started.‘ But whatever may have been the region actually visited by these N asamonian adventurers, their expedition appears to have remained a wholly isolated fact. It is clear from the terms in which it is related by Herodotus himself, that their example was not followed. N 0 commercial relations were established with the tribes beyond the great desert, and it is evident that with the exception of this single vague notice, Herodotus had no idea of the existence of the vast regions known in later times as __Soudan or N egroland.5 8. The only other passage in Herodotus which appears to point to any knowledge of the more distant regions of Africa to- wards the south, is that already alluded to, in which he describes the, Mgcrobian Ethiopians, a people who were evidently regarde by him as the remotest inhabitants of Africa, of whom he had any knowledge. His information concerning them was how- ever very vague and strongly mixed with fable. They dwelt, 4 Note H, p, 305. '" Note I, p. 307. 272 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. ' according to the account which he had received of them—an account evidently derived from Egyptian authorities,—~i“~upon wtheflslggrespf the sea to thensouth QfAfrica ;” they were the tallest and handshrhest race of men in the world, and lived on the average to the age. of 120 years: gold was so abundant in their country that it was used even for the chains and fetters of prisoners, while bronze was on the contrary extremely scarce : they enclosed the dead in pillars of transparent crystal instead of coflins: their food consisted solely of meat and milk ; bread and wine being alike unknown to them. In addition to the simplicity of their diet, a marvellous fountain in which they bathed, was supposed to be one of the chief sources of the extraordinary health and longevity that they enjoyed.6 It was against this people that Cambyses undertook an expedition, in which both he and his army were very near perishing utterly in the desert. But the circumstances of this expedition are so briefly related by the historian as to be almost wholly unintelligible, and can hardly be said to throw any light upon the geographical position of the people against whom it was directed. Previously to undertaking the enter- prise Cambyses sent an embassy to the king of the Macrobians, and for this purpose sent for interpreters from the Ichthyophagi, a people who are not elsewhere mentioned by Herodotus, but who are noticed by later writers, as settled on the west coast of the Red Sea, extending as far as the straits at its mouth. These Ichthyophagi therefore, it may be presumed, were a people of the same race with the Macrobians, and probably situated near them: a supposition which would concur with the expression of Herodotus that the latter dwelt on the southern sea at the farthest extremity of the earth.’I But the route by which Cambyses attempted to arrive at this remote region is very obscurely indicated, or rather can hardly be said to be indicated at all. On arriving at Thebes, we are told by our author, he sent off a detachment of his army 6 iii. 17—24. 7 See above, p. 267. SECT. 1. HERODOTUSI AFRICA. 273 to attack the Ammonians and destroy the temple of Ammon; with the rest of his forces he proceeded against the Ethi- opians, but having neglected to make due provision of food for his troops, before they had completed the fifth part of the journey, their provisions utterly failed, and they were reduced to eating their beasts of burden. Cambyses neverthe- less still persevered and it was not till they had entered upon the sandy desert, where for want of all other resources his soldiers began to eat one another, that he was persuaded to abandon the expedition and return to Thebes, having lost a large part of his army.8 § 9. Strabo and other writers of subsequent times, when the geography of Africa was better known, regard the expedition of Cambyses as directed against the Ethiopians of Meroe: 9 and on this supposition there would be little doubt that the desert in which his army suffered so severely was that between Korosko and Abu Hamed (on the direct route from Syene to Meroe) the horrors of which have been described in forcible language by all travellers who have crossed it.1 But it is clear that this view of the subject was not that of Herodotus. These Macrobian Ethiopians (whom he mentions only in connection with this expedition of Cambyses) are distinctly described by him as dwelling on the southern sea, in the remotest regions of the earth; while the Ethiopians of Meroe dwelt on the banks of the Nile, and the Automoli or Deserters lived far beyond them, but still on the same river. N o mention occurs of the Macrobians in the description of the Nile and the people that occupied its banks: nor any hint that they adjoined the Ethiopians of this quarter.2 The fact appears to be (as already 8 iii. 25. 9 Strabo, xvii 1, p. 790. 1 See Hoskins’ Travels in Ethiopia, pp. 1932; Lepsius, Briefe aus Aegypten, pp. 125-136. '2 Mr. Cooley’s attempt to identify the Macrobians with the Automoli (Claudius Ptolemy and the Nile, pp. 20, 21) appears to me utterly untenable. VOL. I. The fabulous and mythical notions con- cerning the former people could never have attached themselves to a race who had established themselves in Ethiopia within historical, and even compara- tively recent, memory; and the very existence of the tradition concerning the Automoli proves that they must have borne a marked resemblance to T 274 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. indicated) that Herodotus, misled by hisv erroneous notion that the Nile flowed from the west, conceived the Macrobians to be situated in quite another direction, that Cambyses had alto- gether left the course of the river, and struck across the deserts towards the south, with the view of reaching their country, when the want of supplies compelled him to return. ‘What the real circumstances of the expedition may have been we are wholly unable to udge : it is not improbable that some fabulous tales of an El Dorado on the southern sea may have tempted Cambyses to undertake a march of the extent of which he knew nothing. That the expedition, whatever its purpose or destination, proved an entire failure, and occasioned the loss of great part of his army, is all that we know about it.3 § 10. It may therefore be safely asserted that, westward of Egypt and the valley of the Nile, the knowledge of Herodotus was confined to the regions of Africa north of the great desert. But with this portion of the continent he shows a degree of acquaintance, and a general conception of its physical features exceeding what might reasonably have been expected, con- sidering the imperfect nature of his sources of information. In more than one passage he dwells upon the marked division of the regions in question into three zones or tracts, clearly characterized by nature. The first of these, extending along the sea coast of the Mediterranean, all the way from the fron- tiers of Egypt to Cape golqgis beyond the Pillars of Hercules, with the exception of the portions occupied by Greek and the Egyptians: aresemblance of which some traces are said to be still found in the inhabitants of Sennaar. But the accurate information: as he confuses the expedition of Cambyses himself with his main army, with that of the whole description of the Macrobians represents them as a people unlike any other. Their fabulous characteristics are the essential part of the tradition concerning them; as their position on the southern ocean distinctly separates them from the Ethiopian races in the valley of the Nile. 3 It is clear that the view taken by Strabo was not founded upon any more detachment sent against the Ammo- nians; and speaks of “the army of Cambyses” having been buried in the sands on the way to Meroe' (Strab. xvii. p. 820). It is not impossible that the expedition was in reality directed against Meroé; but it is certain that this was not the view taken by Hero- dotus himself, or by the persons from whom he derived his information. SE01‘. 1. HERODOTUS: AFRICA. 275 Phoenician settlers, was inhabited by' Libyans and races of Libyan origin; and though the greater part of these werg merely nomad tribes, the whole tract was regarded by Hero dotus as “ the inhabited region ” of Libya. Beyond this was) what he calls “the wild beast tract,” which was apparently, considered as too much infested by wild beasts to be susceptible of permanent habitation. South of that again was a mere! desert of sand, destitute of water and producing nothing.4 Though this division is somewhat too strongly marked, and was evidently conceived by Herodotus—with that love of sym- metry and generalisation congenial to the Greek mind—as more definitely characterized than it really is, it is not without a considerable foundation in truthfiid has been adopted, with some modifications, by the most eminent modern geographers, as well as by the Arabic writers. of the middle ages.5 i'7MItiimiswsingular' that it is’much more clearly marked in the western part of Africa, with which Herodotus was comparatively little acquainted, than in those portions concerning which he had more definite information: but, speaking in ‘a general way, the three zones may be fairly regarded as extending across Africa from the Atlantic Ocean to the confines of Egypt. /The principal interruption to its symmetry, arising from the projec- tion of the Carthaginian territory to the north, was unknown to Herodotus, whoundoubtedly shared the error of almost all his successors in regarding the coast line of northern Africa as ‘4 ii. 32; iv. 181. Thus the young N asamonians, on their exploring expe- dition into the interior, at first passed through the inhabited region, then through that of wild beasts, and after- wards through the sandy desert. 5 Heeren, African Nations, vol. i. pp. 6—9; Humboldt, Aspects of Nature, vol. i. p. 58; Bitter, Erdhunde, vol. 1. p. 838. The “ wild-beast” zone of Herodotus (i1 6npa68ns) correspondslin a general way with the Biledulgerid or “Land of Dates ” of the Arabs, a tract which they regard as comprising the whole of the district south of Mount Atlas to the borders of the desert, known to the Roman writers under the name of Gaetulia. In the eastern half of the continent it is much less clearly marked, there being no such important mountain range as the Atlas to constitute a great physical division. The Harudsch or Black Mountains (the Mons Ater of Pliny) are however regarded by sys- tematic geographers as a kind of repre- sentative or continuation of the same chain to the south of the Cyrena'ica, though of such inferior elevation as hardly to deserve the name. T2 276 HISTORY or ANCIENT GEOGRAPHY. can». VIII. comparatively straight,6 so that the parallelism of the three zones would be much more nearly preserved than is really the case. § 11. Closely connected with this division of Africa into three zones, was another of its physical peculiarities, on which Herodotus lays great stress, and in respect to which we again see his love of symmetrical arrangement cropping out in a remarkable manner. Immediately on the borders of the “ wild beast zone,” before one came to the pure unmitigated sandy desert, was a brow or ridge of sand,7 in which ~t_;_l_1,ere .Qcgurred {reisfilntaimfiance . inhabited Spots, each. efnbieh was marked presence Qllelgfi.1¥1@§$§§,_Q§S<, heapedup into hillocks, out of the summit of which gushed forth streams of ‘clear, fresh water. The spots thus characterized were found at "the distance of about ten days’ journey from one another, forming a continuous chain from the borders of Egypt to the Atlantic.s“/The existence of these detached fertile spots in the midst of the sandy desert is a fact that must in all ages have impressed itself upon the minds of travellers and geographers : but it is singular that Herodotus, or his informants, were less struck with the contrast presented by the fertility of these oases9 with the arid waste around them, than with the occur- rence of large masses, or, as he calls them, “ hills ” of salt, in immediate connection with springs of fresh water." The state- ment that these springs in all cases gush forth from the very 6 Note K, p. 308. 7 His expression of Znppzin d/dp/tov, literally “a brow of sand,” has been generally translated as a ridge; and considered as showing that he was acquainted with the true conformation of the tract in question-—a broad table- land of considerable elevation, in which the so-called Oases occupy depressions or hollows of limited extent. But it seems more probable that Herodotus, in common with almost all writers down to a very recent period. thought of the desert only as a vast sandy plain; and that the peculiar expression of Gapi’m was used by him merely to describe a broad strip or band stretching across the continent. A recent traveller in- deed remarks, “ how admirably his ex- pression of ‘a ridge of sand,’ rather than a plain, describes the edge of the northern Sahara” (Tristram's Great Sahara, p. '75); yet Herodotus had certainly never seen any part of the regions visitel by Mr. Tristram, and had nothing but the vaguest infor- mation concerning them. 8 Herodot. iv. 181. . 9 Note L, p. 308. SECT. 1. HERODOTUS : AFRICA. 277 midst of the salt hills is indeed an exaggeration; but all the oases abound in salt—'—sometimes found in such masses as to be used by the inhabitants for building purposes: while springs rise out of the sand, and even on the top of hillocks of sand, in the midst of the salt-covered plains.1 Strabo has aptly compared the fertile patches thus scattered amidst the deserts of Libya tqihe spots on a leopard’s skin ; 2 and they are in fact dispersed overflthe interior of the continent with as little order or symmetry as the spots in question : but Herodotus on the contrary supposed them to follow one another at regular intervals, and in a general direction parallel to that" of sandy, Z0116... itself, 6.8., from eas_tw_rtqwweksft.3 , His informa- tion was probably derived from traders, who knew only the particular line of route which they had themselves followed, and had doubtless paid little attention to the bearings of the track from one station to another. But the permanent cha- racter of these fertile patches—which must in all ages have been equally marked by nature, and have formed as it were the stepping-stones by which alone it was possible to carry on communication across the desert—enables us in many instances to identify with certainty the localities intended by Herodotus, though both the distances and the directions will often be found erroneous. / § 12. A singular mistake at the very outset tends to vitiate his whole line of positions. The first point which he mentions l- is the {hmin’onium or/Oasis of Ammon, rendered so celebrated ; at a later period by the visit of Alexander—the site of which is well ascertained to correspond with that now known as the Oasis of Siwahf" Here there can be no doubt as to the locality /, 1 See the extract from General Mi- nutoli’s account of the Oasis of Siwah, cited by Heeren (African Nations, vol. i. p. 207); and Tristram’s Great Sahara, . 75. p 2 Strabo, ii. 5, p. 130. The com- parison has been frequently reprated by later geographers. 3 Note M, p. 309. 4 The Oasis of Siwah was first visited and described by Browne in 1792; and its identity with that of Ammon fully established by Major Rennell (Geogr. of Herodot. pp. 577-591). It has been since frequently visited and fully described by Hornemann. Oaillaud, Minutoli, Mr. J ames Hamilton and others. The results of all their inves- 2,78 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. meant, and its distance from the nearest point of Egypt is correctly given: but Herodotus erroneously conceived it to be situated west of Thebes, while it really lies about due west of the Lake Moeris, or the district of the Fayoum, from which it is in fact distant about ten days’ journey. The effect of this error is to bring down the whole line of stations more than three degrees and a half of latitude too far to the south. It is probable that in this instance Herodotus had confounded the Great Oasis, which is really situated due west of Thebes, and where there was also a temple of Ammon, with the Oasis of Ammon properly so called, from which it is more than four hundred miles distant.5 § 13. His next station is Apgila, a place which still preserves its name unaltered in its modern appellation of Aujileh—a rare instance in African geography—and is correctly placed about ten days’ journey to the west of the Ammonians.6 It is still visited by the Arabs from the nearest sea-coast, as it was by the Nasamonians in the days of Herodotus, on account of the excellence of its dates.7 But beyond this again, at the same distance to the west, he places the Garamantes, whom he himself calls a very great nation, though he appears to have conceived them as dwelling around a mere oasis like the others.8 Here there can be no doubt that the same people are meant as those designated by later geographers under the name of the Garamantes,—the inhabitants of the region now called Fezzan, a district which has indeed very much of the character of an oasis, though of far greater extent. This tigations are well brought together by M. Parthey (Dos Oral-cl a. die Oase des Ammon, 4to Berlin, 1862, reprinted from the Transactions of the Berlin Academy). The site of the celebrated temple and oracle of Ammon was first discovered by Mr. Hamilton in 1853. 5 Note N, p. 309. 6 Hornemann reached Aujileh from Siwah after nine long days’ journey, but the caravan travelled two whole nights, and both men and boasts were quite exhausted (Travels, pp. 45, 46). The Arab geographers give the dis- tance at ten days’ journey. 7 Hornemann, ~18. 8 I‘apoi/aaz/rres. . . tom ,ué'ya io'xvpéis, iv. 183. The whole district of Fezzan is not less than 300 miles in length from N. to S. by about 200 from E. to but only a small portion of this is capable of cultivation. SE01‘. 1. HERODOTUS: AFRICA. 279 identification is confirmed by his notice, that the land of the Garamantes was distant in a direct line thirty days’ journey from that of the Lotophagi, on the northern coast. This is in fact just about the true distance from Fezzan to the sea-coast at Tripoli or Mesarata, with which it must always have had a direct communication.9 But on the other hand the distance from Aujileh to the nearest point of Fezzan is not less than sixteen days’ ourney, and the direction instead of west is nearly south-west. The effect of this second error is in great part to correct the former one, as Mourzuk, the capital of Fezzan, is in fact almost exactly in the same latitude as Thebes. § 14. At the present day a frequented caravan route proceeds from Mourzuk through Aujileh and Siwah to Cairo,1L and it is probable that this was already the case in the days of Hero- dotus, and that the stations thus far were derived by him from ' communications with persons who were really acquainted with the route. The notice of the direct distance between the Garamantes and the Lotophagi on the sea-coast also points to the existence of intercourse in this direction, which would probably be that followed by traders from Cyrene. But it is much more difficult to determine the positions of the suc- ceeding stages. According to our author there was another oasis ten days’ journey beyond the Garamantes, inhabited by a people called the Atarantes,2 and another again at the same distance beyond that, at the foot of a lofty mountain called Atlas, from whence the inhabitants derived the name of 9 Captain Lyon in 1820 took just ‘ ing ’A'rcipavres is restored from Stepha- thirty days (exclusive of stoppages) on | nus of Byzantium and Eustathius (ad his return journey from Mourzuk to Dionys. Per. v. 66), who however both the sea. ) cite the distinction between the Ata- 1 This was the route by which Hor- ,1 rantes and Atlantes from an author nemann travelled. He took sixteen ; named Rhianus, not from Herodotus; days of rapid travelling from Aujileh l and there seems no doubt that the cor- to Zuilah (the first town where cara- l ruption in the text of Herodotus is at vans halt on their arriving in Fezzan), : least as old as the time of Eustathius. but the distance is commonly reckoned ; Yet it seems impossible that he can at twenty days’ journey. } have called both tribes Atlantes, with- 2 It must be remarked that this name out any further notice or explanation. has been corrupted in all existing MSS. ' Both names are in all probability only of Herodotus into "Arman/res. 'l‘hcrcad- forms of the same. 280 HISTORY OF ANCIENT GEOGRAPHY. OIIAP. VIII. Atlantes. Mount Atlas is described, in terms which certainly iseem to be derived from the old Greek fables connected with {the name, as a lofty mountain of a circular form and so high ;that its summit could never be seen, being covered with clouds ‘both in winter and summer, on which account the natives called it “ the pillar of heaven.”3 Both the name and the description, however fanciful and distorted, seem to require us to seek this last station in the regions to the south of the great chain of Mount Atlas, and not far from the foot of that chain. In this case the line of route must have turned up towards the north-west, from Fezzan by way of Ghadamis to Wargla or Tuggurt; and the next oasis from the land of the Garamantes would be that of Ghadamis,—the Cydamus of Pliny and Ptolemy—a place ' that could hardly have been at any time unknown to the nations on the coast. It is about the same distance from the borders of Fezzan towards the north-west as Aujileh is towards the north-east. This may therefore be fairly presumed to be the spot where our author places the Atarantes. But the position of the Atlantes cannot be determined with anything like even plausibility; no ‘part of the real chain of Mount Atlas approaches sufficiently near to the central zone of desert to agree even approximately with the account ‘reported to Herodotus, and the whole description is so fanciful as to render it very unlikely that it was based upon any authentic in- formation!‘ § 15. Thus far, however, Herodotus himself supposed that he possessed accurate information; but beyond this he himself tells us that he was not able to give the names of the tribes that inhabited the sandy belt which he is describing. But he assures us that this peculiar zone itself extended to the sea beyond the Pillars of Hercules; and contained mines of salt, with the masses of which the inhabitants built their houses.5 3 Herodot. iv. 184. notice this peculiarity only in regard to 4 See Note C, p. 310. the most distant of these Cases; the 5 It is singular that Herodotus should practice being really found much nearer SEOT. 1. HE RODOTUS Z AFRICA. 281 To the south of this tract, which, sandy and sterile as it was, was thus studded from distance to distance with inhabited spots, the regions towards the interior of Libya are described as utterly desert, without trees, without water, without rain, without even wild beasts, and as Herodotus expressively adds, without any trace of moisture.6 As applied to the vast desert of the Sahara this description is scarcely exaggerated; of the fertile and Well-watered regions to the south of it Herodotus had evidently no suspicion, and his ideas of physical geography were too vague to lead him to the conclusion that if these regions were really traversed, as he supposed them to be, by a great river like the Nile, they could not be condemned to unmitigated sterility. I § 16. The only notice which appears to refer to a people to the south of the tract which we have been here considering, is where he speaks of the Garamantes as carrying on expeditions in pursuit of the Troglodyte Ethiopians, whom he describes as the swiftest of foot of any people he had heard of, on which account they were pursued by the Garamantes in chariots and four. These Troglodytes, he adds, fed on serpents, lizards and other reptiles, and their language was like the squeaking of bats.7 At the present day the inhabitants of Fezzan carry on similar razzz'e or slave-hunts, against the Tibboos of the interior; many of whom, inhabiting the more rocky portions of the desert, are still “ dwellers in caves ;” their agility is pro- verbial, and their language is still compared by their more civilized neighbours to the whistling of birds.8 In calling these Troglodytes (of whose position he had probably no clear home. Thus it is mentioned by Mr. Hamilton ( Wanderings in North Africa, p. 294) as still prevailing at Siwah, and by Dr. Oudney in the western parts of Fezzan (Denham and Clap- perton’s Travels, p. 46). Pliny also notices it as the practice of a tribe in the interior from the Syrtis (H. N. v. c. 5,§ 34). Herodotus justly observes that it is a _ conclusive proof of the perfect dryness of the climate, wherever it is found: “ for if it rained, the walls built of salt could not stand ” (iv. 185). ‘5 iv. 185. M11 liquor/.505‘ e’a'rl 6’1/ aim-f)‘ obaév. 7 Herodot. iv. 183. 8 Hornemann’s Travels, p. 119; Lyon’s Travels in Northern A frz'ea, 4to. Lond. 1821. 2.82 HISTORY or ANCIENT GEOGRAPHY. GHAP. VIII. idea) Ethiopians, Herodotus doubtless uses the term merely in the sense of “black men.” The Tibboos are in fact, as a race, almost black, though they have not the distinctive features of the negro. It may well be doubted whether Herodotus ever saw a true negro, the curly hair, of which he speaks as dis- tinctive of the Ethiopians, being found more or less markedly in all the black races of Northern Africa. § 17. With the Libyan tribes that bordered on the coast of the Mediterranean, from the confines of Egypt to those of the Carthaginian dominions, the information possessed by Hero- dotus, as might be expected, was comparatively complete and satisfactory. His knowledge of them was evidently derived from the Greek settlers at Cyrene and the neighbouring colonies, which were at this period flourishing and civilized communities; and not only carried on trade with the neigh- bouring tribes, but exercised in some degree a civilizing influence over them, so that those who most nearly adjoined the Greek territories—the Asbystae and the Auschisae—to a great extent imitated the customs of the colonists.9 These last tribes he describes as being particularly skilled in driving chariots with four horses,1 a practice which they may have derived from the Cyrenaeans-—who were renowned throughout Greece as charioteers; though Herodotus himself, in another passage, asserts that the Greeks themselves first derived the practice of yoking four horses abreast from the Libyans—a statement which it is difficult to accept.2 But there seems no doubt that the use of chariots was general in his days among the Libyan tribes, though now wholly unknown in Northern Africa.3 9 Herodot. iv. 170. 1 'reBpnr'zroBd'raz 8% of": 'ifilao'ra, dAAd yoimo'ra ALBuwV 6201. (l. c.) 2 Ibid. 189. The Greeks of Cyrene were undoubtedly distinguished for their skill as charioteers. Pindar cele- brates the victories of Arcesilas of Cy— rene at the Pythian games (Pyth. iv. v.); and Sophocles speaks of two Libyan chariotccrs as contending with those of the Greeks in the race where Orestes was supposed to have perished (Electra, v. 702). But we cannot sup- pose that the poet here alluded to native Libyans. 3 Thus our author, as just mentioned, represents even the Garamantes of the interior as pursuing the Troglodytes with chariots and four ! (iv. 183.) SE01‘. 1. HERODOTUS: AFRICA. 283 The enumeration of the tribes along the shores of the Medi- terranean presents few geographical difficulties, though, as is always the case with nomad races, both their names and boundaries are in some instances different from what we find in later geographers. Thus the Giligammae who, according to Herodotus, occupied the coast-district eastward of the Cyrenai'ca, nearly to the confines of Egypt,‘ are not mentioned by any later writer; while the name of the Marmaridae, who appear in all the other geographers—even in Scylax, not more than a century afterwards—as the chief people in this region, is not to be found in Herodotus at all. In this case there is little doubt that the same people is meant, and the change is merely one of name, though it is one of which we can offer no explanation. § 18. The coast of the Cyrena'l'ca, from the neighbourhood of Derne to that of Euesperides (the modern Benghazi) was wholly occupied by Greek settlements; here, therefore, the Libyan tribes of the Asbystze and the Auschisae were confined to the interior.5 But to the west of these again on the sea- coast adjoining the bight of the Great Syrtis, we find the N asamonians, a people of Libyan race, who are mentioned by all succeeding writers down to the time of the Roman Empire. The Psylli, who had formerly been the neighbours of the Nasamonians, had according to Herodotus ceased to exist in his time, having undertaken an expedition into the interior, in which their whole nation perished, being overwhelmed with storms of sand.6 The story that they had thus marched into the desert “to make war on the south wind” is obviously a 4 iv. 169. ’ of this story is found in the fact that 5 This was the country which pro- ; no mention appears of the Psylli in duced the famous silphiam, which was . Scylax, a century after Herodotus; one of the principal sources of the though they reappear in Strabo (xvii. wealth of Cyrene. It is singular that p. 838). No mention occurs in Hero- this peculiar and valuable plant is only dotus of their skill as serpent charmcrs, mentioned incidentally by Herodotus ' an art for which they were so cele- {iv.169, 192), though he enumerates . brated in later times. (Lucaml’harsal. all the wild animals of these parts of ix. vv. 890-940; Plutarch, Cat. Min. Libya. 3 56; Plin. II. N. Vii. 2, §l‘l.) 6 Herodot. iv. 173. Some confirmation 284 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. mere fictitious embellishment, but curious as showing the same kind of feeling in regard to this wind—the bane of all these regions of Africa—as that which led the Atarantians to curse the sun as he rose in the heavens.7 The long strip of coast from the bight of the Greater to that of the Lesser Syrtis (the modern Pashalik of Tripoli) was occupied in the time of Herodotus by three tribes, the Macae, the Gindanes and the Machlyans, who succeeded one another from east to west, though their exact limits cannot be defined. Besides these the Lotophagi--a name evidently of Greek origin and retained from its connection with the Homeric poems—are mentioned as occupying a promontory projecting into the sea, which can be no other than the peninsula of Zarzis, adjoining to the Lesser Syrtis, and opposite to the island of Meninx, which was regarded by most ancient writers as the island of the Lotophagi. But Herodotus himself tells us that the use of the lotus-fruit as an article of diet was not confined to these Lotophagi alone, but was common to the neighbouring tribe of the Machlyans; it was doubtless in fact in more or less general use along this part of the coast, as it continued to be .in the days of Polybius.8 The Machlyans, according to Herodotus, occupied the eastern shore of the Lake Tritonis ; on the opposite side of the same lake were situated the Auseans, who were still a mere nomad tribe, like the others just enumerated; but beyond them, towards the west, were agricultural races, differing in many respects in their manners and customs from those to the east- ward. It is evident that we here come in contact with the more civilized tribes, occupying the fertile regions of the modern Regency of Tunis, which, both under the Cartha- ginians and the Romans, was one of the richest countries of the Mediterranean. But the knowledge of them which had reached the ears of Herodotus, was still very imperfect. It is evident that his Cyrenaean informants were very little 7 Herodot. iv. 144. Strabo relates the same thing of the Ethiopians in the upper valley of the Nile (xvii. p. 822). 8 Note P, p. 311. SECT. 1. HERODOTUS: AFRICA. 285 acquainted with any part of Africa west of the Lesser Syrtis; probably they were prevented, by the commercial jealousy so frequent among ancient nations, from trading directly with any of the tribes subject to the Carthaginian rule, or from touching at any seaports, except Carthage itself. The informa- tion which Herodotus had received concerning the tribes beyond Lake Tritonis was further confused in his mind by the erroneous notions which he entertained of the geographical conformation of this part of Africa. He evidently conceived the tribes which he enumerates in succession—the Maxyes, the Zau'e'ces, and the Gryzantes,9 as occurring in that order from east to west; while they probably in reality trended more towards the north, and the neighbour- hood of Carthage. But no notice is found in Herodotus of the proximity of that great city. Nor does he appear to have been aware of the great fertility of this part of Africa; as he distinctly tells us that the district on the banks of the little river Oinyps 1 was the only portion of Libya that could bear any comparison in point of fertility with the richer districts of Europe and Asia.2 The island Oyraunis, which Herodotus places opposite to the Gyzantes, and describes as long and narrow, is clearly the Oercina or Oercinna of later geographers—still called Kar- kineh; though it is strangely supposed by N iebuhr to be the same with the island of Oerne on the Atlantic coast of Africa. Herodotus, as is justly observed by Major Rennell, “un- questionably intended an island in the Mediterranean, and 9 Two of these tribes—the Maxyes i ern portion of the Roman province of and the Gyzantes—are evidently the Africa. The attempt to connect the same that were already mentionel by name of the Zaueces with that of Zeu- Hecatzeus under the names of the gitana is much more questionable. Mazyes and the Zygantes (Hecatzeus, 1 Note Q, p. 312. Fr. 304, 306), but they are otherwise 2 Herodot. iv. 198. Even Timeeus, unknown. It has been. however, sug- nearly two centuries after Herodotus gested with some probability that the appears to have been ignorant of the name of the Gyzantes or Zygantes great fertility of the Carthaginian ter- reappears in that of the Byzantes or ritories in Africa, for which he is de- Byzacians, who gave name to the dis- servedly censured by Polybius. trict of Byzacene, forming the south- 286 HISTORY OF ANCIENT GEOGRAPHY. CHAT‘. viii. that near Carthage.” 3 It is true that no gold is now found in the island of Karkineh; but the story‘ of its being found at Cyraunis, “in a lake, out of which it was drawn up by young maidens, by means of feathers dipped in pitch ” is an obvious fable, and is related by Herodotus himself with an expression of doubt.’* § 19. In some respects, however, he had formed a correct general idea of this western portion of Libya, as altogether different from the fiat and sandy tracts towards the east, being in great part mountainous, well wooded and abounding in wild beasts, especially lions, elephants and serpents of vast size.5 It was here also, according to the stories told by the Libyans, ' " that were found the fabulous beings who were placed by a i, vague tradition somewhere in the interior of Africa—the men Eyith heads like dogs, and those who had no heads at all, but 'ad eyes in the middle of their breasts. But of such tales as these Herodotus is careful to express his disbelief. The stories of the existence of wild men and wild women may perhaps have arisen from. the occurrence of large apes or Pongos, like those met with by Hanno on the western coast, though none such are now found to the north of the Great Desert. § 20. Two other points of interest in the geography of Northern Africa remain to be noticed. Herodotus repeatedly mentions “ the Syrtis: ”’_but always without any distinguish- ing epithetfiiddfliltwisv cllelar'lth'at‘ he knew of only one gulf 'Hf—Thflewname—that called by subsequent geographers the Greater Syrtis, between Berenice (Benghazi) and Leptis Magna (Lebdeh). Nor does he make any mention of its peculiar dangers, of which such exaggerated notions pre- vailed at a later period.7 On the other hand he speaks of the Lake Tritonis, in a manner that may almost lead to the inference that he con- founded the inland lake of that name with the deep gulf known 3 Rennell, Gcogr. of Herodot. p. 639, 4to. ed. 4 iv. 195. 5 iv. 191. 6 ii. 32, 150; iv. 169, 173. 7 Note R, p. 313. SE01". 1. ' HERODOTUS : AFRICA. _ 287 to later geographers as the Lesser Syrtis, and now called the Gulf of Cabes. It seems certain at least that he regarded the Lake Tritonis, which he supposed to be of very large extent, as communicating directly with the sea.8 It is not improbable that considerable physical changes may have taken place in this part of Africa, during the historical period, and that the Lake Tritonis may in the time of Herodotus have been much more extensive than the shallow salt lake (called Sebkah-el Faraoun) that now represents it. It appears probable also that it then had a narrow outlet by which it communicated with the sea, from which it is still separated only by a narrow isthmus of sand: but the information of Herodotus concern- ing these countries is far‘ too vague for us to rely upon it as proving the extent of the changes that have taken place since his time. It must be added that the Lesser Syrtis is correctly described by Scylax, who wrote only about a century after our historian.9 § 21. We have seen that Herodotus evidently possessedvery I imperfect information concerning the portion of Africa ex- tending from Garthageuto the Pillars of Hercules, though_,_it was the part of the continent which was best known ____to “later geographers. But he has preserved to us one curious piece of information concerning the trade of the Carthaginians with the regions beyond the Straits, which he derived (as he himself tells us) directly from Garthaginian informants.1 Outside the Columns of Hercules (he says) there was a district, inhabited by natives of Libyan race, which was frequented by Cartha- ginian merchants. These were accustomed, on arriving at the spot, to land their cargoes and set out their Wares on the sea- shore: they then kindled a fire and retired to their ships. The natives were guided to the spot by the smoke, and after ex- amining the goods offered, deposited by them a quantity of gold, and retired in their turn into the interior. If the Car- 8 Thus he speaks in iv. 179 of Jason being driven among the shoals of Lake Tritonis before he saw the land. 9 Note S, p. 314. 1 Herodot. iv. 196. 288 HISTORY or ANCIENT GEOGRAPHY. can». VIII. thaginians on landing again were satisfied with the amount of gold offered, they took it and went away: if not, they again withdrew to their ships, till the natives had increased it to an amount that they deemed satisfactory. Thus was the tratfic carried on, with mutual confidence, but without any direct communication between the two parties. N o indication is furnished by Herodotus of the locality where this “dumb commerce” was carried on: but the fact of gold being the object of the trade leads to the inference that it was at a considerable. distance towards the south, there being very little gold found north of the Sahara. The practice itself is found by modern travellers to exist in several countries of Africa, where the Moorish merchants have to deal with Negro tribes, who are probably afraid to show themselves for fear of being kidnapped as slaves.2 The only point on the Atlantic coast of Africa which Hero- dotus mentions by name’ QapewSgloeis, which he regarded as the most westerl‘j'rwpioint of the continent. Of this also he had doubtless heard from Carthaginian traders: but we must not assume that, because he was thus rendered familiar with the name, he had any definite idea of its true geographical position.3 There is certainly no foundation for the supposition that he as acquainted with the island of Cerne, so well known to later writers, as one of the principal stations of the Carthaginian commerce on the shores of the Atlantic.4 § 22. In order to complete this review of the knowledge of the African continent possessed by Herodotus, it remains only to consider the very curious and important narrative upon which he founds his belief that it was surrounded by the ocean on the south, so that, as he expresses it, the Erythrman Sea (the Indian Ocean) and the Atlantic were one and the same. We have already seen that this was the prevalent idea in his time: 2 See the account given by Captain practice in his day on the west coast Lyon (Travels in Northern Africa, p. of Africa (p. 100 in Ramusio, vol. i.). 149), and by Shaw (Trrvcls, p. 239). 3 See Chap. V1. p. 163, note. Ca da Mosto also mentions a similar , 4 Note T, p. 316. SECT. 2. HERODOTUS : AFRICA. 289 ‘A most, if not all, preceding geographers having assumed the notion of a circumfluent ocean, surrounding all the three con- tinents. This view however Herodotus rejected as not war- ranted by any credible testimony; but while he expresses his doubts of the existence of an ocean to the north of Europe and Asia, with regard to Africa, he says, no doubt can exist that lit is surrounded by the sea (with the exception of the narrow ' isthmus of Suez) inasmuch as it had been circumnavigated by order of the Egyptian king, N echo.5 La" SECTION 2.--C'irer.tmnavigation of Africa by Necho. § 1. According to the information collected by the historian—— the source of which he has unfortunately not indicated—the Egyptian monarch, who appears to have been a man of an enterprising character, and had previously made an attempt to connect the Nile by a canal with the Red Sea, having been foiled in this undertaking, turned his attention to warlike expeditions both by sea and land, and fitted out two fleets of triremes, one on the Mediterranean, the other at the head of the Red Sea, where he constructed docks, the remains of which were still visible in the time of Herodotus.6 At the same time he sent out a squadron of ships, manned by Phoenicians, with orders to sail- round Africa, and return by way of the Pillars of Hercules into the Mediterranean. These Phoenicians, setting out from the Red Sea, sailed along the southern ocean. When the autumn came on, they landed at the point of Libya where they found themselves for the time being, and having sown a crop of corn, waited there till it was ripe, reaped it, and then 5 Herodot. iv. 42. A186” pév yap 817M? was the son of Psammetichus, and éwirrrp/ éoiiaa weplpfiv'ros, 106711 3001/ ai’rr'fis reigned from 3.0. 610 to B.O. 594-. He 1rpbs 'r'ilv ‘Aa‘lny oilplget, NGKZB ‘r03 Aiymr- therefore belonged t0 a period con- rlwv Baa'zkéos 'II'PGI’TOV 16311 1511.613‘ ‘layer cerning which Herodotus had good Ka'raSéEw/ros. historical information. 6 Id. ii. 158, 159. Necho, or Neco, I VOL. I. U 290 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. proceeded on their voyage. In this manner two whole years elapsed, and in the course of the third year, having doubled the Pillars of Hercules they returned in safety to Egypt.7 Herodotus adds a circumstance, which, he says, “ appears to him incredible, but others may perhaps believe ”—that accord— ing to the report of the navigators, while sailing round Africa, they had the sun on their right hand. We now know that this must really have been the case, had the voyage been actually performed, however strange it appeared to Greek hearers, V and there is no circumstance that has had so much weight in obtaining credence for the narrative in question as this very statement, which was regarded by Herodotus himself as unworthy of belief. § 2. There are few questions in ancient history or geography that have been the subject of more discussion than the reality of this supposed circumnavigation. N o doubt can reasonably be entertained that Herodotus has reported correctly the information he had received: and it is certain that he himself believed iri the truth of the narrative, and looked upon it as establishing beyond a doubt the peninsular character of Africa, and the connection between the eastern and western oceans. But it was far from obtaining general credence in ancient times. Its authenticity was doubted by Posidonius as not attested by sufiicient evidence,8 and in this judgment Strabo concurs. Both these authors however believed in the geogra- phical fact that Africa was surrounded by the Ocean, though they doubted any one having ever actually sailed round it. Polybius on the other hand expresses doubts whether the sea was really continuous to the south of Africa,9 and Ptolemy at a later period wholly disbelieved it : they must therefore have rejected as unworthy of credit the narrative reported by Herodotus. 7 Herodot. iv. 42. authorities, that Posidonius (or, Strabo) 8 Posidonius ap. Strab. ii. p. 98. It refers to the expedition as having been is worthy of notice as showing how sent out by Darius. carelessly the ancients often cited their \ 9 iii. 37. SECT. 2. HERODOTUS : AFRICA. 291 Among modern writers on the contrary there has existed the greatest diversity of opinion. We now know with certainty that Herodotus and those who agreed with him in considering Africa as surrounded by the sea were geographically correct, and the possibility of the voyage cannot therefore be denied. But this is a very‘ different question from that of its actual accomplishment. ~ § 3. Among those who have believed in the authenticity of the narrative, Major Rennell is undoubtedly the one who has done the most‘ to support it. He has shown,1 1. That the time allotted for the expedition was amply suficient for its accom- plishment even according to the slow rate of ancient navigation —a point which it appears strange should ever have been contested; 2. That from the time the voyagers passed Cape Guardafui (the north-eastern point of Africa) supposing them to have chosen the right season of the year, they would be favoured by the northern monsoon, as far as the southern tropic, and would also have a strong current in their favour the whole way round the Cape of Good Hope ;2 3. That along the western coast of Africa, so long as they were within the southern hemisphere, they would for the most part have been able to' reckon upon a favourable wind, as well as a current to the northward. The combination of these circumstances un- doubtedly proves that the circumnavigation of the continent would be much more easily effected from the East than from the West; and thus tends to dispose of the argument against its probability that might be derived from the failure of other attempts made in the contrary direction (such as those of Hanno 3‘ and Sataspes), and from the very slow progress of the Portuguese discoveries in the same quarter. knowledge in this direction. 1 Geography of Herodotus, pp. 672— 71-1. 2 It is remarkable, however, that the Arabian navigators in the middle ages, who undoubtedly visited the eastern coast of Africa as far south as Sofala, never passed beyond Cape Corrientes, which formed the extreme limit of their 3 There is, however, no evidence that the voyage of Hanno was under- taken with any view to the circum- navigation of Africa. It is, indeed, frequently referred to by later writers as such an attempt, and is even de- ‘ scribed by Pliny as having accomplished U2 292 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. § 4. At the same time it must be admitted that the enormous length of the voyage—so entirely out of proportion to all that we know of as accomplished in ancient times, either before or afterwards; the difi'iculty of carrying provisions for such long intervals as would be required, notwithstanding the supposed halts for the sowing and reaping of successive harvests : 4 and the very imperfect account that appears to have been preserved of so important and remarkable an achievement—all contribute to raise a prirna facie improbability against the whole story which it is difficult'to surmount. The absence of all geographical details prevents the possi- bility of testing the truth of the narrative (as in the case of Hanno) by an examination of such statements. But it must he confessed that this total absence is in itself a suspicious circumstance. We do not know from whom Herodotus derived his information, whether from Phoenician merchants at Tyre,5 or from the Egyptian priests: the latter is the more probable: but in either case it is singular that no single fact concerning the wild tribes with which the navigators must necessarily have come In contact, no “ traveller’s tales ” about the marvels of such distant lands and the perils of so long a navigation, should have been preserved to the time of Herodotus, or have been known to his informants. it (l); but the original narrative of ’ Africa, making arrangements to cm- the voyage represents it only as under- ' ploy the same expedient (Strab. ii. p. taken for the purpose of founding 100). Hence it would naturally be colonies along the west coast of Africa. supplied, as a necessary incident, by 4 Herodotus undoubtedly conceived those who believed in and reported the the Phoenicians as repeating this ope- * story. ration more than once, though Major 5 Herodotus certainly visited Tyre Rennell appearsto have supposed it to (see ii. 44): but, whether from the have taken place only once. This is shortness of his stay or from some other evident from the form of his expressions cause, appears to have collected but (1rpoo'lo'xou'res 3w onefpemcov . . . Bepi- little information there. At least we o'av'res at Thu cr'irov) as well as from the do not find him referring to the testi- duration of the voyage. Such a mode mony of Tyrian or Phoenician inform- of meeting the probable difficulty from ants with regard to any geographical want of provisions seems to have sug- questions. The Carthaginians, whom gested itself naturally to ancient navi- he repeatedly cites, he may have met gators, as we find Eudoxus, when pre- either at Tyre or at Cyrene. paring for the circumnavigation of SECT. 2. ' HERODOTUS: AFRICA. 293 § 5. The one only exception to this total want of circum- stantial evidence is the fact reported by the navigators that in sailing round Libya they had the sun on their right hand. Such would undoubtedly have been the case if they had really pene- trated into the southern hemisphere; but as no Greek had ever done this, Herodotus—whose theoretical knowledge of astro- nomy was certainly of the vaguest description—was naturally led to reject the statement as incredible. In modern times on the contrary, it has been frequently regarded as the strongest proof in favour of the whole account. A recent translator of Herodotus even goes so far as to say that “few would have believed the Phoenician circumnavigation of Africa, had it not been for this discovery.” 6 Yet it may well be doubted whether we are warranted in hastily assuming that such a statement must necessarily have been derived from personal observation. The Egyptian priests were well aware that the sun was vertical at Syene at the time of the summer solstice: and it was an inference so natural as to be almost inevitable that any one proceeding further south would have the sun to the north of him. The frequent inter- course with Meroe would confirm this view. It is probable moreover that Phoenician navigators had already frequented the coasts of the Erythrzean Sea, considerably to the south of the Tropic of Cancer: and even in the particular voyage in question—if we suppose that the narrative had any founda- tion in fact, and that an exploring expedition was really sent out by N echo, it would easily have attained to latitudes where the phenomenon in question might be observed during a part of the year. Nothing is more common than to have theoretical inferences converted into statements of fact; and if the in- formants of Herodotus supposed—as he himself undoubtedly did, in common with almost all the Greek geographers in later times—that the continent of Africa trended away rapidly to the west, from near the mouth of the Red Sea, the assumption ‘ Rawlinson’s Herodotus, vol. iii. p. 35, note. 294 HISTORY or ANCIENT GEOGRAPHY. CHAP. VIII. that navigators proceeding along its southern coast, from east to west, would have the tropical sun to the north of them. (and therefore on their right hand) would certainly not require a greater amount of astronomical knowledge than was possessed at this period both by Egyptians and Phoenicians.’z On the other hand it is certainly remarkable that no notice is taken, or at least no mention preserved, of the change of seasons in the southern hemisphere—a circumstance which must havebeen the more strongly impressed upon the minds of the navigators from its intimate connection with the choice of times for halting, with the purpose of sowing and reaping corn for their own supply.8 Nor is anything said of other changes in the celestial appearances, such as‘ the disappearance of the Great Bear and the pole-star, by which the Phoenicians were accustomed to steer, and the loss of which must therefore have been a source of great perplexity to them in the southern hemisphere. It may be said indeed, that the extreme concise- ness of the narrative, as reported by Herodotus, renders all such negative arguments of little value : but unfortunately it is that very brevity, which, by depriving us of all incidental corroboration, leaves us simply to choose between the bare statement of the fact on the one side, and its great intrinsic improbability on the other. Of course the statement of Herodotus, that the fleet returned by way of the Pillars of Hercules, would be conclusive evidence that it had really sailed round Africa; if we could rely upon its accuracy. But if we suppose the story to have acquired general currency, it would be readily seen that this must have 7 Mr. Cooley’s remarks on this sub- ject (in his edition of Larcher’s Notes on Herodotus, vol. ii. p. 30) are cer- tainly well-founded. See also Mr. Blakesley’s note to the passage, iv. 42. 8 It is perhaps stretching this argu- ment too far to lay much stress (as has been done by M. Gosselin) upon the mention of autumn (firms 6% 'yfuoz'ro ¢0w61rwpou) as‘the season for sowing their corn: such a phrase might natu- rally have been introduced by Hero- dotus himself in relating the story he had heard. But the arrangements of the voyagers for this object must have been so materially influenced by the changes of the seasons, both in the southern hemisphere and in the tropics, that it is difficult to understand how they could accommodate themselves to these alterations, without any previous knowledge to guide them. SE01‘. 2. HERODOTUS! AFRICA. 295 been the case, and such a fact would naturally be added by one of the successive narrators of the tale. § 6. The argument derived from the total neglect of this voyage and the important discovery supposed to have been thus made, in subsequent times, till it came to be totally dis- believed by the most judicious ancient geographers—an argu- ment on which great stress'has been laid by some modern writers, is certainly far from conclusive. Similar instances- may be cited in comparatively recent times. Major Rennell has pointed out that the discovery of the eastern coast of New Holland, generally attributed to Captain Cook, had really been made by a Dutch voyager near 150 years before; a circumstance wholly unknown to the great navigator, as well as to all his contemporaries.9 A still more striking instance (probably unknown to Major Rennell himself) was that of the discovery of the continent of North America under the name of “Vinland ” by the N orthmen in the eleventh century—a discovery which had been entirely forgotten, until it was brought to light again in our own days. The expedition sent out by N echo—if we are to believe that the narrative had any foundation at all—was intended solely for the purpose of settling the geographical question, and resembled in this respect the voyage of Scylax of Caryanda, who explored the Erythraean Sea by order of Darius. The great length of time employed would in itself be sufficient to deter future navigators from attempting to follow the example for commercial purposes; and the discovery, even if it were really made, would not have had the effect of opening out a new commercial route to other lands. The circumnavigation of Africa by the Portuguese, in the fifteenth century, was accomplished with a view to arriving in 9 Rennell’s Geography of Herodotus, discovered by the Portuguese at a still p. 714. See also Major’s Prince Henry earlier period, though never recorded the Navigator, p. 442. From the re- in an authentic form, or at least in searches of this recent writer it appears such a form as to attract general at- clear that a considerable portion of the tention. coasts of Australia had been actually _______- ._.__.______~- 296 HISTORY OF ANOIENT GEOGRAPHY. carp. _VIII. India; and the result fully justified the importance attached to it on that account.1 N 0 such prize would reward the Phoenician voyagers, or lead other navigators to follow in their track. Yet it certainly seems strange that they should not have held sufficient intercourse with the natives to discover the great abundance of gold existing among them, either on the coasts of Mozambique or those of Guinea; and gold was- undoubtedly, in ancient, even more than in modern, times, the "attraction which none could resist.2 § 7. On the whole it may be said that the alleged voyage of the Phoenicians under Necho is one of those statements that cannot be disproved, or pronounced to be absolutely impossible; but that the difficulties and improbabilities attending it are so great that they cannot reasonably be set aside without better evidence than the mere statement of Herodotus, upon the authority of unknown informants.3 We have abundant evi- dence in later times how easily such reports were got up andt believed—as in the case of Eudoxus of Cyzicus (which we shall have occasion to consider hereafter), and the Indians, who were reported by Cornelius N epos to have circumnavigated the‘ north of Asia and Europe from the Erythraean Sea to the Baltic (l); and we shall see that in the case of Hanno, an expedition which did not really advance as far south as the Equator was magnified into a complete circumnavigation ‘of Africa.‘ §8. Herodotus, however, did not rest his belief in the peninsular character of Africa solely upon the history of this expedition, conclusive as he deemed it. “ N echo (he tells us) was the first to prove it; but besides this (he adds) the Cartha- ginians assert the same thing.” Whether they based their ‘ Hence the name of Cape of Good 2 Sofala, on the coast of Mozambique, ' Hope, which was given to the southern was during the middle ages the tract promontory of Africa by John 11., king from which the Arabian traders derived of Portugal, in anticipation “ of the their principal supply of gold. realization of the long-coveted passage 3 Note V, p. 317. tq India.” (Major’s Prince Henry, p. ‘ See the next Chapter. 3 5.) Seer. 2. HERODOTUS : AFRICA. " 297 belief upon any alleged discoveries of their own, he has un- fortunately left uncertain, though it appears probable that they did so.5 But the historian breaks off abruptly to tell us of another attempt to circumnavigate Africa, which did not prove successful, and leaves us in the dark as to what the Carthaginians may have really accomplished.6 §9. The unsuccessful attempt, just adverted to, is thus related by Herodotus.7 Sataspes, a Persian nobleman of high- rank, having incurred the displeasure of Xerxes, and been condemned to death by that monarch, was allowed, at the intercession of his mother, to obtain a respite of his sentence on condition of sailing round the African continent. For this purpose he went down to Egypt, and having there obtained a ship and a crew, he set sail for the Pillars of Hercules. After passing through the Straits, he doubled the Libyan headland of Cape Soloeis, and proceeded from thence towards the south for a long distance. But after a voyage of many months, having passed over a vast extent of sea, and finding it still stretch farther and farther before him, he made up his mind to return. According to his report, the coast, at the farthest point which he was able to reach was inhabited by a race of men of dwarfish stature, who were clad in dresses made of digression to recount the voyage of Sataspes, who did not succeed in cir- cumnavigating Africa. This account he had probably heard at Samos, whither the eunuch of Sataspes had fled with his treasures. It is certainly a mistake to suppose, as Larcher and Ukert have done, that it was this his— tory of Sataspes, which Herodotus had heard from the Carthaginians. The words e’rrel Ear-dorms 'ye . . . or’: rep:— e’rrAwa'e Azfizinu, e’rr’ az’rrb 'roz'i'ro ‘Ire/“peels are conclusive on this point. 6 This is the more to be regretted, as there is great probability that their views were founded on the voyage of Hanno, concerning which we have confidence, had he not unfortunately authentic information from another been led away (as in so many other quarter. passages in his history) by the love of 1 7 iv. 43. 5 Mr. Rawlinson indeed translates the passage of Herodotus (iv. 43) as distinctly asserting that “ the Cartha- ginians, accordin¢r to their own account, made the voyage:’ and Mr. Blakesley also understands the words in the same sense. But it seems to me clear that the true meaning of the passage is oii'rw ,uev 01.5717 (1‘; 1118151)) é'yVa’io'Br; 'rb 'n'pcb'rov (so. weplppvros £08011) para 5% Kapxn- Edi/rot do‘: of Aé'yov'res (so. rrepfppv'rov ell/ac). This is the proposition that he sets out with asserting (iv. 42), and that all the other statements are in- tended to confirm. He would probably have proceeded to tell us the grounds on which the Carthaginians based their —_—-__ l__h__ __~_____. . .‘nm 298 HISTORY or ANCIENT GEOGRAPHY. GHAP. VIII. leaves of the palm-tree. They were a harmless race, dwelling in towns and possessing cattle, but fled to the mountains on the approach of the strangers. He further alleged, as a reason why he did not continue his voyage, that his ship stuck fast, and would not advance any farther.8 But this account, which Herodotus himself appears to have considered a mere pretence, did not satisfy Xerxes, who ordered the unfortunate voyager to be executed, as having failed in his enterprise. Whatever may have been the real cause of his return, it is clear that Sataspes, by attempting the circumnavigation from the west encountered difficulties and delays which would not have presented themselves had he taken the contrary direction; but there seems no reason to doubt that he had advanced far enough to the south to come ‘in contact with the negro tribes, beyond the great desert, who were comparatively civilized, having fixed abodes and dwelling in towns. Beyond this it is impossible to form any safe conclusion from such a meagre narrative. 8 This is afavourite excuse, or fancy, ] preventing their continuing their voy- with timid voyagers in all ages. The 1 ages beyond Sofala around the Cape of same reason was alleged by the Ara- Good Hope. bian navigators in the middle ages, as NOTES A, B. HERODOTUS! AFRICA. 999 NOTE A, p. 263. USE OF CAMELS. MR. KENRICK (Ancient Egypt, vol. i. p. 76) notices the “almost insurmountable ” difficulty of traversing the Sahara before the introduction of the camel, “which never appears in the monuments of the Pharaonic times :” and he adds in a note: “ ‘N e have such ample representations of Egyptian life, that if the camel had been naturalized there as a beast of burden, it must have occurred in the paintings.” Herodotus tells us that the Arabians brought camels laden with water—skins to supply the army of Cambyses on its march through the desert from Syria into Egypt (iii. 9). But notwithstanding this, they do not appear to have ever come into use among the Egyptians. Quintus Curtius indeed mentions their employment by Alexander the Great during his march to the Oasis of Ammon (iv. 30, § 12), but this appears to have been quite an isolated instance: and strange as it appears to us, it seems to be a well- established fact that the use of camels was practically unknown in Africa until after the Mahomedan conquest. (See the dissertation by Hitter, in his Erdlcnnde von Asien, vol. viii. part ii. pp. 705-710.) Hence the accounts given by Heeren 9 and by Mr. Hoskins 1 of the supposed extent of the commerce of Meroe in very ancient times, which are based throughout upon the supposition of its being a centre from which large caravans of camels traversed the deserts in all directions, are almost wholly illusory. NOTE B, p. 263. DISTANCES ON THE NILE. Yet even in regard to this part of the Nile his notions of the distances are very inaccurate. So much time and ingenuity have 9 African Nations, vol. i. pp. 430-47 2. 1 Travels in Ethiopia, 4to, Lond. 1835. 300 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. been spent in endeavouring to reconcile or explain the statements of ancient writers with regard to distances, in countries which were very imperfectly known, that it is important to observe how widely the statements of Herodotus differ from the truth, even in a country so well known as Egypt, which he had himself visited, and where there is no possibility of error as to the localities. Herodotus tells us in one passage that the distance from Heliopolis to the sea is 1500 stadia exactly (ii. 7), which would be equal to 150 geo~ graphical miles, or about 173 English miles: while in another passage (ii. 9), though apparently referring to the same oomputa~ tion, he allows only 1260 stadia for the same interval. The actual distance to the old Sebennytic mouth of the Nile—the nearest of the three, and therefore certainly the one that we should naturally suppose to be meant—is, according to Sir G. Wilkinson, only about 110 English miles, following the course of the river. If on the other hand we suppose the distance to be reckoned from the Canopic mouth, which was that generally resorted to by the Greeks, the distance would be about 150 miles, a tolerable approach to the truth : but the necessity of having recourse to this supposition in itself shows the vagueness of such estimates, considered as geo- graphical data. Again, the distances above Heliopolis, where no such ambiguity can arise, are equally overstated. Thus Herodotus reckons the distance from Heliopolis up the river to Thebes at 81 Egyptian schoeni or 4860 stadia, equal to about 552 English miles, and that from Thebes to Elephantine at 1800 stadia, or more than 2.06 English miles. But the fogmer distance (according to Sir G. Wil- \ kinson) is really only 421 English miles, and the latter does not ‘ exceed 124 milesf The source of his error is in this instance not difficult to discover. The distances are doubtless given by him according to the estimates of the Egyptian boatm en, who would express them in schceni (axo'ivot), as indeed Herodotus himself does in several passages: and these he reduced into stadia, at the rate of 60 stadia to the schoene (ii. 6). But this Egyptian measure was, as we learn from Strabo (who had himself made the voyage up the Nile), a very vague and unsettled one, varying, he assures us, from 30 to 120 stadia. (Strab. xvii. c. 1, § 24, p. 804). No doubt the distances were in fact merely estimated, not measured at all : and this glaring instance sufiiciently shows us how little such estimates are to be relied on. On the other hand it is remarkable that Herodotus reckons it NOTE 0. HERODOTUS : AFRICA. 301 only nine days’ voyage up the river from Heliopolis to Thebes, probably speaking from his own experience : but such a passage is unusually rapid, and twenty days is not more than “ a fair average passage ” from Cairo to Thebes. Wilkinson’s Handbook for Egypt, . 2. P It )may be added that his enormous exaggeration concerning the size of the Lake Moeris, which he describes as 3600 stadia (360 geographical miles) in circumference (ii. 149), while according to the recent careful investigations of M. Linant de Bellefonds, cited by Parthey (Zur Erdhunde des Alten Aegyptens, p. 511), it could never have exceeded 48 to 50 geographical miles in circuit, shows how little dependence can be placed on such estimates: even where, as in this instance, Herodotus had himself visited the locality. NOTE 0, p. 264. THE DODECASCHCENUS. Herodotus tells us that this part of the navigation, requiring the boats to be dragged by main force up the rapids, continued for 12 schoenz' (equal to 720 stadia) ; this required four days on account of the rapidity of the current and the consequent slowness of pro- gress. This distance of 12 schoeni from Tachompso to Syene or Elephantine is confirmed by an inscription of the time of Tiberius, discovered at Philee: it afterwards gave name to a district called the Dodecaschoen us (AwdeKoio'Xou/os), which occupied precisely this portion of the valley of the Nile. (Ptolemy, Geography, iv. 5, § 74.) The error of Herodotus appears to have arisen from his confounding the interval thus designated with that of the rapids, and sup- posing the difficulties of navigation to extend equally through the whole distance. In other respects his description of the rapids themselves, and the mode of passing them, is very correct, and presents a striking con- trast to the absurd fables current on the subject in later times. The First Cataract, as it is commonly called, is in fact merely a succession of rapids : the entire descent in a space of five miles being only 80 feet. (Kenrick’s Egypt, p. 31.) The Tachompso of Herodotus appears in Ptolemy (l. 0.) under the form of Metacompso, which he places opposite to Pselcis, a well- 302 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. known locality, the ruins of which are still visible at Dakkeh. But Herodotus was misinformed‘as to the size of the island, there being nothing in this part of the Nile but mere islets. NOTE D, p. 266. THE AUTOMOLI. There seems no doubt that these Automoli of Herodotus, whom he mentions only by this Greek form of their name, are the same people noticed by Eratosthenes and several later writers under the name of the Sembritee or Sebritae, to whom a similar origin is ascribed: and the data furnished by these writers enable us to place them with reasonable certainty in the region of the mo- dern Sennaar, in 13° to 14° N. latitude: and about twenty days’ journey above Meroe. (See this subject fully discussed by St. Martin, Le Nord de l’Afrique dans l’Antiqui-té, pp. 24-29), and by Mr. Cooley (Ptolemy and the Nile, pp. 20-23) ; a work with which the French author seems to have been unacquainted.) It is impos- sible to account for the enormous exaggeration of the distance by Herodotus, except from the tendency to vague overstatements com- I mon in speaking of all remote and imperfectly known regions. The contrast it presents with the accurac of his information concerning Meroé is an instructive example of t He difference in the value of his statements according to the nature of his materials—a distinction which cannot be too carefully borne in mind in discussing the statements of ancient writers on geography. NOTE E, P. 266. DISTANCES ON THE UPPER NILE. This is a very fair approximation to the truth. Timosthenes, the admiral of Ptolemy Philadelphus, also estimated the distance from Syene to MeroéatptiO days’ journey, but whether by land or by water is ii’dt statedlh(Plin. ‘H. N. vi. 0. 99, § 183). The distances given by Herodotus in detail give only 56 days, viz., 4 days for the ascent-of the rapids, 40 days by land, and 12 more NOTE F. HERODOTUSZ AFRICA. 303 by water to Meroé : but the omission is obviously to be supplied by the supposed lake, and the navigation thence to the beginning of the upper rapids. This would correspond with the interval between Dakkeh and the second cataract at Wady Halfa, which must always have been the point at which voyagers up the Nile quitted their boats and took to travelling by land. The navigation in this part being unimpeded, the distance might probably be accomplished without difficulty in four days. NOTE F, p. 266. COURSE OF THE NILE FROM WEST TO EAST. Niebuhr is almost the only writer upon the subject who has seen the necessity of applying the words of Herodotus ([5661. 3% a.’ éc-are’pns 76 m2 v‘yMov Svcuéwv, ii. 31) to the course of the river from the country of the Automoli to Elephantine.2 But Valckenaer, in a note to Herodotus (ii. 84), understands and explains the words in the same sense : “ Supra Eyyptum fiuit ab Occidente.” Unless we thus understand them, the comparison of the course of the Nile with that of the Ister becomes quite unintelligible: for there would be no reason to assign to it this long course from the west. Colonel Leake and Mr. Rawlinson both translate the passage as if Herodotus had meant to say that at that point—the farthest to which it was known with certainty, the river came from the west, though up to the land of the Automoli it had had a course from south to north. But Herodotus would surely have given some clearer intimation of such a marked change in its direction had such been his meaning. The passage as it stands at present (,aéxpl ue’v vvv Teo'o'e'pwv uni/6w rrhdov Kai. 580i) ywe'io'Ke'rat 6 Nelhos rroipef 701') 6’1/ Aiyv’rrrcp fieiijuaros . . . . fie'et 3% drr’ e'o'rre'p'qs "re Kai v‘yhiov dueue'wv), certainly seems intended to convey the same idea as we find elsewhere more distinctly expressed in the case of the Borysthenes, the sources of which were also unknown—a passage in all respects very parallel to the one now under consideration : ,ue'xpl. ,ae'v vvv Ile'fifiov Xa'ipov, e’; 751/ reo'o‘epolrcov'ra {Wasps/(av rrAdos éo'n', 'ywo'io'ice'rat fie'wv drrd ,Bope'w dVéILLO'U (iv. 53). In both cases the natural interpretation of the words seems to be that 2 Geogr. of Herodotus, pp. 15, 20. 304 HISTORY OF ANCIENT GEOGRAPHY. OHAP'. VIII. the general direction of the course of the river is meant, not merely its direction at the farthest point indicated. It is singular that almost all the commentators conceive Hero- dotus to have assumed this westerly origin of the Nile, in con~ sequence of the story of the N asamonians. It appears to me that the course of his reasoning was just the converse. He knew, or supposed he knew, that the Nile in the upper part of its course, as far as it was known, came from the west. Beyond that, he says, no one knew anything about it. But when he hears of these Nasa— monians having discovered a great river, with crocodiles in it, flowing from the west, he immediately concludes that that river can be no other than the Nile; an inference natural enough from the data that he possessed. NOTE G, p. 268. SUPPOSED SOURCES OF THE NILE. The only exception was a tale told him by a certain priest, who was registrar of the sacred property in the temple of Athena at Sai's in Lower Egypt, which Herodotus justly rejects as utterly unworthy of credit (ii. 28). According to this fable the real source of the Nile was just on the frontiers of Egypt, between Syene and Elephantine, where there were two mountains with sharplfiieaked summits, called Crophi andwMophiLand Vbeptweeg the, two was an unfathpmabletabyfiss, from which arose the waters that were rS—aIgnu'ces of the .Nilef'one half of which flowed towards the noiith into Egypt, the other half towards the south into Ethiopia. He added, that the abyss had been found to be unfathomable by fPsammetichus, who had in vain attempted to sound its depths with a rope many thousand fathoms in length; a circumstance which, Herodotus justly observes, the fact was true, might be accounted for by the violence of the eddy'at this point, occasioned by the passage of the river between the two mountains. It is unnecessary to point out the absurdity of a story, which represents the Nile above Syene as flowing towards Ethiopia instead of from it; but it is certain that traces of the same idea—whether really of Egyptian origin, or in consequence of its being told by Herodotus—are found current at a later period. Thus when Ger- NOTE‘ H. HERODOTUS : AFRICA. 305 manicus visited Egypt in the reign of Tiberius, he was shown the unfathomable abyss in question; and Seneca alludes to it as regarded by the natives as the source of the Nile (Tacit. Annal. ii. 61; Seneca, Queest. Nat. iv. 2). Herodotus, however, appears to have heard nothing of it, when he himself ascended the river as far as Elephantine, and it was doubtless on his return to Lower Egypt that he first heard the tale from the priest at Sai's, so that he had no opportunity of making enquiries on the spot. It is certainly hard upon Herodotus that he has been frequently censured, both in ancient and modern times (Strabo, xvii. p. 819; Mure’s History of Greek Literature, vol. iv. p. 38 7), for his credulity in relating such a story as this, though he in fact cites it only to express his utter disbelief of it, and thought the priest was making game of him (Ejaot'ye rrouzew e’ddxee gbdnevos eidéval. drpem'ws). But, as often happens in similar cases, an idea once started, however absurd, found supporters who attempted to give it a rational interpretation, and it was doubtless the tale told by Herodotus, which gave rise to the theory, found in some of the Arabic writers, of two Niles taking their rise together, and the one flowing to the north, the other to the south. But they judiciously placed these sources beyond the limits of their own positive knowledge. It was probably this last theory, combined with a very confused and imperfect recollection of the passage in Herodotus, ‘that led an eminent traveller in our own time to look for “ the fountains of Hero- dotus ” in the interior of the continent, far south of the Equator; forgetting that the “ fountains ” as described to the historian were to be found between Syene and Elephantine, within the confines of Egypt itself! (Livingstone’s Last Journals, vol. ii. pp. 50, 169.) NOTE H, p. 271. EXPEDITION OF THE NASAMONIANS. This view is maintained by Colonel Leake, in a paper published in the Journal of the Geographical Society, vol. ii. p. 1—28,3 as well as by Major Rennell (Geogr. of Herodot. p. 432, 4to edit.), in whose 3 The substance of this paper has been reproduced by Mr. E. B. James in the article NIGER in Dr. Smith’s Diet. of Ancient Geography, vol. p. 428. VOL. I. X 306 HISTORY OF ANCIENT GEOGRAPHY. GHAP. VIII. time, however, the geography of Central Africa was still very imperfectly known. It is strongly opposed by Mr. Cooley (in his edition of Larcher’s Notes on Herodotus, vol. i. p. 241), who contends that the river could be no other than the Yeou,_the main river of Bornou, and regards it as “in the highest degree improbable” that the Nasamonians should have reached the Quorra, or river of Timbuctoo. But, without insisting too closely upon the direction assigned by Herodotus, it must be admitted that we do less violence to his statement by supposing them to have travelled in a south- westerly direction, than due south, which they must have done in order to reach Bornou. And with regard to the distance to be traversed, supposing them to set out from Fezzan, which is admitted by both hypotheses, the distance to be traversed before reaching the river of Bornou is not less than 12 degrees of latitude or 7 20 G. miles, as measured in a direct line on the map; while that to the nearest point of the Quorra hardly exceeds 900 G. miles. Supposing the one journey to be possible, it is difficult to say that the difference is such as to render the other impossible. The theory of M. de St. Martin, which agrees in substance with that of M. Walckenaer, that the Nasamonians never crossed the great Sahara at all, but that the river visited by them was only the river of Wargla, to the south of Mount Atlas,‘ appears to me to satisfy none of the conditions of the problem. There is no permanent river there, but a mere wad-i, with a lake, which becomes a mere bed of salt in the summer; there could never have been crocodiles in such a stream, nor could the inhabitants have been materially different from the other Libyan races. The oasis of Wargla is indeed exactly one of those described by Herodotus as situated in the “ belt of sand ;” the broad expanse of the desert stretches out from thence both to the south and west? It was this desert that he conceived the Nasamonians to have crossed; and there appears no alternative, but either to believe that they had really done so, or to reject the whole story as a fiction. Even if we adopt the latter course, it seems difficult to account for it without supposing that some vague rumour of the existence of such a river in the interior of the continent had reached the tribes on the sea-coast. 4 St. Martin, Le Nord de l’Afrique l’Afrique Septentrionale, p. 353. _ dans Z’Antiqnite‘, pp. 16-20; Walck- 5 See the description of it In Mr. enaer, Recherches Géographiques sur Tristram’s Great Sahara, p. 218. NOTE I. HERODOTUS: AFRICA. 307 NOTE I, p. 271. DIFFICULTY OF COMMUNICATION WITH INTERIOR. It has indeed been assumed by several modern writers, more especially by Heeren, that the caravan trade of Central Africa must always have been substantially the same; and that as there are now, and have been ever since the middle ages, frequented caravan routes from Tripoli by Fezzan to Bornou, and again from Morocco to Timbuctoo and the regions on the Niger, similar com- munications must have existed in ancient times. But there seems no doubt that the existing system of caravan trade dates only from the introduction of Islamism‘ into Africa. It was the Arabs who first introduced the camel into Northern Africa, and without camels any extensive intercourse with the interior was impossible. The Negro races have never shown any disposition to avail themselves of this mode of transport, and at the present day the commerce of the interior is carried on almost entirely by Moorish, that is, by Mohammedan, traders. The spread of Islamism has doubtless led to increased communication from another cause, the necessity for the Mohammedan inhabitants of the outlying and detached regions of the continent to make the pilgrimage to Mecca. Even in the most flourishing times of the Carthaginians they do not appear to have made any use of camels; and as late as the days of Strabo the communications with the tribes of Western Africa who dwelt beyond the Sahara were scanty and irregular.6 In the time of Herodotus there is certainly no indication that either the Cartha- ginians or the Greeks of the Cyrenai'ca had any commercial inter- course with the regions beyond the Great Desert. Heeren indeed dwells more especially upon the traffic in gold, and gold dust, which he supposes to have attracted the merchants of antiquity, as it has done the Moorish traders in later times, to brave the perils of the desert, on the north side of which, as he justly observes, little or no gold is to be found.7 But there is in fact no indication to be found in Herodotus, that gold was brought from skins of water under the bellies of their horses ” in order to cross the desert; a remarkable proof of the shifts by which they endeavoured to supply the want of camels. 7 African Nations, vol. i. p. 183. 6 Strabo, xvii. p. 828. The Pharu- sians (as he tells us), whom he con- sidered as dwelling beyond the great desert in Western Libya, held only rare and occasional intercourse with the inhabitants of Mauritania, “tying x2 O 308 HISTORY OF ANCIENT GEOGRAPHY. CRAP. VIII. the interior of Libya at all. It was indeed supposed to be found in fabulous abundance among the Macrobian Ethiopians, who dwelt on the southern sea (iii. 17), but all that is related of them has an air of mere fable. The historian also describes in detail the trade in gold carried on by the Carthaginians with the tribes beyond the Pillars of Hercules on the shores of the western Ocean, but no trace is to be found of the existence of any such trade by land with the interior of the continent. 9 NOTE K, p. 276. ERRONEOUS POSITION ASSIGNED TO CARTHAGE. Strabo supposed the parallel of Carthage to be only 1000 stadia (100 G. miles) to the north of that of Alexandria, and the inner- most bight of the Great Syrtis to be less than 2000 stadia south of Carthage. The real difference is not less than six degrees and three-quarters of latitude, or more than 400 G. miles. Even Ptolemy fell into a Similar error, and placed Carthage less than two degrees to the north of Alexandria. The influence of this error upon their _conception of the geography of the Mediterranean will have to be considered hereafter: but its efi‘ect upon that of the interior of Africa was to bring the land of the Garamantes (Fezzan) nearly on a parallel with that of the Gaetulians, the Biledulgerid of the Arabs : a change which must be carefully borne in mind in discussing the views of Herodotus upon the subject. NOTE L, p. 276. THE OASES. The name Oasis or Auasis, which is of Egyptian origin, is not, indeed, found in Herodotus in this sense: it was probably first introduced by the Alexandrian writers, and was already familiar to Strabo. Herodotus in one passage speaks of an expedition sent by Cambyses against a city Oasis, where the capital of the Great Oasis, west of Thebes, is undoubtedly meant (iii. 26), but he does not seem to have had any idea of the term otherwise than as a NoTEs M, N. HERODOTUS: AFRICA. 309 proper name. Even at the present day the spot thus designated is commonly known as “ el Wah,” “ the Oasis,” in contradistinction to all others. NOTE M, p. 277. SYMMETRICAL ARRANGEMENT OF OASES. This tendency to symmetry and regularity is not peculiar to Herodotus. The Arabian geographers, and the Arabs of the present day, commonly reckon “ten days’ journey” from one of these halting-places to another, with little reference to the exact dis- tance. The journey from the Oasis of Ammon (Siwah) to that of Augila is, in fact, just about ten days’ journey, and this was pro- bably the one best known to Herodotus, from the proximity of these two stations to the Cyrenai'ca. Edrisi, in his geographical work, reckons ten days’ journey from Aujilah to Zala, and again ten days more to Zawilah, which was in his time the capital of Fezzan. It must be remarked also that no allowance is made for the extent of the Oases themselves, though these in several instances occupy a district of considerable size, and that of Fezzan spreads out into an extensive region. Herodotus, indeed, appears to have regarded the “ hill of salt,” which he supposed to exist in each case, as the centre of the habitable district around it, and took no account of the extent of the latter. NOTE N, p. 278. CONFUSION OF GREAT OASIS AND THAT OF AMMON. This appears to me by far the simplest solution of the difficulty. But with their usual unwillingness to admit that Herodotus can have made a mistake, several writers have suggested that an inter- mediate station has fallen out of the text, and that the route he has traced lay through the Great and Little Oasis to that of Ammon. It is not impossible that this may have been the case with regard to his original information, but there is no room to doubt the accu- racy of the text of Herodotus as it stands : and this must be taken therefore as correctly representing his. view of the matter. In like manner Heeren has suggested that another station. may have been 310 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. omitted between Augila and the Garamantes, the real distance being nearly double that indicated by Herodotus. But, as N iebuhr justly observes, the soundness of the text in all these passages admits of no doubt : and we have only to deal with them as we find them. Whether the original error lay with Herodotus or his informants, it is impossible for us now to determine. It is, however, not improbable that the confusion in this instance may have arisen from the circumstance that the same Egyptian deity, whom Herodotus calls the Theban Zeus, and who was generally known to the Greeks as Zeus Ammon,had a temple also in the Great Oasis. Herodotus may have obtained his information concerning this temple at Thebes, while his accounts of the true Ammonium— the Oasis of Siwah—were almost certainly collected at Cyrene. Under these circumstances it seems not unlikely that he may have confused the two. It must be added that his account of the expedition sent by Cam- byses against the Ammonians (iii. 25, 26) is based upon the same misconception, as he represents the king as first detaching the troops for that purpose from Thebes : though the Ammonians really lay much nearer to Memphis, and nearest of all to the Lake Moeris. In this narrative, however, he clearly separates the Oasis, where the city of that name was situated, from the Ammonium. It was in the sandy desert between the two that the army perished. That Herodotus should have fallen into error as to the true geographical position of the Oasis of Ammon is not surprising. Even Ptolemy—as M. Parthey has pointed out—brought it down much too far south, so as to make it almost precisely equidistant from Memphis and from Thebes. (Parthey, Die Oase des Ammon, p. 148: and see his map.) NOTE 0, p. 280. THE ATLANTES OF HERODOTUS. Mr. Tristram suggests that the description of Mount Atlas as given by Herodotus may have arisen, in part at least, from con- fused accounts of Mount Zaghouan, the only conspicuous isolated mountain in the Regency of Tunis. (The Great Sahara, p. 77.) But Mount Zaghouan is situated far to the north of the line of oases which we are following, in the “inhabited district” of Herodotus NOTES 0, P. HERODOTUSZ AFRICA. 3 I I and among the agricultural Libyans. It is quite clear that, according to the conception of Herodotus, the belt of sand in which these oases were situated extended continuously in a straight line to the western ocean (iv. 181, 185), and had nothing to do with the caravan routes to Carthage. Such a chain of oases, with or without salt rocks and hills, might, in fact, be traced throughout the whole tract: but his information concerning it was .evidently of the vaguest description. Still the name of Atlas and the Atlantes certainly seems to indicate a connection—however confused and imperfectly understood—with the far-famed Mount Atlas, of which the name at least was so familiar to the Greeks. At the same time the name, whether Greek or Phoenician, was cer- tainly not of indigenous origin, and the question. naturally suggests itself, whether the description of this semi—fabulous mountain was not transferred by the Greek traders to a mountain of the interior, with which it had no real connection, on account of the name Atlantes (or some native form resembling it), which they had met with in that part of the country. It is certain at least that Herodotus did not himself conceive these Atlantes as being situated in the neighbourhood of the Ocean: as he distinctly tells us that the ridge or belt of sand was continued beyond them to the Pillars of Hercules, and the region outside Of them (OmjKel 8, div i7 dgbpiir] ,ae'xpt cHpakhe'an/ o'rghe'wv, Kai TO 35w 'rov're'wv, iv. 185), but that he is not able to give the names of the stations; thus implying clearly that he supposed there was room for more than one of them. NOTE P, P. 284.. THE LOTOPHAGI. Scylax extends the name of the Lotophagi to the inhabitants of the whole coast, between the two Syrtes, from the river Cinyps to the commencement of the Lesser Syrtis (§ 110, ed. Muller). But he describes the island, to which he gives the name of Brachion, though it is certainly the same with the Meninx of later writers, as particularly abundant in the fruit. Polybius, who visited the region in person, has left us a particular account of the lotus-tree and its fruit, which evidently excited great interest among the 312 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. Greeks, on account of its connection with the Homeric fable. (Polyb. xii. 2.) The tree in question (the Rhamnus Lotus, or Zizyphus Lotus of botanists) is still found in abundance on the island of Djerba, or Gerba (the Meninx of Strabo and Ptolemy), and its fruit is still used as an article of diet by the Arabs, though modern travellers are far from finding in it the attraction that operated so powerfully on the companions of Ulysses (Barth, Wanderungen in Nord-Afriha, p. 262; Guérin, Voyage Archéologigue dans la Régence de Tunis, vol. i. p. 206). They, however, speak in the highest terms of the beauty and fertility of the island, which is called by Barth “a terrestrial paradise,” a circumstance which may have contributed to its being selected by tradition as the abode of “the dreamy lotus-eaters.” NOTE Q, p. 286. THE RIVER CINYPS. The river Cinyps of Herodotus, which is mentioned also by Scylax and the later geographers, can he certainly identified with the little stream called Wadi Quaam (but known also by various other Arabic appellations), which flows into the sea a few miles to the east of Lebdeh (Leptis). The plain on both sides of it, extending from Lebdeh to Mesarata, though now uncultivated, is still remark- able for its natural fertility; forming a striking contrast to the barren tracts on each side of it. (Beechey’s Northern Africa, p. 51 ; Barth, Wanderungen, pp. 317—319.) In ancient times it possessed an almost proverbial celebrity for its productiveness, which pro- bably arose in great measure from this exceptional mention of it in Herodotus (See Mela, i. § 37, and Ovid, Ea: Ponto, ii. 7, 25. Cinyphiee segetis citius numerabis aristas). It was doubtless well known to the Greeks of Cyrene, from whom he derived his infor- mation; and at an earlier period the Spartan prince Dorieus had attempted to found a colony there, but was driven out by the Libyan tribes, supported by the Carthaginians, who naturally resisted this attempt to extend the chain of Greek colonies so much nearer to their own frontiers (Herodot. v. 42). Scylax, however, speaks of the existence of a town upon its banks (§ 109), though he says that it was no longer inhabited in his time. But within a few NOTE R. HERODOTUS: AFRICA. 3 I 3 miles of it arose the city of Leptis— called for distinction’s sake Leptis Magna—which became, first under the Carthaginians, and afterwards under the Romans, one of the most important cities of Northern Africa. NOTE n, p. 286. THE SYRTIS. It is singular that Herodotus, though well acquainted with the name and extent of the Syrtis, should say nothing of its physical peculiarities or the dangers of its navigation, which were well known to Scylax (§ 110). These last, though greatly exaggerated by ancient writers, are not without a foundation in fact; and arise from the extremely flat and low character of the coast, together with a number of sunken rocks, as well as the violence of the north winds, which have here an uninterrupted sweep across the broadest part of the Mediterranean. The tides, which are represented by ancient writers as one of the great sources of danger in both Syrtes, are in fact almost imperceptible in the Greater Syrtis, though they attain to greater importance in the Lesser Syrtis than at any other point in the Mediterranean, except Venice (Smyth’s Mediter- ranean, pp. 187, 188 ; Rennell, p. 651). In this respect, as in several others, the writers in question seem to have confounded the characteristics of the two gulfs. The extensive quicksands, which gave rise to the notion that the coast was “ neither land nor sea, but a kind of mixture of both,” have now in great measure disappeared; and along the whole of this line of coast Captain Beechey is of opinion that the land has been gaining on the sea, for that we find “ the ancient parts filled up, the lakes converted into marshes, and the quicksands to have become solid and firm ” (Beechey’s Northern Africa, p. 272). Hence it is by no means impossible that the peculiar characters of this region were more strongly developed in ancient times, and afforded somewhat more foundation for the fables that were built upon them. The silence of Herodotus cannot fairly be cited on the other side as proving that these pecularities did not exist in his time, or even that they were unknown to the Cyrenaeans in his day. It is quite unreasonable to commend Herodotus (as has been done by some modern writers) for his superior information and accuracy, 3I4 HISTORY OF ANCIENT GEOGRAPHY. CHAP. VIII. merely because he says nothing about the dangers of the Syrtis, of which he never mentions the name except incidentally. NOTE s, p. 287. THE LAKE TRITONIS. Major Rennell, in whose time the geography of this part of Africa was still very imperfectly known, was the first to suggest that the Lake Tritonis of Herodotus was in fact identical with the Lesser Syrtis of later writers, or rather comprised that and the inland lake of Lowdeah united (Geogr. of Herodot. p. 662): and this view is supported by Mr. Rawlinson, who speaks of the Lake Tritonis as “ an inner sea ” which stood to the Lesser Syrtis in the same relation as the Sea of Azof to the Euxine. (Rawlinson’s Herodotus, vol. iii. p. 154, note 1.) But I confess I cannot see any necessity for its adoption. The terms in which Herodotus speaks of the Lake Tritonis (1‘; Tpt'rwvis Mum) and the tribes that dwelt around it are certainly such as to imply primd facie that it was a lake or inland piece of water: he nowhere alludes to its saltness, but calls it “ a large lake ” e’s Mitt/1711 lueyoihrlv Tpt'rwvida, iV. 179) and represents it as the boundary between the nomad Libyans and the agricultural tribes. Even at the present day the salt lake known under the various names ‘of Chott el Fejij, Chott el Melah, and Sebkah Faraoun (which is termed by Shaw Shibkah el Lowdeah), is not less than 110 miles in length: and there can be no doubt that at an earlier period it was much more extensive and was united with various other salt lakes in the same region, so as to cover an area of nearly double that extent. (See the description of the recent French travellers, M. Guérin in the Voyage Archéotogique dans la Re'gence de Tunis, vol. i. pp. 247-250 ; and M. Charles Martins in the Revue des Deuce Mondes for July, 1864.) It is at present sepa- rated from the sea only by a low sandy isthmus not more than ten miles in width, and there is every reason to believe that this is nothing more than a bar of sand gradually thrown up by the action of the winds and tides. It is therefore not improbable that in the time of Herodotus, as well as in that of Scylax, it communicated with the sea by a narrow channel, or opening, which has gradually silted up. Thus far the views of Major Rennell may be admitted to be well- NOTE S. HERODOTUS 2 AFRICA. 3 IS founded and to display his usual Sagacity. But when he argues that because Herodotus describes Jason as driven by a storm “into the shoals of the Tritonian Lake ” before he saw the land, he must therefore have supposed it to be a gulf of the sea, not an inland lake, and that that gulf could be no other than the Lesser Syrtis (p. 663); he certainly seems to be requiring an unreasonable amount of accuracy from a writer who is relating a mere poetical legend, and applying it to a country which he never visited. Sup- posing the name of the Lesser Syrtis to be still unknown to fame, “ the shoals of the Tritonian Lake” would not be an unapt desig- nation of the shallows which were in fact situated close to its mouth. The mention by Herodotus (iv. 17 8) of “a large river,” called the Triton, flowing into the Tritonian Lake, is a difficulty which admits of no satisfactory solution. No such river exists at the present day, nor could there ever have been any considerable- perennial stream in that region of Africa. But Herodotus had evidently no idea of the real nature of the Tritonian Lake—a vast expanse of very shallow salt water, which was probably, even in his day, often dry in many places : he supposed it to be a lake like any other, and that a lake of such extent should have a large river as its feeder was but a natural assumption. The same idea was as usual retained by later geographers, who ought to have been better acquainted with this part of Africa: Pliny (v. 4, § 28) speaks of a vast lake receiving the river Triton, from which it derives its name. Mela gives a precisely similar account (i. 7, §36), and Ptolemy describes the river Triton as rising in the mountain of Vasaleton, and constituting three lakes, to one of which he gives the name of Tritonitis. The three lakes in question are probably only distinct names for three portions of the large expanse, which is sometimes united into one sheet of water, more often separated into three by dry intervals of sand covered with salt. (See the descriptions above cited.) Scylax, who wrote only about a century after Herodotus, has left us (§ 110, p. 88, ed. Miiller) a much more particular account of the lake Tritonis, as well as of the Lesser Syrtis, which he designates by- that name, and describes as 2000 stadia in circumference, and much more dangerous and difiicult of navigation than the other Syrtis. He then speaks of an island called Tritonis, which he places (apparently by a corruption of the text) in. the Syrtis, and a 316 HISTORY OF ANoIENT GEOGRAPHY. CHAP. vIII. river Triton. The lake (he adds) has a narrow mouth, in which there is an island, so that sometimes at low water there is no appearance of an entrance at all. The lake is of large extent, being about 1000 stadia in circumference—a statement much below the truth. Here it is not quite clear whether the river Triton is the same with the narrow channel communicating with the sea, or not, though this is the most probable explanation. Ptolemy also distinctly speaks of the outflow of the river Triton into the sea, which he places ten miles to the west of Tacape, the modern Cabes (Ptol. iv. 3, § 11) ; and there can be no doubt that he here means the same river, which he elsewhere mentions as having its rise in the interior and flowing into the lake (Ib. 3, § 19). Pliny and Mela add nothing whatever to our information. The question is an interesting one, because it appears probable from recent geological researches that a great part of the Northern Sahara was at no very remote period covered by an inland sea, communicating with the Mediterranean at the Lesser Syrtis, and that it has been gradually elevated to its present level. Could we therefore prove that this communication was still open to some extent in the time of Herodotus, we should be able to trace the last stage of this geological change by historical evidence. Unfor- tunately the testimony of Herodotus is very vague, and apparently derived from imperfect information; while that of Scylax, which is more complete and definite, is in some degree marred by a. corruption of the text, which seems to arise from an accidental omission in our manuscripts. (See C. Miiller ad loo.) NOTE T, p. 288. THE ISLAND OF CYRAUNIS. N iebuhr (Geogr. of Herodotus, p. 20), as already mentioned, supposes the island Cyraunis of Herodotus to be the Same with the Cerne of later authors, and there appears no doubt that Diodorus confounded the two: but this is certainly not consistent with the text of our author, who, after describing the peculiarities of the island of Cyraunis, which he places opposite to the Gyzantes, and therefore certainly in the Mediterranean, adds (after a short digres- _ sion) this other tale told by the Carthaginians concerning a place in Libya, outside the Columns of Hercules. It is this introduction NQ'I‘E V. HERODOTUS I AFRICA. 3 I 7 of the one narrative, just after the other, that has apparently led to the confusion of the two, but Herodotus, in fact, carefully dis- tinguishes them. This Cyraunis is clearly, as Rennell has pointed out (Geogr. of Herodot. p. 638), the same as the Cercina or Cer- cinna of Strabo and later authors, which agrees both in position and extent with the description of Herodotus. The account given by Scylax (§ 112) of the mode in which trade was carried on by the Carthaginian merchants at Oerne has undoubtedly some resemblance to that related by Herodotus con- cerning a place (not named) on the Atlantic coast of Libya, but the similarity may have been easily produced by the actual occurrence of similar conditions. Moreover, the two accounts differ in one important particular, that while Herodotus mentions gold as the principal, or sole, subject of barter on the part of the natives, it is not even included by Scylax among those obtained at Cerne. NOTE v, p. 296. THE CIRCUMNAVIGATION OF AFRICA; OPINIONS OF MODERN WRITERS. The narrative of Herodotus has been believed, and its authen- ticity supported, among modern writers, by Major Rennell (Geogr. of Herodot. pp. 672-714), by Larcher (Notes on Herodot. vol. ii. pp. 26-30), by Heeren (Researches, vol. ii. pp. 76-81, Engl. transl.), and Prof. Rawlinson (Herodotus, vol. iii. p. 45). Mr. Grote in his history (vol. iii. pp. 377-385) accepted the narrative as a historical fact, but I have reason to believe that he subsequently changed his opinion. On the other hand it is rejected by Gossellin (Ge’ogr. des An- ciens, tom. i. pp. 204-216), Mannert (Geogr. der Gr. u. Romer, vol. i. pp. 19-26), Mr. Cooley in his English translation of Larcher’s notes, pp. 30-32), by Dr. Vincent (Commerce and Navigation of the Ancients in the Indian Ocean, vol. ii. pp. 195-204), and by Sir G. Lewis (Hist. of Ancient Astronomy, pp. 508-515). Ukert justly observes (vol. i. p. 48) that the question is one upon which opinions will always continue to be divided. The argu- ments on both sides may well be considered as exhausted: and the absence of all details precludes the possibility of adding to them by farther investigation. (318) CHAPTER IX. VOYAGE OF HANNO. § 1. IT is unfortunate that, while Herodotus has dwelt at some length on the unsuccessful voyage of Sataspes, which added very little to the information possessed concerning Africa, he seems to have had no knowledge of the far more important and interesting expedition of Hanno in the same direction.1 The details of this voyage, which have been preserved to us by a singular chance, while almost all other records of Cartha- ginian and Phoenician enterprise have perished, are well worthy of a careful consideration, and will find an appropriate place here, as there is every probability that the voyage itself, though not mentioned by Herodotus, must have taken place before the time of that historian. The narrative that has been transmitted to us purports to ' be that of the commander of the expedition himself, inscribed on a tablet dedicated by him on his return in the temple of Cronos or Saturn, a name undoubtedly meant to designate the Phoenician deity Moloch. Such a dedication has nothing in itself improbable, and the brevity and the somewhat official style of the narrative itself corresponds with its supposed origin. The document in its present form being in Greek, must probably be regarded as a translation of the original; for though we have numerous instances in later times of bilingual inscriptions in Punic and Greek,2 it is scarcely 1 It is indeed not improbable that to the marvellous character of which he the accounts on which the Cartha- could hardly have failed to advert. ginians based their confident belief 2 Hannibal, indeed, as we know, that Africa could be circumnavigated, before leaving Italy, set up a bilingual were connected with this voyage; but inscription in the temple of the Laci- it is impossible to believe that Hero- nian Juno, recording his exploits, in (lotus had any knowledge of its details, Panic and Greek: but there was an CRAP. IX. VOYAGE OF HANNO. 319 likely that such a practice would be resorted to at so early a period.3 The primary object of the expedition—as We are told at the outset—was not so much discovery, as colonization; hence Hanno sailed with a mixed multitude of men and women, amounting, it is said, to not less than 30,000 (a number in all probability exaggerated), and a fleet of sixty ships, all pente- contersf“ After sailing through the Straits of Hercules and continuing their voyage for two days, they established their first colony, to which they gave the name of Thymiaterium; it commanded a great plain. Thence standing on towards the west they came to the headland of Soloeis, the promontory of Libya, where they established a temple to Poseidon. From thence they turned to the east, and after half a day’s voyage came to a large lake or marsh near the sea, in which many elephants were feeding. After passing this lake and pro- ceeding another day’s voyage, they founded five towns by the sea-coast, to which they gave the names of Caricon Teichos, Gytta, Acra, Melitta and Arambis. Setting out again from thence, they came to a large river called Lixus, flowing from the interior of Libya. Its banks were inhabited by nomad Libyan tribes, who had flocks with them, and entered into friendly intercourse with the navigators. But the interior of the country, according to the statement of these Lixitae, was occupied by wild and inhospitable tribes of Ethiopians, in- habiting a region abounding in wild beasts, and intersected by high *mountains, from which the Lixus took its rise. These Ethiopians were Troglodytes, dwelling in caves and obvious reason for this, as the record was left in a country where Greek was spoken. No such reason could apply to an inscription set up atCarthage. 3 Note A, p. 332. 4 It is not to be supposed that the colonists were all conveyed in the pen- teconters, which would have been much too small for the purpose. They were doubtless embarked in merchant vessels (6Mccl8es), the greater part of which would be left behind as the successive colonies were founded. The pente- conters would serve for escort, and to explore the coast in advance: and from Cerne onwards they might probably have gone on alone. 320 HISTORY OF ANCIENT GEOGRAPHY. CHAP. IX. holes in the mountains, of strange aspect, and swifter of foot than horses.5 Hanno remained some time with the Lixitm, and took with him interpreters from among them to accompany him on his voyage. Proceeding onwards they coasted along the desert for two days, holding a southerly course; and then turning eastward for one day’s voyage, they came to a small island (only five stadia in circumference) in the inmost recess of a gulf, where they established a settlement, and called its name Cerne. Reckoning up the length of their navigation they came to the conclusion that Cerne was opposite to Carthage ;6 for the distance from the Columns to Cerne was the same as that from Carthage to the Columns.7 § 2. From Cerne the navigators made two separate voyages to the south. In the first of these they came (after what interval is not stated) to a lake, containing three islands, com- municating with the sea by a large river, and of such extent that it took them a day’s voyage to penetrate to its inner extremity, which was overhung by lofty mountains, inhabited by wild men, clothed in the skins of beasts, who prevented the voyagers from landing by throwing stones at them. Sailing from hence they came to another broad and large river full of crocodiles and hippopotami. Here (for what reason we are not told) they turned about and returned to Cerne.8 §3. Setting out a second time from thence they held a direct course towards the south for twelve days, at the end of which time they arrived at a headland formed by high moun- tains, covered with thick woods of trees of many kinds. After doubling this headland in two days’ voyage, they came to a vast gap or opening of the sea, on the other side of which was ‘ Periplus, §§ 2—7, ed. Miiller. “able analysis” of the voyage of 6 Note B, p. 333. Hanno; and is at a loss to understand 1 pewjpL § 8 his having put back “without any 8 lb. §§ 9—10. It is singular that Mr. assignable motive. ’ What his motive Major (Prince Henry the Navigator, p. may have been we know not, but the 92) should treat this return to Cerne as fact is distinctly stated by himself. if it were a part of M. de St. Martin's CHAP. IX. VOYAGE OF HANNO. 321 a plain; from whence many fires were seen at night. After laying in a stock of water, they continued to coast along the land for five days, till they came to a large bay, called by their interpreters the Western Horn. In this was an island, on which they landed, but found no signs of inhabitants, seeing nothing but the forest in the day-time; but in the night many fires were seen to be burning, accompanied with the sound of musical instruments, flutes and drums and cymbals. The Carthaginians were seized with a panic terror and immediately quitted the island.9 Sailing from thence in all haste they passed along a district which seemed all ‘in a blaze of fire; streams of fire as from a volcano pouring down from thence into the sea. Terrified at this appearance they hastened on, and came in four days to another place where the land was again all blazing with fire; in the midst of which was one fire that rose much higher than any of the rest, and appeared to touch the stars. By daylight it was seen that this was a very lofty mountain, which was called the Chariot of the Gods (Theon Ochema). Three days’ farther navigation (still passing by streams of fire) brought them to a gulf called the Southern Horn. Here there was an island, containing a lake, with another island in it, which was full of wild men and Women,'with hairy bodies, called by the interpreters Gorillas. The Carthaginians were unable to catch any of the men, but they caught three of the women, Whom they killed and brought their skins back with them to Carthage. This was the farthest limit of their voyage, as they were com- pelled by want of provisions to return.1 § 4. Such is in substance the brief narrative of this remark- able voyage, which in many respects stands alone among the records of ancient geography. Notwithstanding the apparently marvellous character of some of the incidents recorded, it bears the unquestionable impress of being an authentic record of a real voyage; and even the geographical data will be found, on ~ 9 Peripl. §§ 11-14. 1 Id. §§ 14—18. VOL. I. Y 322 HISTORY oE ANCIENT GEOGRAPHY. can». 1X. a careful examination, to be for the most part easily reconciled with existing facts. Their simplicity and clearness, when considered alone, will ,indeed be found to present a striking contrast to the confusion in which they are involved, in the hands of later geographers. There has indeed been great discrepancy of opinion among modern writers with regard to the distance actually traversed, and the farthest point to which the navigators attained. Gossellin refused to believe that they advanced farther than Cape Noun (in 28° 40' N. lat.) ;2 but this view, though adopted by Walckenaer,3 may be safely rejected as utterly untenable. It does not indeed satisfactorily explain or agree with any of - the principal facts recorded, and is in fact based solely on the assumption that the ancients could not make voyages of any considerable length.4 Major Rennell on the contrary, in a very valuable and elaborate examination of the question,5 came to the conclusion that the Southern Horn—the farthest point actually attained by the expedition—was identical with 'Sherboro’ Sound, just beyond Sierra Leone (in N. lat. 7° 45'), and startling as it may at first appear that the ‘voyagers should have penetrated so far to the south, the arguments in favour of this view may be regarded as almost, if not quite, con- clusive. It has been adopted both by the most recent editor of the Periplus of Hanno C. Muller), and by M. de St. Martin in-his elaborate and valuable work on the ancient geography of Africa. Both of these writers have supplied important corrections and additions, arising in part from our improved acquaintance with the west coast of Africa, since the time of Major Rennell, but the merit of having first established the true view of the question undoubtedly rests with the great English hydrographer.6 2 Recherches sur la Geographic des \ 5 Geography of Herodotus, pp. 719- Anciens, vol i. pp. 70-106. \ 745,4to ed. 3 Iter-herches sur la Geographic de l 6 See Rennell’s Geography of Hero- Z'Afrique, p.362. dotus, §26; St. Martin, Le Nord de 4 Note C, p. 334. ' ' Z’Afrz'qae dans l’Antiquite‘, pp.330-400; CHAP. IX. VOYAGE OF HANNO. 323 § 5. The main point upon which the geography of the whole voyage may be considered to rest is that of the position of Cerne, the place from which the Carthaginian commander set out on his two separate exploring voyages to the south; and where he founded a colony, which undoubtedly continued to exist for a long period of time, and carried on a considerable trade with the nations of the interior.7 Now the data for de- termining the position of Cerne are given with unusual pre- cision. It was a small island, situated in the bight of a deep bay; and it was, according to the computation of the Cartha- ginian navigators—derived from their sea-rechoning—the same distance from the Straits or Pillars of Hercules on the one side that Carthage was on the other.8 Major Rennell (concurring on this point with D’Anville and several other writers) identified the island of Cerne with that of Arguin, a short distance to the south of Cape Blanco,9 which became at one time a considerable trading station in the hands of the Portuguese;1 and the same view was adopted by Ukert and Movers. But in the first place the size and position of the island of Arguin corresponded but imperfectly with the description of Cerne, and what was a more important objection its distance from the entrance of the Straits greatly exceeded that of Carthage. Major Rennell indeed sought to avoid this difficulty by pointing out that in these latitudes there is a constant southerly current, setting along the coast of Africa, which would undoubtedly have carried the Carthaginian ships much farther to the south than they were aware of, reckoning and the commentary of Dr. C. Miiller in his Geographi Grwci Minores, vol. i. . -14. 7 Note D, p.2334. 8 Peripl. § 8. 9 The first to make this identification was a Portuguese pilot, cited by Ra- music in his commentary on the voyage of Hanno, referred to in Note A. . 1 Ca da Mosto (Viaggio, p. 99, in Ramusio, tom. _ i.; Major’s Prince Henry, p. 254). The existence of this trade, in many respects resembling that of Cerne in ancient times, was one of the reasons which led to the identifi- cation of the latter with Arguin; but there is no natural communication with the interior to determine it; the trade was in both cases merely the result of the establishment of a factory on the coast. Arguin is now abandoned, just as Cerne was, when the Carthaginian commerce declined, and is a poor and desolate island. v 2 324- HISTORY OF ANCIENT GEOGRAPHY. CHAP. IX. only by their rate of sailing. But the difference, amounting to not less than 320 G. miles, was one that could hardly be thus accounted for; and it was principally this difficulty that led several modern writers to place Cerne ten degrees farther north in the neighbourhood of Aghadir or Santa Cruz, though no such island is now to be found there, and this position is as much too far to the north, as Arguin is to the south. We are indebted to Dr. C. Muller for first pointing out the ex- istence, at a point intermediate between the two, of a small island, still called on the French charts Herne, in the bight of a deep bay, at the mouth of the Rio do Curo. The situation of this island thus exactly agrees with the description of Cerne, while its position on the coast, more than 200 G. miles to the north of Cape Blanco (in lat. 23° 50') reduces the allowance to be made for the current within a very probable amount. It will be found also that the account of the subsequent pro- ceedings of the voyage accords much better with the position thus assigned to Cerne than with that of Arguin; and on the whole it may fairly be said that the solution thus proposed of this long disputed question may be considered as established with reasonable certainty.2 § 6. Starting then from the position of Cerne, as a fixed point of departure, it will be found that there is little difficulty in determining all the more important points visited farther to the south --the physical peculiarities of the localities being such as to render their identification almost certain, even if there were more difficulty than actually exists in reconciling them with the distances stated. It must be observed that from Cerne onwards the voyage appears to have been one of discovery merely, no trace being found of any attempt at colonization to the south of that island. 2 The suggestion thus put forth by but we are assured that it is used by M. Miiller (Prolegomena, p. xxvi) is the Moors of the continent. The Rio adopted by M. Vivien de St. Martin , do Ouro was already known to the (Ge’ogmphie de l’Afrz'que, pp. 382-3), Portuguese navigators of the 15th and confirmed with additional argu- century; but there has been no settle- ments. The name of Herné appears ment therein modern times. for the first time on the French charts, i ' Omar. IX. VOYAGE OF HANNO. 325 In regard to the first expedition we have no statement of distances, or of the time employed, but we learn that its limit was the mouth of a large and broad river, full of crocodiles and hippopotami. This description leaves no doubt that the river attained could be no other than the Senegal, called by later Greek and Latin writers the Bambotus, and described by Polybius in almost the same terms as by Hanno.3 There is no river of any considerable size to the north of this, after leaving the confines of Mauritania, and certainly none in which either crocodiles or hippopotami could ever have lived.‘* §7. In like manner the first point indicated in the second voyage can admit of no doubt. The lofty headland covered with large trees of various kinds, can be no other than Cape Verde, the only point of elevated land that projects into the sea along the whole of this line of coast, and which derives its modern name from the rich verdure of the woods with which it is covered.5 The distance of twelve days’ voyage agrees much better with the supposition that Cerne was at the Rio do Ouro than at Arguin. Beyond the promontory of Cape Verde, the coast again becomes flat and low, and the mouth of the Gambia forms a broad estuary with flat shores, which is evidently the “ chasm” or wide opening of the sea next men- tioned by Hanno. It was here that they first saw the fires, which afterwards figure so prominently in their narrative. The distance from thence to the deep gulf called the Western Horn6 leads us to identify the latter with the bay or gulf of 3 “ Flumen Bambotum, crocodilis et hippopotamis refertum.” Polyb. ap. Plin. V. i. §10. The statements of Polybius in this passage will be exa- mined in a future chapter. 4 Note E, p. 335. 5 See Ga da Mosto (in Ramusio, Viaggz', tom. i. p. 105 1)), who calls it “molto bel capo, ed alto di terreno.” The expression of the Carthaginian navigator of “ high mountains " is un- doubtedly an exaggeration, but it was natural that they should over-estimate the importance of the only high land they had seen for hundreds of miles. 6 It is a striking instance of the wu- fusion into which the statements of the Periplus were thrown by later writers, that they transformed the deep gulfs, or inlets of the sea, described by Hanno under the name of the Western and Southern Horn into promontorz'es, and applied the names accordingly. The Hesperi Gornu of Ptolemy and Pliny is no other than Cape Verde, which is also called Hesperium Promontorium. 326 HISTORY or ANCIENT GEOGRAPHY. can». Ix. ‘Bissago, in which there are many small islands; but‘ the identification of these inlets must remain somewhat uncertain; the coast of Africa in this part being indented by many deep bays. It is otherwise with the Theon Ochema, which may be safely assumed to be the mountain called by the Portuguese Sagres (a name afterwards corrupted into Sangaree) but which figures in some modern maps as Mount Souzos, in others is called Mount Kakulima. It. is described as a lofty conical mountain, forming a conspicuous landmark in the midst of a flat coast,7 thus according well with the appearance of the mountain as described in the Periplus; and the occurrence of a striking isolated peak in such a position is too remarkable a coincidence to leave any doubt upon the subject. Three days’ navigation from hence brought them to the limit of their voyage, the Southern Horn—a deep gulf or inlet of the sea, which may be probably identified with Sherboro’ Sound, the next opening beyond that of Sierra Leone, more familiar to us at the present day from the establishment of an English colony. The distances given along this part of the coast from Cerne to the Southern Horn agree remarkably well with the real positions: but the natural characteristics of certain important points, especially the Cape Verde and the Mount Sagres, as well as the river Senegal, afford much the most satisfactory means of identification, and leave no reasonable doubt of the voyagers having really advanced as far as the point indicated. It may be added that this was the farthest point reached by any ancient navigators. Even in the days of Ptolemy the Theon Ochema was still the limit of his knowledge of the west coast of Africa. §8. The circumstancmhat seems to have tended most to discredit the narrative of Hanno in the eyes of subsequent writers was the marvellous account that he gave of the “ streams 7 Rennell, p. 734; Roussin, Mémoire stance that the companions of Pedro sur la Navigation aux Cotes Occiden- dc Cintra, who first discovered the .tales dc Z’Afrique, p. 95 (cited by St. mountain in 1462, considered it to be Martin, p. 394). Its striking character the highest mountain they had ever is sufficiently shown by the circum- - - seen (Rennell,-l. c.).- . - CHAP. IX. VOYAGE 0F HANNO. 327 of fire” and “pillars of fire ” that he saw after passing Cape Verde.8 Nor can it be doubted that the terror which these appearances excited in the minds of the navigators, and which is very naively confessed in the journal, led them to consider- able exaggeration of what they really saw. But the pheno- menon is readily explained by the practice, generally adopted by the negroes in this part of Africa, of setting fire to the long dry grass in the autumn, by which conflagrations are kindled of such an extent as readily to give rise, in the excited ima- gination of the voyagers, to the description that they have left us.9 § 9. Another statement that was treated as fabulous in ancient times was the account of the wild men and women covered with hair, that they found in the island of the Southern Horn. Yet the fact that they brought back the skins of two of them to Carthage might suffice to show that the assertion was not without some foundation in fact. Pliny indeed adds that the skins in question were dedicated by Hanno in the temple of Juno at Carthage, and continued to be visible there till the destruction of the city.1 There can be no difficulty in supposing these “ wild men and women” to have been really large apes of the family of the Chimpanzee or Pongo, several species of which are, in fact, found wild in \Vestern Africa, and 8 These statements were (as was so often the case) distorted by the exag- gerations of subsequent writers. Thus Mela says (iii. 9, §94) “ Ultra hunc sinum mons altns, ut Graaci vocant Theon, perpetuis ignibus flagrat ,' ” and Pliny has the same assertion (H. N. vi. 30, § 197), “Imminens mari mons ex- celsus aeternis ardet ignibus, Theon Ochema dict-us Grsecis.” 9 See the passages quoted from modern travellers by Major Rennell (p. 720), and by C. Miiller in his notes to the Periplus (p. 12). _ This explanation was long ago sug- gested by Ramusio in his Commentary on the Periplus ( Viaggi, tom. i. p. 113._b.) from the information of a Portuguese pilot, notwithstanding which, various other suggestions, one more absurd than another, have been put forward by modern writers. The word that} was undoubtedly most commonly applied to a stream of lava from a volcano, and hence it seems to have been generally assumed that volcanic appearances were those indicated (see the passages cited in preceding note). But, independent of the extent of the phenomena de- scribed, which in itself would exclude such a supposition, there are in fact no appearances of recent volcanic action on this part of the coast of Africa. ' 1 Plin. H. N. vi. 31, § 200. 328 HISTORY or ANCIENT GEOGRAPHY. CHAP.1X_ some of them, as is now well known, attain to a stature fully equal to that of-man.-2 . § 10. It is curious and instructive, after seeing how well the leading geographical facts related by Hanno accord with our present knowledge of the regions in question, to observe how confused and uncertain were the notions concerning his dis- coveries entertained by later writers in ancient times. Pliny even speaks of him in one passage as having circumnavigated Africa from Gades to the borders of Arabia 13 In another place he stigmatizes him as the original author of many fables, repeated after him both by Greek and Roman writers, of which he places in the front rank the statement of his having founded many cities, no trace of which was visible in the days of Pliny! That author here speaks of his “ Commentarii,” but it appears certain that he had not seen them himself :‘-1 and his account of the western coast of Africa, though containing some facts undoubtedly derived from Hanno, is, for the most part, taken from other authorities. Mela, on the other hand, correctly describes Hanno as having sailed round a great part of the continent, and been compelled to return, not from any diffi- culties of navigation, but merely from want of provisions. He adds also many particulars which are certainly taken, either 2 The species to which modern natu-r this subject presents a curious contrast ralists have appropriated the name of to the exaggerations of later writers. Gorilla (derived originally from this narrative of Hanno) does not seem to be now found north of the equator, but large anthropoid apes are still found in the forests of Senegal, and near Sierra Leone; and it is probable that these were more common in ancient times, and spread over a wider area. It is reported by modern travellers, in accordance with the experience of the Carthaginian navigators, that the males are never taken alive. Even the females (as the Carthaginians found) bit and fought with such violence that , they were forced to kill them in self- ' defence. The accuracy of Hanno’s report on Thus Mela, though directly citing Hanno as his authority, tells us: “ Grandis littoris flexus grandem in- sulam includit, in qua tantum feminas esse narrant, toto corpora hirsutas et sine coitu marium sua sponte fecundas !” (iii. 9, § 93). 3 Plin. H. N. ii. 67, § 169. “Et Hanno Carthaginis potentia florente circum- vectus a Gadz'bus ad finem Arabiaa navi- gationem eam prodidit scripto.” 4 This is apparent even from his own expression: “ Faere et Hannonis Car- thaginiensium ducis commentarii Pu, nicis rebus fiorentissimis explorare ambitum Africae jussi.” (v. 1, § 8.) OHAP. IX. VOYAGE OF HANNO. 329 directly or indirectly, from the narrative of Hanno, but he jumbles them together without any regard to geographical order, and thus involves them in inextricable confusion.5 Arrian again, who evidently quotes Hanno only at second hand, unaccountably describes him as having sailed (after issuing through the Straits into the Ocean) for thirty-five days towards the rising sun, and then turning to the south, where he met with great difiiculties from want of water, as well as the burning heat, and streams of fire flowing into the sea, which compelled him to return.6 We should, indeed, have known very little about the voyage of Hanno, had not a fortunate accident preserved to us theoriginal narrative. § 11. It is singular that, while the geographical statements preserved by Hanno from,.-the remoter regions of Western Africa are thus found to be easily reconciled with the truth, it is much more dificult to arrange with any certainty the details of the earlier part of the voyage, from the Straits of Hercules to the Island of Cerne. Fortunately these are of comparatively little importance. The promontory of Soloeis is undoubtedly Cape Gantin, which the ancient navigators seem to have re- garded as occupying a much more important position than it really does : 7 and as the extreme western point of Africa.8 Hence, probablyfit was selected by Hanno as the site of a temple to Neptune. But beyond this the geographical data 5 Mela, iii. 9. i I The Cape Soloeis of Hanno and 6 Arrian, Indica, c. 43. 7 The same thing was the case in the early voyages of the Portuguese in these parts with regard to Cape Noun and Cape Bojador. Ga da Mosto, how- ever, a little later, speaks of Cape Cantin as the most considerable head- land along this coast. Its abrupt ele- vation, “rising precipitously 211 ft. above the sea ” (see a paper by Lieut. . Arlett in the Geographical Journal, vol. vi. p. 308), coincides with the de- scription of it in Scylax (9; bwéxel ,ud- Mara. és 'rbx/ 1r611'r011,§ 112), and explains the choice of so commanding a site for a temple to Neptune. l Scylax certainly corresponds to the E Solis Mons and Solis Promontorium of I later geographers: the Soloentium of Ptolemy, on the contrary, is situated much farther to the south, and has no connection with the headland in question. 3 It is hardly necessary to point out the gross error involved in this as- sumption. Cape Cantin is really situated in 9° 17' W. longitude, while Cape Verde, the real westernmost point of Africa, is in W. long. 17° 32', or more than eight degrees of longitude farther west. 33o HISTORY or ANCIENT GEOGRAPHY.‘ can». IX. are either imperfect or erroneous. It is obvious that the Car- thaginians would never ,have founded five towns in immediate proximity to one another, but no mention is made of the intervals between them, or of the distance from thence to the river Lixus. The latter is called “ a large river,” which would lead us to identify it with the Wady Draa, much the largest - river in this part of Africa, which descends from the southern slopes of the Atlas chain and falls into the Atlantic a few miles south of Cape Noun. But the statement which follows, that the voyage from thence continued for two days by the side of the desert, is certainly erroneous ; the distance from the mouth of the Draa to the gulf of Rio do Ouro being not less than 420 G. miles, throughout the whole of which space the voyagers would undoubtedly be coasting the barren sandy shore of the desert, without a break or opening of any kind. On the other hand, if we read twelve days for two—the most natural sug- gestion, and that adopted by the latest editor of the Periplus—- the number appears greater than necessary, and does not correspond with the supposed relation to Carthage. But how- ever this difficulty may be solved, the importance attached to the river Lixus, and to its communications with the interior, certainly raise a presumption in favour of its being the same stream which is now called the Draa. The river Lixus of later authors, at the mouth of which there was ,a city of the same name, was certainly distinct from the one here mentioned, being situated only at a short distance to the south of Cape Spartel.9 ‘ § 12. N o mention is found in any subsequent writers of the colonies founded by the Carthaginians to the south of Cape Soloeis; they probably never rose to any importance, and as the power of Carthage declined, the trade with the western coast of Africa seems to have been gradually given up, and these settlements would naturally be abandoned. Even the 9 The site of the Roman town of Lixus undoubtedly corresponded with that of the modern Al Araish, or Laraiche, at the mouth of a small river called the Wadi a1 Khos. ‘llu E? 5i s~ V5 5; ii |1|".|‘ \ ‘A. 3. ‘ \ . ksl‘lzgxnfi- ‘ ‘ mu‘, 5,; _ H1‘ ‘PM 210 z k w M if‘ 7 e’. a; I y 1 _ _g;@ §gb 30 225; .'.,__:-73' MAP I to illustrate ’ THE VOYAGE .0F c mum; ' 20 M _ Q _...*_,~.~_~k~_-—~—~~—-—"' 20 V5 t s 1' ‘‘ Pklai’i ., GL- _ Q. )- . " “w .whw: ‘ ¢\‘ 4-_- um \ F10 _ Siérrdleone South‘ ~v "=* -__ herborol¢fi$gl I gzFET-‘fg smm'a; swim}! London. JohnMmray. CHAP. IX. VOYAGE OF HANNO. I trade at Cerne, which (as we have already seen) was still active in the days of Scylax, had ceased to exist before the fall of Carthage: and the very site of the island was a subject of doubt with later geographers.1 How easily all trace might be lost of such a site, as soon as it ceased to be resorted to as a trading station, is shown by the parallel case of Arguin, which was for a considerable time under the Portuguese a commercial station of no little importance; but is now a barren and deso- late island, inhabited only by a few Arab fishermen.2 1 Eratosthenes, indeed, still admitted the existence of the island, for which he is undeservedly censured by Strabo (i. 3, p. 47), and was probably ac- quainted with its true position. It is net unlikely that the other fabulous tales which he reported concerning the west coast of Africa (‘Iranian-ewes 5% ml wepl 76311 55m orr'nAcTW ‘Hparmeiwv woAAoTs pvfiu’aoeo'r, Képv'nv 76 11650011 rcal Nord de l’Afrz'que, pp. 356-380) and the tiAAous 'ro1rovs ouopo'nQou robs ,unSapofi commentary of Dr. C. Muller in his z/w/l Semz/vnévovs: Strabo, l. 0.) were ' edition of the Periplus. taken in like manner from the voyage of Hanno, with which Strabo seems to have been wholly unacquainted. 2 See the description of it by Capt. Grover in the Journal of the Geogr. Soc. vol. xvi. pp. 165—167. For the determination of the other details of this first part of the voyage, the reader may consult St. Martin (Le 332 HISTORY OF ANCIENT GEOGRAPHY. GHAP. IX. NOTE A, p. 319. EDITIONS OF THE PERIPLUS. THE narrative of Hanno was certainly extant in Greek at an early period. It is cited in the work ascribed to Aristotl on Marvellous Narratives (§ 37), which belongs to the third century B.C.; as well as by Mela, Pliny, and many later writers; and Pliny expressly speaks of it as the source from whence many Greek and Roman writers had derived their information, including, as he considered, many fables (Plin. H. N. v. 1, § 8). The authenticity of the work now extant under the name of the Periplus of Hanno, may be fairly considered as unquestionable. Though assailed by Dodwell, in a dissertation (De vero Peripli, qui Hannonis nomine circumfertur, tempore) inserted in Hudson’s Geographi Minores, vol. i. it was successfully defended by Falconer, in the preface to his edition (1797), and is admitted by all the later editors without a doubt. Indeed the internal evidence, when fairly examined, is conclusive upon that point. But there has been great diversity of opinion as to the period to which the expedition is to be referred; on this point the narrative itself gives us no information ; and the name of Hanno was so common at Carthage as to afford us very little clue to his identity (see Smith’s Diet. of Biogr. art. HANNO). But it has been generally agreed among recent writers that the most probable conclusion is, that he was either the father or the son of the Hamilcar who led the great Cartha- ginian expedition to Sicily in B.0. 480. In the former case the Periplus may be probably assigned to a date about B.0. 520 ; in the latter it must be brought down about 50 years later. This last view is that adopted by C. Muller in his edition of the Periplus (Geographi Grceci Minores, vol. i. pp. xxi—xxiv), where the whole subject is fully discussed; but, as between him and his grandfather, the choice must be admitted to be little more than conjectural. M. Vivien de St. Martin, however, prefers the date of B.0. 570, which had been previously adopted by Bougainville (Mémoires de Z’Acade'mie ales Inscriptions, tom. xxviii. p. 287). The Periplus was first published at Basle in 1533 (as an appendix to the Periplus of Arrian), from a MS. in the Heidelberg library—- the only one in which it is found: it has been since repeatedly NorE B. VOYAGE OF HANNO. 333 republished in a separate form, with copious commentaries and illustrations. Of these separate editions those by Falconer, 8vo, 1797, and by Kluge, 8V0, Lips. 1829, are the most valuable. The treatise is also included in the editions of the Geographi Grcecz' Minores by Hudson, Gail, and C. Miiller. The valuable and elabo- rate commentary of the latest editor may be considered as in great measure superseding all others. Besides all these editions, it has been made the subject of elaborate investigations by Gossellin, Bou— ' gainville, Major Rennell, Heeren, Ukert, M. Vivien de St. Martin, and other‘ geographical writers. Indeed there are few ancient writings that have been the subject of more copious commentary in proportion to its very limited extent. The earliest of these com- mentaries, inserted by Ramusio in his collection of voyages, is curious and interesting as being derived from Portuguese sources, who were in modern times the earliest explorers of these coasts. That by the Spanish writer Campomanes (El Periplo dc Harmon ilustrado, appended to his Antiguedad Mam'tima dc Cartago, 4to, Madrid, 17 56) is, on the contrary, utterly worthless. NOTE B, p. 320. POSITION OF CERNE. It is rather difficult to understand the exact meaning of the expression here used, mm’ @361‘) Keiadai Kapxiydcivos. It is fortun- ately explained in part by the subsequent addition that the voyage to Cerne from the Straits was of the same length as that from Carthage to the Straits: but still the sense of the phrase remains obscure. The first impression would be that it is used in the same sense as we should say, it was on the same meridian with Carthage: just as Herodotus describes the mouths of the Nile and the Danube as lying opposite to each other (ii. 33, 34). And it cer- tainly appears that the passage was so understood by later writers, who interpreted it as “ ea; adverse Carthaginis.” Corn. Nep. ap. Plin. vi. 31, § 199. But these geographers, from Eratosthenes onwards, all conceived the west coast of Africa as trending away rapidly towards the east, immediately after passing Cape Soloeis, so that the island of Cerne would thus be brought approximately (though not of course exactly) to the same meridian with Carthage. Hanno, however, gives no countenance to such a view, which seems 334 HISTORY, OF ANCIENT GEOGRAPHY. CRAP. IX. ( to have been adopted solely on theoretical grounds. His state- ments of the bearin gs of his course, wherever given, are generally correct, and the inference certainly is that the general direction of their voyage lay, as it really must have done, towards the south. The distances from the Straits to Cerne and to Carthage, being supposed equal to one another, would form the two sides of an isosceles triangle, the base of which would be formed by a straight line joining the two: but certainly if the phrase means nothing more than this, it appears a strange one to have employed. NOTE 0, p. 322. VIEWS OF HEEREN. The views of Gossellin have been satisfactorily disproved by Major Rennell (see particularly p. 438), and are justly rejected by Heeren (African Nations, vol. i. p. 492), who admits that the expedi- tion advanced as far as the Gambia. But he appears to have over- , looked the important identifications of Cape Verde and the moun--- tain called the Theon Ochema (both of them already pointed out ' by Major Rennell), and dwells mainly on the distances. It appears to me far safer, in such a case as this, to rely mainly upon the description of marked natural features, where these are really prominent and striking, than upon statements of distances, which are liable to so many causes of error. At the same time the dis- tances given, on the voyage beyond Cerne, agree very well with the truth, as soon as the position of Cerne itself is rectified in accordance with the view stated in the text. NOTE 1), p.323. NOTICE or‘ OERNE ‘IN SOYLAX. The existence of this trade is distinctly attested by Scylax (§ 112), who appears to have derived it from some authentic source, though he was imperfectly acquainted with the geographical posi- tion of Cerne, as he places it only twelve days’ voyage from the Straits, and seven from the promontory of Soloeis. - He was evidently unacquainted with the voyage of Hanno, as he treats Cerne as the farthest point to which navigation was possible: an NOTE E. VOYAGE or HANNO. ' 335 idea that was very probably intentionally circulated by the Car- thaginian traders. Nor does he allude to the circumstance that this island was said to be equidistant with Carthage from the Straits: a statement which is repeated by several later writers, who undoubtedly derived it in the first instance from Hanno. The statement of Scylax has, indeed, been a great difficulty with many modern writers, who have sought to reconcile it with the narrative of Hanno’s voyage, though the two are, in fact, wholly incompatible, unless we make considerable allowance, as sug- gested by Rennell, for the effect of a southerly current. The voyage from Carthage to the Straits is estimated by Scylax himself at seven days and seven nights’ voyage (equivalent to fourteen days on the ordinary mode of computation) under the most favourable circumstances (r013 KaAM/orov 'n'Aoii, § 111, ed. Miill.) NOTE E, p. 325. THE RIVER CHREMETES. It is much more difficult to determine the intermediate point ‘ mentioned in this first voyage. There is nothing to indicate its distance from Cerne, or its relative position with regard to the river next mentioned. But the description of a great river flow- ing from a lake, of such extent as to take a day’s voyage to pene¥ trate to its inmost extremity, can only be explained by supposing (as suggested by M. de St. Martin) that this was another arm of the Senegal, which in fact forms a great expansion or lake previous to its outflow into the sea, and may very probably have in ancient times had an outlet more to the north than any new existing. Yet the difficulty still remains, that the head of the lake is .de- scribed as “overhung by lofty mountains,” a circumstance that can hardly apply to any such lake in the delta of the Senegal. The river in question is called in our existing text of the Periplus the Chretes (Xpéms), but it has been supposed that this is a corruption, and that the true reading is Chremetes (Xpqaéms), a name which we find mentioned by Aristotle as that of a large river on the West Coast of Africa (Aristot. Meteorol. i. 13; the name is found also in Nonnus, Dionys. xiii. 374, xxxi. 103), but is unknown to the later geographers. If the river alluded to by Aristotle is really the Senegal, a vague tradition of its importance may perhaps have been derived from this voyage of Hanno. (336) CHAPTER X. WRITERS AFTER HERODOTUS. SECTION” 1.——Ctesias. § 1. WE have very imperfect means of tracing the progress of geographical knowledge among the Greeks during the interval of at least a century, which elapsed between the publication of the History of Herodotus and the expedition of Alexander into Asia. This period was indeed one of the most eventful in the history of Greece, and was, in many respects, the epoch of the greatest prosperity that that country ever witnessed. It was also one during which the intellectual activity of the Greek world developed itself on all sides; and among other forms of scientific inquiry that which regarded the form and constitution of the world in which they lived could hardly fail to attract the attention of the speculative Greek philosophers. But while the first foundations of such a scientific knowledge of the earth in its general relations,-——or of geography con- sidered as a part of cosmography,—must be assigned to the period in question, it does not appear that any great progress was made in that more detailed knowledge of the countries and nations occupying the known portions of the earth’s surface, which is commonly understood by the term geography. Nor were the limits of these known portions materially enlarged. The Greek world, with all its numerous colonies was still limited to the lands that surround the Mediterranean Sea; no Greek navigator had yet explored the waters of the Atlantic, and the Persian Empire on the east still included within its confines all that was really known to them of the continent of Asia. Within these limits their knowledge was SE01‘. 1. CTESIAS. 337 doubtless more complete and accurate in detail; and many of the more outlying nations were become more familiar to them than they had been to their predecessors. But any approach to scientific geography was rendered impossible, not‘ only by the very imperfect nature of their cosmographical notions, but still more by the want of instruments with which to make those scientific observations upon which all accurate geography must be based. I Yet it cannot be doubted that, had the works of some of the writers who flourished during this period been preserved to us in their integrity, we should have been able to form a much more complete judgment of the real extent and limits of the geographical knowledge of the Greeks. Unfortunately all those authors who would have been most valuable to us in this respect, have perished; and we are condemned to glean from the scanty fragments preserved to us by later writers some idea of the nature and value of their contributions. § 2. The works of the two principal historians that flourished in the generation after Herodotus were not calculated to throw any additional light on geographical knowledge. From the nature of its subject the great work of THUoYDIDEs was limited to a narrow area; and though the clearness of his geographical descriptions, in the few cases in which he has thought it neces- sary to give them—as in .that of Sicily in the beginning of the sixth book, and that of Thrace and Macedonia in the second—- corresponds with the definite and philosophical character of his mind, his narrative was generally concerned with countries, and even localities, so well known to his readers, that he had little opening for the display of his talent in this respect. § 3. His contemporary ANTIooHUS of SYRACUSE would un- doubtedly have added much more to our information had his works been preserved to 'us. These consisted of a history of Sicily, and one of Italy, in both of which, but especially in the latter, he appears to have introduced numerous notices of a geographical character. But the few extant fragments are not sufficient to enable us to judge how far he entered into VOL. I. z 338 HISTORY OF ANCIENT GEOGRAPHY. carp. x. any general geographical description, or to estimate the extent of his knowledge of those parts of the Italian peninsula which had not been occupied by Greek colonists. Almost the only notice of interest that has been preserved to us by the citations from his work is the fact that he was one of the first Greek authors who mentioned the name of Home. It is observed also that he did not concur in the ordinary tradition of its founda— tion by ZEneas, or one of his descendants, but supposed it to date from a much earlier period.1 §4. Among the writers who, in the period immediately following Herodotus, contributed to familiarize the minds of the Greeks with the more distant portions of Asia, a prominent place must undoubtedly be assigned to CTESIAS, a native of Cnidus in Caria, and a physician by profession, who in that capacity resided not less than seventeen years at the court of Persia.2 Such an interval must naturally have given him oppor- tunities of acquiring statistical and geographical information concerning the provinces of that empire superior to what had been possessed by any other Greek. Nor did he neglect to avail himself of the facilities thus presented to him. Among the works with the composition of which he occupied himself after his return to Greece (B.G. 398) we find mention of a treatise on the revenues of the Persian Empire,-—which it would have been interesting to compare with the information furnished us by Herodotus upon the same subject—as well as r of two professedly geographical treatises—the one on rivers, the other on mountains. Both of these are totally lost, and a meagre abstract by Photius is all that remains to us of his principal Work, the Persian history. Of the historical merits of the “ Persica ” in general, it does not fall within our province to‘ speak; had the work been preserved to us in its entirety it would unquestionably have afforded us many interesting ‘ Dionys. Halic. i. 73. published by Baehr (Ctesize Cnidii 2 Concerning the life and works of Operuin Reliquiaa, 8V0, Francofurti, Ctesias, see the Prolegomena to the 1824), and Colonel Mure’s History of excellent edition of his extant remains , Greek Literature, vol. v. pp. 482-500. Seer. 1. CTESIAS. 339. notices and ' casual details of a geographical character. We are told also by Photius that the author had appended to it a detailed account of the routes from Ephesus to Bactria and India respectively, with the number of days’ journeys and the distances in parasangs,——-a document that could hardly have failed to be of the highest geographical interest.3 Unfortu- nately none of these details have been preserved to us; and the extant abridgement of the Persica cannot be said to add anything to our geographical knowledge. § 5. His only other work of importance—of which we possess in like manner a mere abridgement by Photius 4'——treated specially of India and the Indians; and derives its chief interest from being the first professed account of that im- portant region. Unfortunately it was in this short treatise—- for the “Indica” occupied only a single book, serving appa- rently as a kind of appendix to his larger work—that the defects of Ctesias,—his want of critical judgment and love of the marvellous—were the most strongly developed. India was already in his day become the “land of marvels” to all the neighbouring Asiatic nations, which it has continued down almost to our own times; and Ctesias, while residing at the Persian court, appears to have accepted without hesitation all the marvellous tales that he could collect concerning the distant regions of the East. The consequence is that while he laid the foundation for a vast mass of fables and absurdities, which continued to be propagated by successive writers down to the latest period of Greek literature, be contributed almost nothing to the real knowledge of the land of which he wrote. 3 At the same time it is remarkable that no reference is made by any later writer to this itinerary; whence we may perhaps infer that it was not con- sidered as based upon any adequate authority. “ The abridgement is, however, in this case much more copious than that of the Persica : the abstract of the lndica occupying eleven pages (in B'ahr’s edition), though the original was only in one book, while the 23 books of the Persica in the epitome of Photius fill only eighteen pages of the same edition. Besides this we have copious extracts from Ctesias by ZElian, in his History of Animals (iv. 21, 26, 27, 36, 46, 52; v. 3; xvi. 31 ; xvii. 39), some of which are probably ,copied almost literally . from the original. 22 340 HISTORY OF ANCIENT ‘GEOGRAPHY. CHAP. X. So far ‘as we can judge from the imperfect form in which his work has descended to us, his geographical knowledge of India was little, if at all, in advance of that of Herodotus. The enormous extent that he assigned to it,-—representing India alone as equal to the whole of the rest of Asia,5 was evidently a mere vague assertion. N o trace appears of his having been acquainted? even with the name of the Ganges, or with any- thing beyond the Punjab and the valley of the Indus. Of the latter river he says, with his usual exaggeration, that it was forty stadia broad in its narrowest part, and 200 stadia (20 G. miles) in its greatest width.6 But he not only does not notice the existence of crocodiles in it, as had been correctly reported‘ by Herodotus, but expressly says that it produced no other animals than a gigantic worm, seven cubits in length, which could drag into the river, and devour, oxen, and even camels.7 He was indeed familiar with the use of elephants in war by the natives, but here again he distorts the‘ fact by enormous exaggeration, representing the Indian king as marching to battle with a hundred thousand elephants (I), besides 3000 of superior strength and stature, which were employed in destroying the walls and towers of hostile cities.8 His account of the powerful Indian dogs is more reasonable, and his mention of the use by the natives of eagles, kites, ravens, and crows in the pursuit of small game is curious as the first notice of falconry, a practice prevalent in the East from the earliest ages, but wholly foreign to the Greeks.9 §6. It is unnecessary to dwell upon the fabulous tales which he relates, both of marvellous animals,—such as the Martichoras or man-eater, the Grifiins, which he describes as guarding the gold in the mountains of the interior, and the Unicorn. or wild ass with a single horn on its forehead—and of still more marvellous races of mankind, among which we 5 Ap. Strab. xv. p. 689. ‘" Ibid. § 27. Fragm.15. 6 Indicu, § ]. He was, however, 8 Fragm. 2. aware that it flowed for a pait of its 9 Indica, § 11. Fragm. 7. course through a mountainous country. SECT. l. CTESIAS. 34f. find the Pygmies, the Sciapodes, and the Cynocephali, or men with dogs’ heads, who were more commonly assigned to the interior of Africa. But it would appear as if Ctesias had col- lected from all quarters whatever could serve to swell his list of marvels, and excite the wonder of his more ignorant and uncritical countrymen. That such a tissue of fables, “among the most extravagant ever brought within the compass of a single compilation in the most credulous times” should have been presented to the Greek public shortly before the days of Aristotle is indeed, as Colonel Mure justly observes, a singu- lar phenomenon.1L But we must remember, in justice to Ctesias and his contemporaries, that these fables, absurd as they were, were deemed worthy of being repeated by successive writers, and still found readers, if not believers, in the days of Pliny and ZElian. At the same time ‘Ctesias himself early obtained, with the more critical part of the public, the reputation which he deserved of a worthless compiler of fables.2 § 7. Among the few grains of real fact to be gleaned out of this mass of absurdities, is the notice that onyxes, sards or sardonyxes, and other precious stones used for Signet-rings, were brought from the mountains in the interior of India. On the other hand he speaks of a river that produced abundance of amber, a substance that was certainly never among the productions of India. No mention is found (at least in the remaining extracts) of any of the customs peculiar to the true Hindoos, which so strongly excited the attention and curiosity of the Greeks, when they visited India with Alexander ; indeed it may safely be asserted that there is hardly a single state- ‘ Mure’s History of Greek Literature, vol. v. p. 497. The reader, who is curious in such matters will find in the same passage a good summary of all these absurd fictions. Their only interest arises from the persevering manner in which they are quoted by successive writers in later times. 2 Thus Aristotle says of him 66s 4mm K'rno'ias, aim Zlw £151611'L0'7'0$(Hi8t. Animal. viii. 28), and again in citing his testi- mony, el 56? 'ma'reiizraz K'r'no'iqz (Ibid. ii. 1). Arrian, quoting his statement concerning the Indus, adds ei 87’; amp {Kai/bx real K'r'qo'ias e’s relrp'npzéio'w (Anal). v. 4, § 2). See also ZElian (Hist. Am'vn. iv. 21), Lucian (Vera Historia, i. 3; ii. 31), and Plutarch (Artamerm. c. i.). Strabo refers to him among the writers on whose statements no reliance could be placed (i. 2, p. ‘13 ; xv. 1, p. 689). 342 HISTORY OF ANCIENT GEOGRAPHY. CRAP. X. ment to be found in Ctesias, concerning either the country or the people, which has the appearance of being deriyed from any direct or trustworthy information. Yet he had the audacity to assert, while stringing together this tissue of absurd fables, that all his statements were strictly true, and were derived either from personal observation, or from the testimony of trustworthy witnesses.3 SECTION 2.—The Anabasis of Xenophon. § 1. The next author whom we have to consider is one of a very different stamp. The expedition of the younger Cyrus from Ionia to the neighbourhood of Babylon (in B.0. 401), with a view to the overthrow of his brother Artaxerxes, has been rendered for ever memorable, not only by the courage and skill with which the body of Greek mercenaries, who had ac- companied him up the country, made good their retreat from the field of Cunaxa to the shores of the Euxine, but still more from the narrative of their adventures having been transmitted to us by one of those who had himself shared in all the diffi— culties and dangers of the retreat, and had taken an active part in surmounting them. The Anabasis of XENOPHON is certainly one of the most authentic, as well as one of the most delightful, episodes of ancient history; and whatever defects its author may elsewhere display in some of the higher qualities of a historian, the narrative in question will always remain one of the most perfect specimens of historical composition—where the writer is relating events of which he is able to speak from his own knowledge. Geographically considered, the Anabasis must certainly have been an important contribution to the knowledge possessed by the Greeks of the countries to which it related. None of these countries were indeed altogether unknown to them; but their 3 Indica, § 33. SECT. 2. THE ANABASIS OF XENOPHON. information concerning them was undoubtedly very vague and superficial; the mountainous regions traversed by the Greek troops on their retreat were of the wildest character, and had doubtless been as little visited by peaceful travellers in those days, as they had been, until very recently, in modern times. There can hardly be a stronger testimony to the courage and perseverance of the Ten Thousand and their leader, than the fact that the rugged and inhospitable tracts through which they forced their way in midwinter, had remained down to our own days almost unexplored, and unknown to modern travellers, on account of the dificulties by which they were beset. § 2. It was fortunate that Xenophon, who accompanied the march in the first instance as a volunteer, and without the least idea how far it was going to lead him, appears to have preserved from the first a regular record of the route, noting not merely the principal places through which the army passed, but the number of days’ march, and the distance from one point to another} These distances were computed in parasangs, according to the Persian custom; and so long as the Greeks accompanied the Persian army, and their march lay along a line of route well known to the Persian officers, as must have been the case throughout the advance up the country, there can be no reason to doubt that they were sub- stantially accurate.5 But the case was very different with the retreat, especially with that part of it which lay through the mountains and high table-land of Armenia, through which there were certainly no frequented high roads, and where the Greeks were repeatedly left to force their way without the assistance of local guides. Upon what principle Xenophon calculated the distances under these circumstances it is not easy to say. We can only feel sure that any approach to real measurement was impossible, and it is obvious that the natural mode of computation by the time occupied on the march, 4 Note A, p. 359. ~" Note B, p. 359. 344 HISTORY OF ‘ANCIENT GEOGRAPHY. CHAP. X. would be extremely difficult of application. It must always be borne in mind that the Greeks had no portable instruments for measuring time, and that even the division of the day into hours was unknown, or at least unfamiliar to them in the days of Xenophon.6 vHence this ready mode of estimating dis- tances, so familiar to the modern traveller, would be altogether wanting. Still, while the troops were marching at an ordinary rate through a level, or even undulating country, a tolerable comparative estimate might be formed of the length of each day’s march, and it is probably this which Xenophon sought to express in parasangs; but when their route lay through a wild and rugged mountain country, harassed at every step by the hostile tribes which occupied it, or toiling through deep and newly~fallen snow, it is evident how impossible it would be to preserve any reasonable estimate of the distances actually traversed, and how little dependence can consequently be placed upon the statements of them given by Xenophon.7 Making allowance for the inevitable errors resulting from this cause, we are able to trace for the most part in a general way the line of route followed by the army of Cyrus on its advance, as well as that taken by the Greeks in their memorable retreat. It is not to be denied that there still remain con- siderable difficulties to be cleared up in regard to details, arising however in all probability as much from our own imper- fect knowledge of the countries in question, as from any want of accuracy on the part of the historian. Some of the most important points have indeed only been established in a satisfactory manner within a very recent period.8 A brief 6 According to Ideler (Handbuch der History 0 Cyrus’s Expedition, (4to. Chronologie, vol. i. p. 238) the use of the word é’ipa or hour for the twelfth part of a day was not in use among the Greeks till after the time of Alex- ander. 7 Note C, p. 361. 8 For a fuller and more detailed examination of the geography of the route in question the reader is referred to Major Renucll's Illustrations of the Lond. 18 1), as well as to the more recent works of Mr. Ainswolth (Travels- in the Track of the Ten Thousand Greeks, 8vo, 1844), and Prof. Koch (Der Zug der Zehutauseud, 8vo, Leipzig, 1850). Much valuable information has been furnished by Mr. Hamilton (Travels in Asia Minor and Poutus, 2 vols. 8vo, Lond. 1842), and by Colonel Chesney’s Survey of the Euphrates and Tigris SECT. 2.- THE ANABASIS OF XENOPHON. 345 outline of the general line of route is all that can be here attempted. § 3. Setting out from Sardis in the spring of B.0. 401, Cyrus advanced through the centre of Asia Minor by a well-known line of route as far as Celaenae, an important city of Phrygia, on the site subsequently occupied by Apamea, close to the sources of the Maeander. Here he halted thirty days, during which time he received fresh reinforcements of Greek troops. It must be observed that the pretext under which he had assembled these under his standard, was that of an expedition against the mountaineers of Pisidia, who were practically inde- ' pendent of the Persian king, and infested the neighbouring countries with their incursions. As far as Celeenae his route was consistent with this object, as well as with his real'purpose of advancing through the passes of Cilicia and Syria into the _ valley of the Euphrates. But from Celaenae he made a sudden detour—for what reason is not explained to ns—and after striking northwards for some distance, till he approached the frontiers of Mysia, then turned again to the east and proceeded by another frequented and well-known road from the neigh- bourhood of Synnada (near Afium Kara Hissar) by Thymbrium and Tyriaeum to Iconium, the last city in this direction that was included in his own satrapy.9 Entering now upon what might be termed hostile territory, though meeting with no opposition, he marched for five days through Lycaonia, and for four more through Cappadocia, until he arrived at Tyana (called by Xenophon Dana), already a large and opulent city, situated at the entrance of the pass into Cilicia. This cele- brated pass, subsequently well known as the Pylae Ciliciee, is described by Xenophon as impossible for an army to force, if properly defended. It had at first been occupied by the Cilician king Syennesis, but was abandoned by him without striking a blow, on learning that a small Greek force, dispatched (2 vols. 8vo. Lond. 1850). The whole notes to the 69th and 70th chapters subject has been discussed in the most of his History of Greece. elaborate manner by Mr. Grote in the 9 Note D, p. 363. 346 HISTORY OF ANCIENT GEOGRAPHY. CHAP. X. by Cyrus from Lycaonia, under the guidance of the Cilician queen, had crossed the Taurus by another route, and that the fleet of Cyrus had also arrived upon the coast. Cyrus was thus enabled to traverse the formidable pass without opposition, and descended to Tarsus in the midst of the Cilician plains.1 §4. Here another halt was made, of not less than twenty days. It must have been long evident—at least to the leaders of the Greeks, that the march against the Pisidians had been a mere pretence; and that the expedition of Cyrus was really directed against the Great King himself. ‘The soldiers now mutinied, and refused to follow him any farther, but were gradually persuaded to proceed, partly under the flimsy pretext that he only meant to advance as far as the Euphrates, where a Persian general named Abrocomas, a personal foe of Cyrus, was supposed to be posted. Having thus induced the Greeks to follow him, Cyrus advanced through the succession of diffi- cult passes in the neighbourhood of Issus, which played so important a part during the march of Alexander, but which on this occasion were unaccountably neglected by'the Persian generals? At 'Myriandrus, a city on the Gulf of Scanderoon, which was at this time a place of considerable trade,3 Cyrus halted for a week, before breaking off all communication with the sea, his fleet having here met him for the last time. From hence he struck at once into the interior, and continued his march without interruption to the Euphrates. The details of this part of the march present no considerable difficulty, though the line of route is not otherwise known. He appears 1 Anab. i. 2, §21—25. For a descrip- tion of these Cilician gates (see Ains— worth’s Travels in the Track of the Ten Thousand, pp. 44—46; Chesney‘s Ea- phrates and Tigris, vol. i. p. 353 ; Lan- glois, Voyage en Uz'licie, pp. 367-370. All these modern writers fully confirm the accounts of Xenophon and other ancient authors concerning the formid- able character of the pass, which would afford an imprcgnablc position, if pro- pcrly defended. 2 Note E, p. 364. 3 The exact site of Myriandrus has not been determined; but it is of little importance, as its position is well known within a few miles. It is placed by Strabo between Rhosus and Alexandria (xiv. 5, p. 676). The foundation of the latter city, which speedily rose to be the most important place on the Gulf of Issus, had naturally the effect of producing the decline of the neigh- :_ bouring towns. Seer. 2. - . THE ANABASIS or XENOPHON. ‘347 to have reached the river at a point considerably above Thap- sacus, and to have followed its course for some distance down to that place, which was at this period the customary place of passage, and where Cyrus accordingly prepared to cross the Euphrates.‘ Here there was again some hesitation on the part of the Greek mercenaries; but most of them saw plainly that the die was cast, and it was too late to recede; they accord- ingly crossed the river, and prepared to meet the army of the Great King. - The position of Thapsacus, though a point of the highest importance, not only for the geography of the Anabasis, but for that of the campaigns of Alexander, and the subsequent geography of Asia, has only been definitely ascertained within a very recent period. It was situated just above the modern town of Itakka, at the only point in the central course of the Euphrates where that river is fordable (though even here only at certain seasons of the year), for which reason it con- tinued to be used alike by the Persian, Greek and Roman armies during a long period. It was also a commercial route of importance in ancient times. At the present day the place of passage is known as the ford of the Anezeh or Bedouins; and is in fact resorted to only by the wandering Arabs of the desert.5 §5. From Thapsacus the line of march lay along the left bank of the Euphrates as far as a river termed by Xenophon the Araxes, which may be safely identified with‘ the Khabur (the Chaboras of Ptolemy and Pliny), as this is the only con- siderable river, which falls into the Euphrates in this part of its course.6 The march from henceforth lay through a desert 4 Note F, p. 365. 5 Note G, p. 365. 6 All writers who have examined the subject have agreed upon this point; but no explanation has been offered of the name of Araxes given by Xenophon to the river in question. It is curious, however, as tending to show the frequent use of that name in Asia, and thus helping to explain the con- fusion into which Herodotus fell upon the subject. It may be observed that the ditficulty found by Major Rennell on account of the distance from Thapsacus, of 9 days’ march and 50 parasangs, disappears as soon as Thapsacus is placed in its true , position, instead of at Deir, little more 348 HISTORY OF ANCIENT GEOGRAPHY. CHAP. X. tract—regarded by Xenophon as a part‘ of Arabia, though on the left bank of the Euphrates—of which the historian has left us a graphic description, confirmed by the observation of recent travellers. The whole country for five days’ march was a level plain, without trees or inhabitants, and frequented only by wild asses, gazelles, bustards and ostriches.7 Five days’ march through this desolate region brought them to an ‘uninhabited city named Corsote, from whence they had again thirteen days’ march, through an equally sterile and unculti- vated' tract, to a place called Pylaa, situated on the verge of Babylonia, a name which Xenophon apparently applies solely to the rich alluvial country, abounding in villages, and inter- sected by canals of irrigation, which extended from hence without interruption to beyond Babylon.8 § 6. Cyrus now found himself rapidly approaching the great - army of Artaxerxes: and he was met by the king himself on the sixth day’s march after leaving Pylae. Unfortunately the details of the military operations that preceded and followed the decisive battle at Cunaxa cannot be traced with any certainty. Even the scene of the battle itself can only be approximately determined: the name of Cunaxa (apparently a mere Babylonian village) is not mentioned by Xenophon, and is preserved only by Plutarch, who doubtless derived it, as well as other particulars of the battle, from Ctesias, who was him- self present with the army of Artaxerxes.9 According to the same authority it was 500 stadia distant from Babylon, though Xenophon was told that the field of battle was only 360 stadia from that great city.‘ But as this was more hearsay evidence, the statement of Plutarch is probably in this instance the than 20 G. miles above the confluence more trustworthy: and if it be followed, the field of battle may be placed a'few miles to the south of the modern castle and given concerning these wild animals, and the difliculty of their pursuit. 8 Note H, p. 366. strong personal propensity for the \ 9 Plut. Artax. c. 8. chase shows itself in the details he has 1 Anal). ii. 2, § 6. of the Khabur. ’ Anab. i. 5, §§ 1-3. Xenophon’s SECT. 2. 'THE ANABASIS or XENOPHON. ' 349 village of Felujah.2 But the subsequent movements of the Greeks are almost wholly unintelligible to us: even the site of Sittace, the place where they ultimately crossed the Tigris, and which is called by Xenophon “a great and populous city,”3 cannot be identified, though it must probably have been situated a short distance to the south of the modern city of Bagdad. On the other hand recent researches have thrown considerable light on one of ‘the difficulties that had been a great stumbling—block to all previous inquirers—the existence of a great line of wall, called by Xenophon the Wall of Media,‘ which he describes as not less than 20 feet in thickness and 100 feet in height. It was said to extend 20 parasangs in length, and was not far distant from Babylon} Through this mighty barrier the Greeks passed on their way from the field of battle to the Tigris, so that Xenophon could not have been -. misinformed or deceived upon the subject, except as to its extent. It seemed impossible that no trace should be left of such a gigantic work: yet until very lately no remains were discovered, which could by any ingenuity be made to cor- respond with the position required by the narrative of Xeno- phon. But in the recent survey of this part of Mesopotamia, Lieut. Bewsher discovered the ruins of a wall, running from N .W. to SE, which he was able to trace for a distance of 10% miles, and which may probably have been much more extensive.5 Such a wall would lie directly across the route of the Greeks in proceeding from Cunaxa to the Tigris: and this discovery has tended much to clear up the topography of _ this part of Babylonia, as well as to confirm the accuracy of Xenophon’s narrative.6 _ § 7. It was from the time that they crossed the Tigris near Sittace that the retreat of the Greeks may properly be said to have commenced: their march was at first conducted in a 2 Note I, p. 369. p. 169. These remains are now known 3 Note K, p. 370. as Hubl es Sukhr——“ the line of stones 4 Anal). ii. 4,§ 12. or bricks.” 5 Journal of Geogr. Soc. vol. xxxvii. 6 N otc L, p. 370. See the Map. HISTORY OF ANCIENT GEOGRAPHY. CHAP. X; friendly manner, and under the convoy of Tissaphernes, through the open country on the left bank of the Tigris, till they came to the river Zapatas, which is unquestionably the stream now known as the Great Zab, the Zabatus of Strabo. This is the first point on this part of their route which can be identified with certainty: the intermediate halting-places being still undetermined,7 though one of them, Opis, is described as a large city. To the Greeks the- halt on the Zabatus was one of the most important points of their expedition for another reason—that it was while encamped here, previous to crossing the Zab, that Clearchus and the other generals were trea- cherously seized and put to death by Tissaphernes. Hence- forth therefore the Greeks had to continue their retreat through a hostile country, and with only such occasional and local guides as they could procure for themselves. § 8. Thus far their line of retreat had followed apparently that of the high road, or “royal road ” of Herodotus, through the plains of Assyria towards the mountains of Armenia,8 and it was doubtless by that line of route that Tissaphernes in- tended, or professed his intention, to conduct the Greeks back to Ionia.9 But when the latter, under the command of Xeno- phon and their other new generals, after crossing the Zab, had continued their march for 9 or 10 days more along the left bank of the Tigris,1 they had to force their way with much difficulty through a hilly tract, occupied by the enemy’s troops; and after surmounting these obstacles they found their 7 Note M, p. 372. into use by geographers in later days. 8 It is remarkable that all this part 9 This was evidently the route indi- of the march, from the time they . cated also by Ariaeus, the commander crossed the'Physcus, is described by of the Persian troops in the army of Xenophon as lying through Media (ii. ~ Cyrus, when, after the battle of Cunaxa, 4, § 27). Herodotus, as we have seen, he pointed out the impossibility of re- ineluded the same district in the land turning by the same route they had of the Matieni. It seems clear that the followed on their advance, but pro- name of Assyria was not in use as a posed to take another line, which would territorial designation at this period. be longer, but along which there would It had perhaps been intentionally sup- be no difficulty in procuring supplies pressed after the fall of the Assyrian , (Anal). ii. 2, § 11). monarchy, and was only brought again 1 Note N, p. 374. SECT. 2. THE ANABASIS OF‘ XENOPHON. 351'- farther advance effectually barred by the mountains descending abruptly to the stream, so as to leave no passage along the bank. The customary line of route was to cross the Tigris at this point, and take a westerly direction towards the Euphra- tes:2 but the'Tigris was much too deep to be forded, and to cross it otherwise in face of the enemy’s cavalry was impos- sible. Hence the only course that remained for the Greeks was to strike at once towards the north into the mountains of the Carduchians, with a view to reaching the high lands of Armenia, from whence they could descend to the Greek colo- nies on the Euxine. By‘following this course they hoped to pass the Tigris and the Euphrates near their sources, where they would of course be fordable.3 ‘ The Carduchians were a warlike race of mountaineers, who had maintained their independence in the fastnesses of a wild and rugged country against all the efforts of the Great King: and they opposed a fierce resistance to the passage of the Greeks!‘ Seven days of almost continual fighting—during which the Greeks suffered more than they had done from all the armies of the Persian monarch 5—-—at length brought them through the mountains to the valley of the Centrites, which separated the land of the independent Carduchians from the Persian satrapy of Armenia. Two days’ march, after crossing this river, enabled them to surmount the sources of the Tigris, and three days more brought them to the river 3 This was distinctly stated by the captives who were consulted by the Greek generals : “ the road to the west, crossing the river, led to Lydia and lonia” (iii. 5, § 15). Hence the Greeks were well aware that at this point they quitted the line of route which they ought to have followed; and took a i thought. But the captives told them i that, when once arrived in Armenia, it 1 would be easy to proceed from thence . in whatever direction they chose ((31/— '; T617661! 5t ei‘irropou E'cbao'azl Glllal ii'nor 11$ 5 e’fie'Aor 1ropeziecrfiou. iii. 5, 17). v 4 There can be no doubt that these ECarduchi were the ancestors of the direction altogether different. 3 This is distinctly stated by Xeno- phon as the reason for their directing their march towards the north (iv. i. §§ 2, 3). No allusion is made to any intention of reaching the Euxine, which [modern Kurds, who still inhabit the gsame mountain tract, and until very ‘ recently maintained their virtual inde- l pendence against the Turks and Per- . sians alike. 3 5 Anal). iv. s, g 2. was probably altogether an after- _ 352 nrsrosr or ANCIENT GEOGRAPHY. one». x. Teleboas. They were now on the elevated table-land of Ar- menia, where they suffered severely from the severity of the weather, and from deep falls of snow. Struggling on through these difficulties, they reached the Euphrates, in the upper part of its course, and, as they were told, not far from its sources.“ The river was here easily .fordable, and they con- tinued their march over the open, upland country, without any hostile opposition, and meeting with a friendly reception from the Armenian villagers, but encountering great difficulties from the snow. § 9. Thus far the course pursued by the retreating army admits of being traced with tolerable certainty, notwithstanding many difliculties in detail. The researches of recent travel- lers have established several points beyond any reasonable doubt. One of the most important of these—the point where they quitted the Tigris, and began their march northwards towards the Euxine, may be clearly placed near the modern town of Djezireh ibn Omar-é—the Bezabde of the Romans. It is immediately above this that the mountains close in upon the Tigris so abruptly as to render all farther progress along its valley impracticable, on which account the modern road crosses by a bridge of boats, and proceeds westward by N isibin to Diarbekr.7 This the Greeks could not do, and were thus compelled to plunge into the mountain region to the north. The Centrites again may be clearly identified with the Buhtan Chai', one of the principal tributaries of the Tigris, flowing from east to west, and constituting (as Mr. Ainsworth remarks 8) a kind of natural barrier between Kurdistan and Armenia. The Teleboas may likewise be assumed with much probability ‘to be the Kara Su, a tributary of the Euphrates, flowing 6 Anab. iv. 5, § 2. 7 Mr. Lay-ard, who descended the Tigris from the upper part of its course, has given a striking account of the narrow gorge through which the river flows between the village of Tilleh, “where it receives the united waters of Bitlis, Sort, and the upper districts of Buhtan,” and the town of Djezireh (Nineveh and Babylon, pp. 50, 51). The difficulties of the passage are such as to compel even an ordinary traveller to cross the mountains between the two points. ‘3 Travels, p. 166. SECT. 2. THE ANABASIS OF XENOPHON. 353 through the fertile valley of Mush. Xenophon’s statement that in passing from one of these rivers to the other the Greeks had passed beyond the sources of the Tigris, can be readily explained by supposing that he took the northern tributary of that river (now called the Bitlis Chai', or river of Bitlis) for the main stream. The real source of the Tigris, or at least what is considered as such by modern geographers, was far away to the west.9 The place where they forded the Euphrates cannot be exactly determined, but it was evidently the Murad Chat’, or eastern branch of the river, that is here designated.1L That stream flows through the elevated table-land of Armenia, which has here an average height of from five to six thousand feet above the sea-level, an elevation that at once accounts for the great cold and deep snows for which this region has been always noted. Across this dreary tract the Greeks had to plod their weary way in the middle of winter: a feat which has not been performed by any modern traveller, still less by any modern army. § 10. But from this point it becomes impossible to follow the line of their farther progress with any reasonable proba- bility. Its steps, as detailed by Xenophon, may be briefly recounted. From the spot where they crossed the Euphrates four days’ march brought them to some Armenian villages, in the immediate neighbourhood of a palace or castle, where they halted for a week, and then continued their march across the snow-covered plain for three days more, without meeting with any villages. Here their guide, having unfortunately been insulted by Cheirisophus, abruptly quitted them, and they were left to find their own way, during a march of seven days (estimated at 35 parasangs), following apparently—during a part at least of their course—the valley of a river to which 9 See Note O (p. 374). I is still known as Frat. But Armenian ‘ Xenophon is the only Greek writer writers apply the name of Euphrates who gives the name of Euphrates to to both arms, and the same usage pro- this branch of the river; Strabo and bably prevailed among the natives in the other geographers mentioning only the time of Xenophon. the western or northern branch, which VOL. I. 2 A 354 HISTORY or ANCIENT GEOGRAPHY. cm». x. Xenophon gives the name of Phasis? Quitting this valley, they crossed a high mountain pass, where they were encoun- tered by ‘the combined forces of three mountain tribes, the Chalybes,3 the Taochi, and the Phasiani. After surmounting this opposition they made five long days’ marches (called by Xenophon 30 parasangs) through the land of the Taochians, suffering severely from want of provisions, until they came to a fort or stronghold of that people, which they took by storm, and thus obtained abundant supplies. Hence they again marched for seven days through the country of the Chalybes, till they came to the river Harpasus, an important stream: after crossing which they entered the territory of the Scythini, and proceeded through it for eight days to a place called Gymnias, which Xenophon describes as a large and wealthy city.4 It was the first place which they had met with de- serving this name since leaving Opis. Here they were received in a friendly manner by the ruler of the country, who fur- nished them with a guide, promising to conduct them within five days to a spot within sight of the sea. The promise was duly kept, and on the fifth day the Greeks beheld the long- looked-for Euxine, from the summit of a ridge or mountain called Theches.5 There still remained for them however five days’ march—three through the land of the Macrones and two 2 The expression of Xenophon, we. 'roii'ro e’rropez’zflno'av é'rr'rd oraeuoz‘ls dud 'II'éV'I'E napao'ai'yyas 'rfis ripe'pas 'n'apa 'rbv @5011! wo'rapuiv (iv. 6, §3), would un- doubtedly seem to imply that the whole march lay along the valley of the Pha-sis, but it may well be doubted whether this is intended. 3 The Chalybes here mentioned are- obviously wholly distinct (geograph- ically speaking) from the people of that name, whom Xenophon subsequently found settled on the shores of the Euxine (v. 5, § 1), and who were a branch of the Chalybes so celebrated among the Greeks as workers in iron. It is singular that Xenophon himself in another passage (v. 5, 9' 17) refers to these mountain Chalybes under the name of Chaldaeans; and, according to Strabo (xii. 3, p. 549), this was the name by which the Chalybes on the Pontas were known in his time. 4 miMv ye'yaivhv Kai ez’idaiaova Kai oircovae'vnv, ‘i7 e’lcaAe'i'ro I‘v/avias (iv. 7, §19). The name is written by Dio- dorus I‘vpvao'ia (DiOd. xiv. 29). 5 Anab. iv. 7,§ 21. No mention is ound in Xenophon of the title of “ the Sacred Mountain,” which is applied to Mount Theches by later writers. Dio- dorus, who often differs from Xenophon in the names he employs, while follow- ing him in the substance of his narra- tive, calls it Mt. Chenium ('rb X'ljz/wv b’pos, xiv. 29). He erroneously places it fifteen days’ march from Gymnesia, the Gymnias of Xenophon, instead of five. SECT. 2. THE ANABASIS OF XENOPHON. through that of the Golchians—before they found themselves at the Greek colony of Trapezus, or Trebizond. Here their wanderings and dangers may be considered as having come to an end: their subsequent progress along the southern coast of the Euxine—from Trapezus to Cotyora by land, and from thence to Sinope and Heraclea by sea—presents comparatively little interest, though not without value for its geographical details, as well as for the light which it incidentally throws upon the relations of the Greek colonists on the Euxine with their barbarian neighbours. § 11. The details of the march of the Greeks, as above given from Xenophon appear so distinct, and in themselves are so clear and intelligible, that it is the more disappointing to find the extreme difficulty of explaining, or reconciling them with the existing geography of the regions in question. \Ve are still indeed, notwithstanding the researches of recent travellers, but imperfectly acquainted with the countries that must have been traversed by the Greeks, between the high table-land of Armenia and the Black Sea; but enough is known of their physical geography to show that they constitute one of the most rugged and intricate mountain tracts in any part of the world. If we consider the position of the Greek leaders, having to force their way through such a country, in. the midst of hostile tribes, without maps, without compass, with mere local guides, who probably knew nothing of the neigh- bouring countries, and having themselves but a very vague general notion of their situation with regard to any known points, it cannot surprise us to find that the geographical data are in reality wholly incomplete. From the passage of the Euphrates till they arrived at Trebizond, it may be fairly said that not a single point can be identified with any approach to certainty. The river Phasis may indeed be reasonably supposed to be the branch of the Aras or Araxes, which flows through a district still called Pasin in the upper part of its course,6 and the presence of the Phasiani " Brant’s Journey in Armenia, in the Journal of ({mgr. Soc. vol. x. p. 341. 2 A 2 356 HISTORY or ANCIENT GEOGRAPHY. one. x. in the same neighbourhood would tend to confirm this view. It is certain at all events that it had nothing to do with the celebrated Colchian river of the name, though it is not im- probable, as has been suggested, that the Greeks supposed them to be identical, and followed the course of the river farther than necessary under this impression. But it seems impossible to believe supposed by some modern writers) that they wandered for many days’ march from their true direction, and then made a sudden return in order to recover it. N o trace of such an error is found in the narrative of Xenophon, who could not have omitted to mention a circum- stance which would have exercised so important an influence on the fortunes of the army. It is indeed utterly impossible to explain the distances given, and even if we admit these to be greatly exaggerated, the number of days employed on the march remains to be accounted for.7 The river Harpasus cannot be identified with certainty; nor can we fix upon the spot from whence the Greeks first caught sight of the sea. Even the situation of Gymnias is equally uncertain; though it might have been thought that a great and flourishing city (as Xenophon calls it) within a comparatively short distance of the sea, must have been well known to the Greeks in after times. But no mention of the name is found in any later geographer; nor even of the tribe of the Scythini, whose capital it was. The other nations mentioned by Xenophon in this part of his route, the Chalybes or Chaldaeans, the Taoehi, the Phasiani, the Macrones, and the Colchians, are all of them known to us from other writers; but both the names and the abodes of these wild mountain tribes were so fluc- tuating and uncertain that we derive from them little assist- 7 Col. Chesney says of the route he ordinary acccptation of the word ;” and followed from the mountain which he he adds that the marches of the Greeks calls Gaur Tagh, that the journey from through the mountains of Kurdistan thence to Trebizond occupied five days, and Armenia must often have encoun- “ owing to the necessity of passing terod difiiculties and delays of asimilar along what in reality is more a winding character (Euphrates and Tigris, vol. ii. chasm than a mountain valley in the p. 232). SEOT. 2. THE ANABASIS OF XENOPHON. ance in determining the exact geography of this part of the . march.8 §12. It may be observed that all these tribes, from the Euphrates to the sea, seem to have been in fact wholly inde- pendent of the Persian Empire; from the time the army quitted Armenia no trace is found of Persian authority. Even the petty nations that bordered on the Euxine, from Trapezus to the frontiers of Paphlagonia, the Mosynoecians, the Cha- lybians, and the Tibarenians, appear to have enjoyed a state of virtual independence, and the governor or ruler of the im- portant province of Paphlagonia conducted himself towards the Greeks with all the freedom of an independent sovereign, though doubtless owning a nominal allegiance to the Persian monarch. It was not till the army landed in Bithynia that we again find the Persian satrap Pharnabazus taking part in opposing their progress. § 13. The narrative of Xenophon, it‘ must always be remem- bered, was composed with a historical, not a geographical object. Hence the geographical details are introduced almost entirely for the purpose of explaining and rendering intel- ligible the operations of the Greek forces. Circumstances had indeed placed the historian very much in the position of a geographical explorer, or at least of a traveller through almost unknown regions; but the description of those countries and their inhabitants formed no part of his main subject. Such notices as we find of them are introduced merely incidentally from their natural connection with the incidents of the march, and never assume, as in Herodotus, the character of collateral episodes. But in the few cases where he has entered into more detail than usual—as in the description of the desert of Mesopo- tamia, in that of the underground dwellings of the Armenians, and the semi-barbarous habits and manners of the Mosynoecians ——we recognize at once the character of an intelligent and trustworthy observer, and find cause to regret that he has not 8 Note P, p. 375. 358 HISTORY OF ANCIENT GEOGRAPHY. on“. x. more frequently dilated upon such subjects. One peculiar circumstance which he relates—the singular effects produced upon his soldiers by the poisonous honey found in the moun- tains of the Colchians—has been completely verified by the observations of modern travellers.9 The other works of Xenophon are of no importance in a geographical point of view. His Hellenica, like the great work of Thucydides, of which it is the continuation, is confined to the affairs of Greece, and the countries immediately adjacent ; while the Cyropeedz'a, which, had it been a real history of the reign of Cyrus, could not have failed to be a work of much interest to the geographer, is in fact a mere historical romance; and the incidental notices of a geographical character that are actually found in it, are not only very vague and indefinite, but in many cases utterly erroneous and calculated to mislead the reader.1 9 Anal). iv. 8, § 20. According to not found on the south side of the same Mr. Hamilton the deleterious quality range (Hamilton’s Asia Minor, vol. i. of the honey is produced by the bees pp. 160, 164, 166). This observation feeding on the flowers of the beautiful ' exactly agrees with the narrative of Azalea Pontica, which clothes the hill- 1 Xenophon. sides above Trebizond, and ascends the 1 See the remarks of Col. Mure in slopes of the mountains next the coast f his History of Greek Literature, vol. v. for a height of 4000 or 5000 feet, but is ! pp. 384-386. MAP TO ILLITS TRATE THE MARFHES OF XENOPHON & ALEXANDER. 1N LOWER ASIA 28 E.0f Gr. 36 London: John Murray. Route of‘ Xfnnphon D!’ of ' A Lara)! (1?!‘ VI Ina? 3. Cwmma l l l Ba v 1011 .14 vroq in: “"1 ix mrnl. . > NOTES A, B THE ANABASIS OF XENOI’HON. NOTE A, p. are. COMPOSITION OF THE ANABASIS. IT seems absolutely necessary to suppose that some such notes were preserved by Xenophon, otherwise it would have been im— possible for him to have given the details of the march with the care and accuracy which distinguish them in all those parts of the route where we are able to verify them with any certainty. We know that the Anabasis—at least in the form in which it is now preserved to us—was not composed till many years after Xeno- phon’s return to Greece, when he was settled at Scillus in the neighbourhood of Olympia; as he not only gives us a particular description of the grove and temple which he dedicated to Diana at that place, but speaks of his sons as grown-up youths of sufficient age to take an active part in the pleasures of the chase (Anal). v. 3, § 10), though he elsewhere speaks of himself as childless at the time he was with the army in Thrace, after the close of the expedition. (Ib. vii. 6, § 34.) It is however probable that the work may have been in great part composed long before, or at least that he may have committed to writing some brief commentaries concerning the events which he had witnessed, while they were still fresh in his memory. But the whole series of marches and distances traversed could hardly have been preserved otherwise than by being committed to Writing at the time; and there is certainly nothing improbable in such a supposition. NOTE B, p. 343. COMPUTATION OF DISTANCES IN PARASANGS. The Persian measure of the “parasang” appears to have been in use throughout the monarchy in the days of Xenophon, as well as those of Herodotus (see Chapter VII.); and the distances along the frequented highways were doubtless estimated in those days in parasangs, as they are at the present time along the post-roads '. 360 nrs'ronv or ANCIENT GEOGRAPHY. cm). x. J in farsangs or farsakhs. The origin of the name, as well as of the measure itself, is uncertain; but both Xenophon and Herodotus regard it as equivalent to 30 Greek stadia. (Xenoph. Anab. ii. 1, § 6; vii. 26. Herodot. ii. 6; v. 53.) Strabo however tells us that there was considerable discrepancy upon this point; some writers estimating it at 30, some at 40, and others again at not less than 60 stadia (xi. p. 518). He himself regards it, like the Egyptian schaznus, as of variable length. It was doubt- less in reality a mere estimate, or rough itinerary measure, as is still the case with the modern farsahh. Of the latter Colonel Chesney says: “ The modern farsang or farsakh of Persia varies according to the nature of the ground from 3.},- to 4 English miles per hour, and being almost always calculated for mules or good horses, under favourable circumstances it frequently exceeds 4 miles. The ancient parasang appears to have been fixed at 30 stadia, or 3 geographical miles. But this being also a road measure, it no doubt varied as at present, and was regulated ac- cording to the nature of the country.” (Chesney’s Euphrates and Tigris, vol. ii. p. 207.) ' From a comparison of the distances given by Xenophon on the march up the country from Sardis to Thapsacus—two fixed points known with certainty—Colonel Chesney deduces the average length of the parasang as equal to 26 G. miles; but several portions of the route would give only about 210 G. miles. Major Rennell estimated the parasang at 225 G. miles. Mr. Hamilton, computing only from distances in Asia Minor, reckons the parasang as some- thing less than .‘Zé— G. miles. It is clear that no exact result can be attained by any such method, as we neither know the precise line of the ancient routes, nor have We any exact measurements of the modern ones. But it seems clear that while the modern farsalrh generally ewceeds 3 G. miles, the ancient parasang on the average fell considerably short of that length. There is nothing in this to surprise us, or lead us to have recourse to any far-fetched hypothesis to explain it; the tendency to over-estimate distances in travelling being much more frequent than the contrary error. Even along the highways and great lines of route— such as the “royal road” described by Herodotus—there is no reason to suppose that the distances were “ measured and marked” as they unquestionably were along the Roman roads; any more than .we are to suppose the roads themselves to'have had any resemblance to those of the NOTE 0. THE ANABASIS or XENOPHON. 361 Romans. The distances were doubtless mere estimates, as they are at the present day along the so-called post-roads of Persia. Thus the route from Teheran to Tabreez—one of the most frequented in modern Persia—is divided into stages from one post-station to another, corresponding precisely with the omdnoi of Herodotus and Xenophon, and of just the same average length, varying from four to five and six farsakhs, though occasionally extending to seven or even eight. (Ussher’s Journey from London to Persepolis, pp. 647-662.) Mr. Ainsworth assumes the parasang to have been in all cases really equivalent to 30 stadia or 3 geographical miles; a funda- mental error, the eficect of which may be frequently traced through- out the whole of his work. Colonel Ghesney, while justly regarding it as a mere “road measure ”'supposes it to be, like the modern Persian farsakh, the equivalent of “an hour of time;” an assumption which may well be doubted. But even if it were so in its origin, and that the Persians had derived from the Babylonians the division of the day into twelve hours, it is certain that neither they, nor the Greeks, possessed any portable instruments for the measurement of time, and hence any attempt to estimate distances by this process must have been of the vaguest possible character. NOTE 0, p. 344. RATE OF MARCHING. Mr. Grote is almost the only writer who has made (as it appears to me) sufficient allowance for the full operation of these causes. Several of the modern geographical inquirers have been misled by a strong desire to reconcile or explain the distances given by Xenophon, without inquiring what probable foundation there could be for his statement of those distances. This defect is especially prominent in Mr. Ainsworth, in whom it is the more remarkable as, having himself travelled over a large part of the ground tra~ versed by the Greek‘ army, he would be naturally the more familiar with the great difficulties which it presented. Yet he uniformly translates the parasangs into miles, at the rate of 3 G. miles to a parasang, as confidently as if he were dealing with distances given in the Roman Itineraries, which had been really measured along 362 HISTORY OF ANCIENT GEOGRAPHY. CI-IAP. X. lines of high road. Hence he does not hesitate to describe the Greeks as marching forty-five (geographical) miles in three days, through deep snow—in some parts, according to Xenophon’s own statement, not less than six feet in depth. Such a march may be safely pronounced impossible: and the same criticism may be applied, with scarcely lesspositiveness, to many other portions of the march through this mountainous district. In one passage indeed Xenophon himself describes the army as marching for seven days at the rate of five parasangs a day (tea Toiiro e’rropeiifino-av (no.9- ,aoils sen-rd (ii/oi rre'v’re wapao'oiy'yas 'rr'is fque'pas, iV. 6, § 4): and this may serve to give some clue to the mode of his computation. It is pro- bable that he reckoned an average day’s march at five parasangs, and estimated the distances accordingly, without making sufficient allowance for disturbing causes. Such a mode of calculation is the more remarkable, as he himself gives five parasangs as the ordinary rate of daily march, while the army was ascending the valley of the Tigris, under the charge of Tissaphernes, so that they were. proceeding through a friendly, as well as a naturally easy, country (Anal). ii. 4, §§ 25—28). The first instance,iafter quitting the Tigris, in which he describes them as marching more than five para-' sangs a day, was through the country of the Taochi, where they were hard pressed for subsistence, in consequence of the barbarians having carried off all supplies into their strongholds, and they pro- bably therefore were induced to make forced marches in hopes of arriving at fresh villages. Hence they are here said to have marched 30 parasangs in five days, or 6 parasangs a day. Again, after obtaining fresh supplies by storming the fort of the Taochians, they are said to have accomplished not less than 50 parasangs in 7 days, though engaged in continual combats with the Chalybes, whom Xenophon describes as the most warlike people they met with (Auab. iv. 7, §15). On the other hand, from the Harpasus to Gymnias, though apparently unopposed, and travelling in part ~ through a level country (Sui wedt'ov), they resumed their customary rate of 5 parasangs a day: and after they had come in sight of the sea, where we might have supposed that they would be in haste to arrive at Trebizond, they advanced but 10 parasangs in three days through the land of the -Macronians, and 7 parasangs in two days through that of the Colchians. If any confirmation were wanting of the extreme vagueness and uncertainty of all estimates of distance under such circumstances, it will be abundantly supplied by a reference to the recent experience NOTE D. THE ANABASIS or XENOPHON. 363 of the English expedition into Abyssinia. Here the distances tra- versed by the army were afterwards actually measured with a chain, and it was found that a day’s march, estimated by experienced officers at 16 or 18 miles, often did not exceed eight! Both Major Rennell and Col. Chesney repeatedly advert to the difficulties encountered by the Greeks in this part of their retreat, and the impossibility of their having actually performed the dis- tances which Xenophon supposed them to have accomplished. Col. Chesney indeed admits that they could not have advanced more than a mile per hour through the snows of Armenia (vol. ii. p. 230): but he is misled by supposing the parasang to represent an hour’s march, forgetting that such a mode of computation was wholly unknown to Xenophon. NOTE I), p. 345. ROUTE OF CYRUS THROUGH PHRYGIA. This part of the route of Cyrus has been a source of much per- plexity to modern geographers, especially to the earlier writers on the subject. The researches of modern travellers have thrown considerable light upon it, but without altogether explaining the diificulty. (See Hamilton’s Asia Minor, vol. ii. pp. 198-204. Ains- worth’s Travels in the Track, pp. 24—35. Koch, Zug der Zelmtauserwl, p. 19.) It is clear that Cyrus, when he quitted Celaenae, for some reason which is not explained to us, instead of continuing his march direct towards Iconium, made a wide circuit, passing round the mountain group of Sultan Dagh, to the north, instead of taking the line of road to the south of it. By so doing he struck into a well-known and natural line of route, from the neighbourhood of Synnada (Afium Kara Hissar) to Iconium, and there can be no doubt that the towns of Thymbrium and Tyriaeum may be placed along this line, though their exact site is still subject to some doubt. The place called Caystri Pedion must therefore be sought in the neighbourhood of the small lake called Eber Ghiol : but the station before this, called by Xenophon Ceramon Agora, cannot be determined even approximately. The expression, however, that it was “ the last city on the side of M ysia ” (mihw OlKO'UfLél/flll, e’o'xotrnv 7171?); iii Mvm'a Xe’ipa, i. 2, § 10), points clearly to a situation towards the northern frontier of Phrygia (probably in the neighbourhood 364 HISTORY or ANCIENT GEOGRAPHY. crap. x. of Ushak) ; and it seems certain, therefore, that Cyrus, in starting from Celaenee, took in the first instance a direction somewhat to the westward of north, with which the presumed position of Peltae would coincide. The motive for this sudden deviation from the ordinary line of route is unknown to us; but it greatly exceeds what can be supposed to have been made without some special reason. He must also have made an equally sudden turn after visiting Ceramon Agora with the view of regaining his general direction towards the south-east. I concur with Mr. Grote in believing in the accuracy of Xenophon’s description of ;this part of the route, which lay through an open country, easily traversed by an army, and where the roads must have been Well known to the Persian officers. The difiiculty is, that the two most important points—Ceramon Agora and Caystri Pedion-——are not mentioned by any other writer: and though the names of Peltae, Thymbrium, and Tyriaeum are men- tioned by later geographers, it is in a manner that affords little assistance in determining their site. We have thus five stations along a line of march undoubtedly very circuitous—for it- occupied not less than 92 parasangs, while the direct route from Celaenaa to Iconium could not have exceeded 150 G. miles (or about 60 parasangs)—no one of which can be fixed with certainty. It may be added, that this sudden deviation towards the north had the effect of turning his back upon the Pisidians, and must therefore have at once betrayed the fact that his expedition against that people was a mere pretence. NOTE E, p. 346. PASSES BETWEEN CILICIA AND SYRIA. The topography of these passes has been carefully investigated by several modern writers—and Col. Chesney’s survey and map of the localities have thrown much light on the matter. The passes having been undefended, Xenophon does not seem to have given full attention to their difficulties. Thus, while he mentions the actual fortified gates between Issus and Myriandrus, to which he gives the name of the Cilician and Syrian gates, from their forming at this time the frontier of the two countries (i. 4, § 4), he does not notice either the pass through the western branch of Mount Amanus NOTES F, G. THE ANABASIS OF XENOPHON. 365 (the Amanian Gates of Strabo), by which Cyrus must have de- scended to the Gulf of Issus, or the narrow pass of Beilan, by which, after leaving Myriandrus, he must have crossed the main ridge of Mount Amanus to enter into Syria, strictly so called. Neither of these passes however presented difficulties comparable to those of the Pylse Ciliciae, of which Xenophon speaks in such strong terms. N ()TE F, p. 347. ROUTE FROM THE GULF OF ISSUS TO THE EUPHRATES. The stations and distances given, after leaving Myriandrus, are To the river Chalus .. .. .. .. 4 days, 20 parasangs. ,, sources of the river Daradax 5 ,, 3O ,, ,, river Euphrates at Thapsacus 3 ,, l5 ,7 Of these, the Chalus is evidently the river of Aleppo (itself called in later times Chalybon, whence the modern Haleb), but we cannot determine at what point the army crossed it: and the sources of the Daradax—also described as a considerable river, 100 feet in width, and where the Persian kings had a park—have not been identified, though that stream must certainly have been situated in the neighbourhood of the modern Balis. The position of Thap- sacus renders it almost certain that the army must have reached the Euphrates some time before halting at that city, and followed its course downwards: though the words of Xenophon (i. 4, § 11) would certainly in themselves have led us to suppose that this was the first point where they touched on the Euphrates. This is a circumstance worthy of note in its bearing on other disputed questions in regard to the line of march. NOTE e, p. 347. THAPSACUS. Major Rennell in his Illustrations of the Geography of Xenophon (p. 60), fell into the error (in common with D’Anville) of placing Thapsacus at the modern Deir, more than 100 miles lower down the river than its true position, and thus threw the whole geo- 366 ursronv or ANCIENT GEOGRAPHY. 01m. x. graphy of this part of the march into confusion. The true site was first determined by the regular- survey of the Euphrates under Colonel Chesney, when it was found that the Euphrates was ford- able at this point only. The existence of large mounds in the immediate neighbourhood points to the site of a city of importance, such as Thapsacus is described by Xenophon, who calls it “ a large and wealthy city” (mans ,ue-yo’tMy Kai aidatuwv, Ahab. i. 4, § 11): and such it continued to be during the greater part of the Macedonian period, but before the time of Strabo it had declined, and given way to the establishment of the customary passage at Zeugma (the modern Bir), more than 120 miles higher up the river. The Euphrates was only fordable at this point at certain seasons of the year: at other times it was crossed by a bridge of boats. It was in this manner that Alexander afterwards crossed it (Arrian, iii. 7): and under the Seleucidan kings there was a permanent bridge of boats (Zei‘iyna) established there (Strabo, xvi. 1, § 21, p. 746). When Cyrus crossed the river at this point the Thapsacenes asserted that the river had never been fordable before, and had only become so in honour of Cyrus (Ahab. i. 4, § 18): but it is evident that this was a mere piece of flattery, based upon the fact that it was subject to great fluctuations, and often impassable on foot. (See Ainsworth’s Travels in the Track, pp. 7 0-72 ; Chesney’s Euphrates and Tigris, vol. i. p. 416, vol. ii. p. 213.) It may be added that the river has here exactly the width reported by Xenophon of 4 stades or 800 yards (Ainsworth, l. c.). _., NOTE H, P. 366. POSITION OF PYLZE. The position of Pylae unfortunately cannot be determined with accuracy, though it is an essential point in order to arrange the details of the subsequent operations. Its name appears to point to some unexplained peculiarity of its situation, but all that we- learn from Xenophon is that it was apparently situated at the point where the barren, undulating, slightly upland country of Mesopotamia gives way to'the level and fertile alluvial plains of Babylonia. (Compare i. 5, §5, and 7, § 1.) This transition is described by all travellers as very strongly marked, and according to Mr. Ainsworth the point where it takes place in following the NOTE H. THE ANABASIS on XENOPHON. 367 course of the Euphrates is about 14 miles above the village of Felujah. (Ainsworth’s Travels, p. 81.) It is a few miles below that point that at the present day the Saklawiyeh—originally a mere artificial canal, but which has gradually become a tortuous river—leaves the Euphrates, and traverses the low country to join the Tigris below Baghdad. Here therefore Mr. Ainsworth fixes the site of Pylae; but it is impossible to reconcile this with the statements of the march from thence to Cunaxa; it was not till the sixth day after leaving Pylaa that the army of Cyrus came in sight of that of Artaxerxes, and though these were probably short marches, as Cyrus was advancing cautiously, expecting at any moment to meet the enemy, they could hardly be estimated at less than 60 miles. Now the site of the battle was, according to Xenophon 360 stadia (36 G. miles) from Babylon, or according to another account 500 stadia (see the next note), and this would place it not more than 30 miles south of Felujah, or less than 20 miles, if we adopt the statement of Ctesias. Hence Colonel Chesney would place Pylae considerably higher up the Euphrates; and Mr. Grote considers it to have been situated only a few miles below Hit, which is more than 50 G. miles above Felujah. This is confirmed by a statement of Captain Jones that there is at this point (9%- miles S.E. of Hit) a place called Bekaa, an Arabic word signifying the same as the Greek Pylae, and that there is actually a narrow pass at that place (Journal of Geogr. Soc. vol. xxxvii. p. 167). But it does not appear that there is at this point any such marked change in the character of the country as would correspond to the supposed situation of Pylze; and the distance from Felujah appears much too great. Hence the position assigned by Mr. Grote seems to me as much too high up the stream, as that of Mr. Ainsworth would bring it down too low. The position of Pylaa and that of Cunaxa are indeed to a con- siderable extent mutually dependent upon each other; and neither the one nor the other can be fixed on fully satisfactory data; but if the probable site for the battle, which will be considered in the next note, be adopted, Pylae can hardly be thrown so far back as it has been by Mr. Grote. Very little assistance can be derived from the statement of the distances in the other direction. From the mouth of the Araxes (the Khabour), which is the last fixed point that can be determined with certainty, the army of Cyrus advanced (according to Xenophon) (i. 5) five days’ march, amounting to 35 parasangs, through a desert country, at the end of which time 368 HISTORY OF ANCIENT GEOGRAPHY. omax. they arrived at a large deserted city called Corsolte, situated on the river Maskas, which was a hundred feet in width. But no such river is now found flowing into the Euphrates, and the site of Corsote cannot be identified. After laying in a stock of provisions here, they again proceeded through a barren country, affording no sustenance for either man or beast, for thirteen days’ march (reckoned at 90 parasangs) which brought them to Pylae (i. 5, § 5). We have thus an estimated distance of 125 parasangs, or 375 G. miles, with only one halting-place, which cannot be de- termined. The difliculties and privations of the march are described in a lively manner by Xenophon, and he tells us that the difficulty of procuring provisions led Cyrus to press on his advance with all possible speed, and to make forced marches. Hence we find them accomplishing 7 parasangs a day in the first stage of the march, and very nearly as much in the second, notwithstanding its long protracted duration. It may well be doubted whether both esti- mates are not materially over-rated. But besides this, the actual distance marched by an army, following the course of such a river as the Euphrates, cannot be judged without knowing how nearly they kept to it, and to what extent they were able to cut off its numerous windings. From the mouth of the Khabour to the Sakla- wiyeh is about 250 English miles, without following the minor sinuosities of the river; but the distance may be increased to a great extent, if we suppose the army to have kept as close as possible to its banks. In such a case therefore any attempt to fix the position of Pylae by reference to its distance from the Khabour would be utterly futile. The only other place mentioned by name in this part of the march is a city called by Xenophon Charmande, which he terms “ a large and wealthy city ” (we)... seesaw Kal. peyoiAr], i. 5,_§ 10), but which was situated on the other side of the Euphrates, on which account the army passed it by without halting, though they procured pro- visions from thence. For the same reason probably Xenophon has not mentioned its distance, stating only that it was passed during the march through the desert (Karol 1'01‘); e’pfiluovs o'rafiuoizs). The name is not found in any other author; but it certainly seems pro- bable (as suggested by Mr. Grote) that the city designated could be no other than the Is of Herodotus (i. 179), still called Hit, which on account of its bitumen springs, must always have been a place of some importance. But even if this identification be admitted, it throws no light on the other points in dispute connected with NOTE I. THE ANABASIS or XENOPHON. 369 this part of the march, as we are left in doubt as to the distance of Charmande from either Corsote or Pylee. NOTE I, p. 2349. SITE OF CUNAXA. This was the conclusion arrived at by Mr. Grote (vol. ix. p. 56, note 2), and appears preferable to the site fixed on by Colonel Chesney (followed by Mr. Ainsworth), who placed the field of battle near a spot marked on his map by the Mounds of Mohammed, at the mouth of a river or canal called the Kutha (Chesney, vol. ii. p. 217 ; Ainsworth’s Travels, pp. 81, 87 This site is about 20 miles farther south than that adopted by Mr. Grote, and by so much nearer Babylon; thus agreeing with the statement of Xenophon, rather than with that of Ctesias. But, as observed in the text, the testi- mony of Ctesias is in this case probably the best authority. Since Mr. Grote’s note was written, Lieutenant Bewsher, who has made a regular survey of this part of Babylonia, discovered that one of the mounds, which in this country always mark the sites of ancient habitation, is called Tel Kuneeseh; a remarkable resemblance to the ancient name. He himself indeed remarks that, “in a country where names of mounds are frequently changed—- except indeed the larger ones—it seems improbable, and almost too good to be true, that this one particular spot, the site of which has been so long sought for, should have kept its name intact for nearly 2300 years.” (Journal of Geogr. Soc. vol. xxxvii. p. 166.) It must be added that the circumstance which he mentions that Kuneeseh is the Arabic for “ church,” tends to weaken the force of the inference from the coincidence of name; but on the other hand the position of the mound in question, which is situated just to the south of the canal known as the Abu Gharraib, 17 miles from Felujah, and 51 in a direct line from Babel, the northernmost of the ruins of Babylon (Bewsher, l. 0.), agrees so well with the requirements of our historical data, as to lend a strong confirma- tion to the evidence of the name, and to leave little doubt that the plain between Tel Kuneeseh and the Euphrates—which has a width varying from 3 to 4 miles ——was really the scene of the far- famed battle of Cunaxa. (See the map.) VOL. I. :2 n 370 HISTORY OF ANCIENT GEOGRAPHY. CIIAP. X. NOTE K, p.349. SITTACE. The disappearance of cities that have attained to considerable prosperity and opulence is a remarkable fact, of which there are many instances in these countries. Thus the site of Kufah—the original seat of the Caliphs, until their capital was transferred to Baghdad in 7 63—is barely to be identified: some insignificant heaps of pebbles scattered over the plain alone serving to indicate the site of what was once a great city (Ussher, Journey from London to Persepolis, p. 465). Even of the opulent Seleucia, which in the time of Strabo contained 600,000 inhabitants and doubtless abounded in splendid buildings, nothing remains but long lines of earthen ramparts, inclosing the mere vestiges of former habitation (Layard, Nineveh and Babylon, p. 571). We cannot therefore wonder at the disappearance of all trace. of the Sittace of Xenophon, which must have been situated in the neighbourhood of Baghdad, pro- bably between that city and the site of Seleucia. It was certainly above the point of confluence of the Diyala, otherwise the Greeks would have had to pass that river, after crossing the Tigris of which there is no mention. The distance of 20 parasangs from Opis affords us no assistance, as the site of Opis itself is still uncertain. The name of Sittace as a city is not found in later writers, but it must undoubtedly have been the capital which gave name to the province of Sittacene, and this, as we learn distinctly from Strabo, was situated between Susiana and Babylonia, on the east of the Tigris (Strabo, xv. 3, p. 732, xvi. 1,p. 744) though Sittace itself, as is plain from the narrative of Xenophon, was on the western or Babylonian side of the river (Auab. ii. 4, §§ 13—24). NOTE L, p. 34.9. THE MEDIAN WALL. Previous to this discovery by Lieut. Bewsher of the remains of the wall in question, which may be considered as completely cor- responding with what is required by the narrative of Xenophon, it was supposed by many writers that the Median hVall must have been a barrier across from the Euphrates to the Tigris, such NOTE L. THE ANABASIS OF XENOPHON. 371 as we find certainly existing at a later period. But this wall or rampart, the remains of which are still visible, and are known to the Arabs as Khalu Nimrud, or Sidd himrud, leaves the Tigris more than 50 miles above Baghdad, extending in a S.\V. direction towards the Euphrates, and would, if prolonged to that river, strike it about 3 miles above the Saklawiyeh canal, and at ‘least 8 above Felujah. (See the description of it in the Journal of Geographical Society, vol. ix. pp. 445—6, 473-4.) It was evidently this wall, of which the remains, already in a half-ruined state (“ semiruta murorum vestigia ”) are noticed during the advance of the Emperor Julian down the valley of the Tigris (Ammian. Marcell. xxiv. 2, § 6). The objections to such a theory were insuperable ; for it was utterly unintelligible that the Greeks should have re- turned so far northward, after the battle; and if they had passed through this wall, instead of bringing them towards the Tigris, they would have emerged into the barren stony plains of Meso- potamia, the Sidd Nimrud marking exactly the limit between these and the rich alluvial tract of Babylonia. Moreover Xenophon describes the army as passing, on the third day before the battle, a vast trench dug by order of Artaxerxes in order to impede the advance of Cyrus, but which after all he left undefended: and this trench, he tells us, was carried through the plain for a distance of 12 parasangs to the Median u'all (i. 7, § 15). Such a line of defence is unintelligible, if the wall was‘a rampart extending across from the one river to the other: but would be an obvious ex- pedient, if the wall had a direction obliquely through the centre of Babylonia, like the rampart of which the ruined remains were discovered by Lieut. Bewsher. On the other hand Strabo mentions a wall, which he calls “ the \Vall of Semiramis” (To Eentpcintdos 8ia'ret'XLo-na), which appears to have extended from the Euphrates to the Tigris at the point where they approached the most closely to one another (Strabo, ii. p. 80, xi. p. 529) ; and this has been supposed by Major Rennell and lllr. Grote to be the Median \Vall of Xenophon. But in the first place no trace remains of such a bulwark, which, if it really existed at the narrowest part between the two rivers, must have been situated near the modern city of Baghdad, between Cunaxa and Babylon: and moreover the account given by Strabo (from Eratos- thenes) is by no means clear, but seems to place this narrowest point at‘ Opis on the Tigris, which must certainly have been situated considerably farther to the north than the real neck of the‘ 2 B 2 372 ms'ronv or ANCIENT GEOGRAPHY. carp. isthmus. It may well be doubted therefore whether there is not a misconception on the subject, and whether the Wall of Semiramis (of Strabo and Eratosthenes) was not in reality the same of which the remains are still called Sidd Nimrud (the Wall of Nimrud). But even if there ever really existed such a line of wall as that mentioned by ‘Strabo, at the point where the two rivers approach within less than 20 miles of each other, it would not have corre- sponded near so well with the narrative of Xenophon as the wall discovered by Lieut. Bewsher, which must have lain directly across the route of an army proceeding from the Euphrates near Cunaxa to Sittace on the Tigris, whether we place that city below or above Baghdad. , It may be added that the remains of the Hubl es Sukhr show that wall to have been built of bricks cemented with bitumen, in accordance with the description of the Median \Vall in Xenophon (ii. 4, § 12), while the Sidd Nimrud is built of “the small pebbles of the country imbedded in cement of lime.” (Journal of Geogr. Soc. vol. ix. p. 446.) It would be obviously idle to attempt to identify the great ditch dug by order of Artaxerxes, any more than the two canals that were crossed by the Greeks on their way from the Median Wall to the Tigris. Such canals have been in all ages out for the purposes either of irrigation or internal communication: and when neglected readily assume in the course of time the aspect of natural rivers. Such is at the present time the Saklawiyeh, which is still navigable for a small steamer, and such was in the middle ages the Nahr Malcha, or Royal River, which is now dry in the main part of its course. A glance at the map given by Lieut. Bewsher (Journal of Geogr. Soc. vol. xxvii.) will sufficiently show by what a com- plicated network of canals and artificial streams all this part of Babylonia is intersected: the greater part of which date from the period of the Caliphs of Baghdad—and have consequently tended to destroy all possibility of tracing its condition in ancient times. NOTE M, p. 350. MARCH ALONG THE TIGRIS. The stages and distances given by Xenophon (ii. 4, § 25) after crossing the Tigris are as follows: .. pis - , // /// r I, ‘Ar/‘NZ?’ , /' A. l / / .>. 0.1.7:. / , g {0/4 / / 1:», ‘I / a "E . ll / {I f.” t , Qt? i 5 ‘Br’, 08 . lb ' /' q l / Q5. q Q m B p w ‘5 I'lklawl'yeb Canal / I, / \ F17 . f? 0 1111572111 ‘ TBS c". . Selew'ja 6! MAP To ILLUSTRATE .. . THE OPERATIONS OF THE GREEKS BEFORE&AF'I'ER THE ' London : John Murray. BATTLE OF CUNAXA __ o ,_ I ‘ .Smnfi'ra'ls Geo ‘gr-up 1715mm}: ." NOTE M. , THE ANABASIS OF XENOPHON. From the passage of the Tigris near Sittace, to the river Physcus and the city of Opis 4 days, 20 parasangs. thence through Media to the villages of Parysatis .. .. .. .. .. .. 6 ,, 3O ,, along the left bank of the Tigris to the river Zapatas .. .. .. .. .. 4 ,, 20 ,, The last ten days’ march lay through an uninhabited country (masses? Epifqaovs), with the exception of the villages of Parysatis, where they halted, but on the first day’s march from these villages there was a large and wealthy city named Caenae (Kan/at’) on the other side of the river, from which the Greeks received abundant supplies. Unfortunately neither Caenee nor Opis can be fixed with any certainty. Captain Lynch indeed considered some ruins that he had discovered near the junction of the Adhem with the Tigris to be those of Opis, in which case the river Adhem would represent the Physcus of Xenophon (Geogr. Journal, vol. ix. p. 472), and this view‘ is adopted by Col. Chesney (Euphrates and Tigris, vol. i. p. 30): but the identification is far from certain. Opis is again mentioned during the campaigns of Alexander, in whose time it was still a large and important city, as it is described by Xeno— phon, though Strabo speaks of it as a mere village (ii. 1, p. 80). But we have no other clue to its position, except that Eratosthenes (ap. Strab. l. 0.) apparently designates it as marking one extremity of the Wall of Semiramis, a statement which can hardly be reconciled with the position assigned to it by Captain Lynch. Caenae, of which the name is not found in any other writer, has been gene- rally identified with the modern town of Senn; but the resem- blance of name, as pointed out by Mr. Grote (vol. ix. p. 93, note), is really a delusion; and the situation of Senn, just opposite to the mouth of the Greater Zab, is certainly at variance with that assigned by Xenophon to Caenae, which he places distinctly on the first day’s march after leaving the villages of Parysatis, It would seem much more probable that Ceenae occupied the site now marked by the ruins at Kalah Shergat, which have been recently ex- plored by Mr. Layard (Nineveh and its Remains, vol. i. p. 5; vol. ii. pp. 45—60). It is true that the existing remains appear all to belong to the ancient Assyrian city, but it is not at all improbable that it may have continued to be inhabited at a later period. At all events the Caenae of Xenophon must have been situated some- where in its immediate neighbourhood. 374 HISTORY or ANCIENT GEOGRAPHY. 01m». x‘. Another ditficultyarises, in regard to this part of the march, from the omission of all mention of the Lesser Zab, an important stream, which the Greeks must necessarily have crossed. Hero- dotus distinctly mentions the two riversof the same name, both of which had to be crossed in following the line of the “ royal road,” and the passage of such rivers ‘is one of the points in general most studiously noted by Xenophon. It is impossible not to suspect that. in this instance Xenophon’s memory had played him false, and that he had forgotten the circumstance that there were two rivers of the name of Zabatus or Zapatas, both of which had been crossed by the army. In this case the villages of Parysatis were probably situated near the mouth of the Lesser Zab ; the distance between the two rivers being 57 miles, which would just correspond with the 20 parasan gs of Xenophon. NOTE N, p. 350. RUINED CITIES OF ASSYRIA. It was on this part of their march, while the Greeks were ad- vancing through the open country on the left bank of the Tigris, that they passed the two deserted cities named by Xenophon Larissa and Mespila, the gigantic ruins of which seem to have forcibly attracted his attention (Anal). iii. 4, § 7-12). These may be safely identified with the sites of Nimroud and Kouyunjik, which have been rendered so familiar to modern readers by the recent researches of Mr. Layard. But no plausible explanation can be given of the names by which they are here designated, and which are unknown to any other ancient writer. The curious legends which Xeno- phon relates concerning them, and which he obviously heard on the spot, show how entirely all trace of their true history and origin had already been lost in his time. 1: NOTE 0, p. 353. SOURCES OF THE TIGRIS. The Tigris may be considered, like the Euphrates, as formed by the junction of two principal streams ; of which the western branch NOTE P_ THE ANABASIS or XENOPHON. 375 is the more considerable, and is regarded by modern geographers as the true Tigris. This takes its rise in the mountains of Ar- menia, about 40 miles N.W. of Diarbekr, and within a very few miles of the nearest point on the Euphrates. But the stream which joins the river of Diarbekr about 80 miles (in a direct line) below that town is so important a tributary, that it is considered by many geographers as constituting what may be called the Eastern Tigris. This arm is itself formed by the junction cf two others: the Buhtan Cha'i, sometimes also called the river of Sert—which is clearly the Centrites of Xenophon: and the Bitlis Chat, or river of Bitlis, which he appears to have regarded as the true Tigris. This flows from N. to S. and has its sources within a few miles of the Lake of Van: while the Buhtan Cha'i' which flows from the E. rises in the mountains of Kurdistan, in a tract that has only recently been visited by any modern traveller. NOTE P, p. 375. DESCENT THROUGH ARMENIA TO TREBIZOND. It may be worth while to add a few remarks on the geography of this part of the route, though there seems little hope that the difficulties with which it is perplexed will ever be wholly explained. Yet it may not be impossible at least to clear some of them away. Major Rennell was, I believe, the first to identify the Harpasus of Xenophon (which he describes as a large river 400 feet in width) with the stream now called Arpa-su, or Arpa-chai', a northern tributary of the Araxes, and this view has been adopted by Mr. Ainsworth, as well as by most modern geographers. But the sug- gestion of Colonel Chesney and of Dr. Koch (which has been followed by Mr. Grote) that the Harpasus was in reality the Tchoruk Su, a large stream flowing through a deep valley which separates the mountains of Armenia from the ranges adjoining the Euxine (Koch, p. 201), is in reality much more plausible, as that river lay directly in the course of the Greeks towards the sea, while in order to reach the Arpa-chai we must suppose them to have made a great olétour towards the N.E., for which there was really no occasion. Mr. Ainsworth indeed carries them still farther to the north, into the heart of Georgia, a distance of 90 miles (I), and then brings them back again no less than 150 miles through 376 HISTORY or ANCIENT GEOGRAPHY. can). x. the country of the Chalybes to the frontier of the Scythini (Travels, p. 183), the whole of this enormous détour being required (as he himself confesses) only in order to explain the distances given by Xenophon in parasangs of three miles each. Dr. Koch on the other hand, while justly rejecting the theory of Mr. Ainsworth, supposes the Greeks, after reaching the Harpasus (Tchoruk Sn) somewhere in the neighbourhood of the modern Ispir, to have wandered away to the eastward as far as Ardahan, and then returned to the same river in the neighbourhood of Ardvin, a supposition equally improbable and uncalled for. The position of Gymnias would be the key to the whole question, if it could be ascertained; but unfortunately it is wholly uncertain; and the different sites that have been suggested are merely suited to the different views of modern writers as to the route of the army. It has thus been placed by Major Rennell at Comasour, by Mr. Ainsworth at Erzeroum, by Dr. Koch in the SNV. of Georgia; while Mr. Grote inclines to identify it with Giimisch Khana, a place on the high road from Trebizond to Erzeroum. But Giimisch Khana is only two or three days’ journey from Trebizond, and less‘ than 40 G. miles in a straight line; and it seems impossible to account for the ten days employed on the march between this point and the sea. Giimisch Khana is at the present day a place of importance on account of its silver mines, which, according to Mr. Hamilton, are “ the richest and most important in the Ottoman dominion” (Hamilton’s Asia Minor, vol. i. p. 234), and Mr. Grote considers these mines as affording a plausible explanation of the existence of a “great and flourishing city” in the midst of sur- rounding barbarians (vol. ix. p. 161). But there is no evidence that these mines were worked in ancient times, and had they been so, within so short a distance of the Greek colonies on the Euxine, it seems diificult to believe that some notice of the fact would not have been found in some subsequent writer, especially Strabo, who was himself a native of the neighbouring province of Pontus. Yet no mention either of Gymnias or of its silver mines is found in any later writer, excepting Diodorus, who evidently copied Xenophon, though he writes the name Gynmasia (Died. xiv. 29). The determination of the mountain of Theches, from whence the Greeks first caught sight of the sea, must obviously depend upon that of the route by which they were approaching the Euxine. In a country traversed by numerous ranges of high mountains—for NOTE P. THE ANABASIS or XENOPHON. 377 the most part, as will be seen by a reference to Kieper't's map, running in a direction parallel with the coast—there must be many points from which a distant view of the sea would suddenly burst upon the eyes of a traveller, and wherever this first took place, under the peculiar circumstances of the Greek army, the sight would be almost equally impressive. On the modern road from Erzeroum to Trebizond, the sea is first seen from a place called Karakaban, about 25 miles from the sea, and at an elevation of between 5000 and 6000 feet; and is described by all travellers as extremely striking. But it seems impossible to account for the time employed by the Greeks on the descent to Trebizond, as well as for the details of their march thither, if we suppose this to be the point whence they first descried the Euxine. (See Mr. Hamilton’s remarks, Asia Minor, vol. i. p. 166; and those of Mr. Grote, vol. ix. p. 162.) Mr. Ainsworth and Colonel Chesney transfer the scene much farther inland, to the mountains known as Kop Dagh and Gaur (or Gjaur) Dagh, forming part of the second, and more inland, chain, south of the valley of the Tchoruk; and the testimony of Colonel Chesney that he himself saw the sea in 1831 from the summit of Gaur Dagh, which is distant from it nearly 60 G. miles in a direct line (vol. ii. p. 232), shows at least that these inner ranges cannot be excluded from consideration. But either of these points seems as much too far inland, as Karakaban is too near the sea. Mr. Hamilton is inclined to suggest that they first came in sight of the sea from some point in the range of mountains extending from Ispir to Baiburt; considerably to the east of the modern line of road to Trebizond (l.c. p. 167), and the same view has been adopted by M. Kiepert in his remarks on a paper by M. Strecker. He supposes the Greeks to have crossed the mountains from Gymnias,—which he places in the neighbourhood of Baiburt, though not actually on the site of the modern city—by a pass leading into the valley of the Siirmeneh, and to have descended that valley to the sea, which they would thus reach some distance east of Trebizond. This suggestion appears highly plausible; but the route in question has not yet been followed by any modern traveller, and we cannot therefore determine fully how far it answers the required conditions. ()n the other hand the theory of a recent inquirer who believes that he has found the remains of the cairn of stones thrown up by the Greeks to mark the spot, on the summit of a mountain just to 378 HISTORY OF ANCIENT GEOGRAPHY. CHAP. X. the south of the pass called Vavough Dagh, which is crossed by the road from Baiburt to Giimiseh Khana (Journal of Geogr. Soc. vol. xl. p. 463) may be almost certainly dismissed as untenable. Xeno- phon’s account clearly implies that the place from which the sea was first descried lay on the line of march—not that they climbed a peak for the purpose. No general in his senses would have led his whole army up a mountain, for the mere purpose of seeing the sea, when he could himself have ascertained the fact of its proximity by ascending the summit with a reconnoitring party. The same objection applies still more strongly to the suggestion of M. Strecker that the Greeks ascended the Kolat Dagh, and after gazing on the sea from thence, descended again on the south side and made a great détom' to the west before resuming the direct road to Trebizond. The memoir of M. Strecker just cited (Bez'trc'ige zur Geographz'schen Erklo'irung des Rz'éckzuges der Zehntausend durch das Armem'sche Hock- land, Berlin, 1870), together with the counter remarks of M. Kiepert, was unknown to me when the greater part of the above note was written. It was with great satisfaction that I found myself in accord with the eminent geographer of Berlin in respect to several of the leading points in dispute; especially the identification of the Phasis with the upper course of the Araxes, and of the Harpasus with the Tchoruk Su. If these points can be considered as suffi- ciently established, the only points of real interest that remain to be determined are, the site of Gymnias, and the spot from whence the Greeks first caught sight of the sea. (379) CHAPTER XI. WRITERS AFTER XENOPHON. ‘SECTION 1.—-Eph0rus. The Periplus of Sag/law. § 1. AMONG the historical writers who flourished in the period between the expedition of Cyrus and that of Alexander (3.0. 399-334) the one who would have contributed the most to our geographical knowledge, had his writings been preserved to us, is undoubtedly EPHoRUs. A native of Cyme in ZEolis, he composed a great historical work, in thirty books, which may fairly be regarded as the first attempt at a general history.1 In this he endeavoured to comprehend the history of the barbarian nations, as well as that of the Greeks, though the latter of course constituted his main subject; and was related by him in detail from the return of the Heraclidze to the siege of Perinthus by Philip of Macedon in no. 340.2 His merits as an historian it hardly falls within our province to consider; but it may be observed that as in relating the earliest periods of Greek history he seems to have given par- ticular attention to the origin and foundation of the several cities, he bestowed equal pains upon those of the later Greek colonies.3 Hence we find him frequently cited as an authority upon these subjects by Strabo and other writers, and there is no doubt that many other statements are derived from the same source, although his name is not quoted. His work would therefore have supplied us, had it been preserved, with the most important contributions to that interesting subject, 1 Polyb. v. See Mure’s Hist. of Greek Literature, vol. v. p. 531. '-’ Diodor. iv. 1.; xvi.76. 3' Polyb. ix. 1; xxxiv. 1. 380 HISTORY or ANCIENT GEOGRAPHY. CHAP. x1, a history of the Greek colonies!‘ But besides such incidental notices of geographical facts, it appears that two whole books of his voluminous work were devoted to a general and sys- tematic review of geography: on which account the fourth book is frequently quoted under the title of “Europe ” or the Description of Europe : the fifth in like manner as that of Asia and Africa. Unfortunately our extant remains of these books are so imperfect that they afford us hardly any means of judging how far his geographical knowledge was in advance of that of Herodotus. By far the longest extracts that have been preserved to us, relate exclusively to the cities and institutions of Greece itself, and hence, however valuable in themselves, they have no geographical interest.5 § 2. Of his general views of geography we learn only that he regarded the four most distant regions of the earth as occupied respectively by the Indians to the east, the Ethiopians to the south, the Scythians to the north, and the Kelts to the west.6 The latter people he considered as occupying all Spain as well as Gaul, and extending to the Pillars of Hercules, and even to Gadeira.7 For his ignorance in this respect he _. is sharply censured by later writers: and it is indeed a curious proof how vague and imperfect were the notions still entertained by the Greeks of the western regions of Europe. At the same time he seems to have been aware that the tracts occupied by the four nations were not of equal extent, but that the ZEthiopians and the Scythians extended the farthest in space, so that he already regarded the inhabited world as an ob- ‘ See especially Strabo, ix. p. 422; and Scymn. Chius, v. 115. The latter author (or rather the anonymous writer of the geographical fragment commonly bearing his name) appears to have in great measure fol- lowed the authority of Ephorus, and his statements concerning the foun- dation of the Greek colonies, derived from that author, are among the most valuable parts of his work. 5 The fragments of Ephorus were first collected by Marx ; they have been more recently published by G. Miiller in his Fragmenta Hz'ston'corum Graecorum, tom. i. (Paris, Didot, 184i). They are here cited according to this last edition. 6 Fragm. 38, M. ap. Strabon. i. p. 34. The same statement is evidently copied from him by Scymnus Chins, vv. 170— 177, but without citing him by name. 7 Fr. 43, ap. Strab. iv. 4, p. 199. Seer. 1. WRITERS AFTER xENoPHoN. 381 long parallelogram, having its greatest length from east to west. § 3. Of the real extent of his knowledge towards the north we have no clear evidence, but he seems to have described the Scythianfitribes in considerable detail, and the following sum- mary of his information concerning them is extracted by a later writer: “Next beyond the Ister came the tribe of the Carpidse, beyond them the Aroteres (evidently an agricultural tribe), and farther north the Neuri, extending to the limits where the land was rendered uninhabitable by frost. Towards the east again, crossing the Borysthenes, came the Scythians who inhabited the land of Hybla (probably the Hylaea or forest country of Herodotus). Beyond these inland were the Georgi (or agricultural Scythians) and beyond them an ex- tensive tract of desert country: after passing which came the Androphagi (or man-eating Scythians), and beyond them again a vast desert. Crossing the Panticapes e. proceeding east- ward again, as he had done in the first instance from the Borysthenes) came the tribe of the Limnaeans (or Lake- dwellers), and many other tribes not distinguished by separate names, but known by the general appellation of N omades, very righteous people, so that they would not injure any living thing, but living in waggons, and subsisting on milk the produce of their mares.8 Other Scythian tribes had crossed over into Asia, where they were known by the name of Sacm. The most distinguished of the Scythian races were the Sauro- matae, the Geloni, and the Agathyrsi. The Mseoti gave name to the Lake Maeotis, into which the Tanai's flowed. That river had its source in a lake, the limit of which was un- known, and flowed by two mouths into the Lake Maaotis.”9 In another passage, preserved to us by Strabo,1 he pointed out that the manners both of the Scythians and Sarmatians varied greatly in the different tribes: some of them being savage to 8 This description is evidently de- ’ 9 Ephori Fragm. 78, ed. Miiller; rived from Homer, whose expressions Scymn. Ch. v. 841-873, ed. M iiller. are almost literallyeopied. ‘ 1 vii. 3, p. 302. 382 HISTORY or ANCIENT GEOGRAPHY. can. x1. such a degree as to feed on human flesh, others on the con- trary refraining from all animal food, and subsisting entirely on the milk of their mares, leading a life of such simplicity and innocence as to deserve the epithet of Homer, that they were “ the justest of mankind.” If we compare this account with that of Herodotus, it is evident that the knowledge of the interior of Scythia had made little or no progress in the interval, while the arrangement and description of the tribes by Ephorus (at least in the form that it is transmitted to us) is much less clear and intelligible than that of the earlier historian. It must however be admitted that it is preserved to us only at second hand, and in a somewhat fragmentary form. §4. The numerous quotations from Ephorus concerning the geography of Asia relate almost exclusively to the cities and inhabitants of Asia Minor, with which the Greeks were already familiar in his day, and cannot be said to add any- thing to our geographical knowledge. He described the peninsula of Asia Minor as inhabited by sixteen different nations: three of them of Greek origin, the ZEolians, Ionians, and Dorians, and the remaining thirteen barbarians : the Cili- cians, Pamphylians, Lycians, Garians, Bithynians, Paphlago- nians, Maryandini, and Trojans on the coast, and the Pisidians, Mysians, Chalybes, Phrygians, and Milyans in the interior.2 I: this enumeration the omission of the Lydians is unaccount- able, and perhaps merely accidental; but it deserves notice as an attempt to deal with the complicated question of the ethno- graphy of the peninsula. In this respect it is curious to compare it with the similar enumerations in Herodotus. § 5., With’ regard to Africa on the contrary it is interesting to observe that he appears certainly to have known, and made use of, the voyage of Hanno: as he was not only familiar with the name of Cerne, but mentioned also that of a town called Garicon 'I‘eichos,3 which is known only as one of the colonies 1 2 Fr. 80, ap.. Strab. 'xiv. 5, p. 678. siders that the voyage of Hanno was 3 Ephori Fragm. 96 ; Steph. Byz. v. known to Pliny only through the work Kapucbv Ts'ixos‘. C. Miiller even con- of Ephorus. SECT. 1. wuITEns AFTER XENOPHON. 38 3 founded by Hanno .on the west coast of Africa. He evidently considered Africa as surrounded by the ocean, but stated that it was impossible to navigate from the Erythraean Sea to Cerne on account of the excessive heat.4 He must therefore have rejected the story of the circumnavigation under N echo. § 6. The only specimen that has been preserved to us of the speculations of Ephorus on physical geography is very un- favourable. In regard to the much disputed question of the cause of the inundation of the Nile, he rejected the hypothesis that it was the result either of melting snow or rains in the upper part of its course, and considered it as owing to the spongy and porous nature of the soil of Egypt, which gave forth in summer, under the influence of the great heat, the moisture it had previously imbibed, and thus filled the river.5 Of all the solutions of the question that had been propounded, as Diodorus observes, this was the farthest from the truth, and showed the most complete ignorance of the country. §7. The important historical work of THEOPOMPUS, a con- temporary and fellow-pupil of Ephorus,6 would apparently have been of comparatively little geographical interest. It is however deserving of notice that he was, according to Pliny, the first Greek writer who mentioned the name of Rome, having recorded the capture of that city by the Gauls.7 He appears also to have supplied additional information con- 4 Plin. vi. 36, § 199. “Propter ar- I them survived the death of Philip, dores.” These words, as they stand - BC. 336. Theopompus indeed appears in Pliny, can hardly be understood in to have been still living as late as any other sense; but I cannot help ‘ B.C. 305. (See Clinton, F. H. vol. ii. suspecting that Pliny has made some i p. 374.) confusion with respect to the great fires 7 Plin. H. N. iii. 5, § 57 (nam Theo- and streams of fire of which the Cartha- pompus, ante quem nemo mentionem ginian voyagers had related so much, habuit, Urbem dumtaxat a Gallis cap- on the western coast of Africa. tam dixit). He had apparently over- 5 Ephori Fragm. 108 ; Diodor. i. 37, looked the obscure notices of the name § 4, 39, §7-13. of Rome by Antiochus of Syracuse 6 According to Suidas they were and Damastes already cited. But both born in the same year, but his Theopompus appears to have been date is probably erroneous. It is cer- really the first writer who mentioned tain, however. that they were fellow- any historical fact in connection with pupils under Isocrates; and both of the city. 384 _ > HISTORY OF ANCIENT GEOGRAPHY. CHAP. XI.‘ cerning the nations of Italy in general, especially the Vene- tians, the Umbrians, and other nations who bordered on the‘ Adriatic. The foundation of ‘Greek colonies at Ancona, and in the islands of Issa and Pharos, during the reign of the elder Dionysius at Syracuse, must have naturally led to increased knowledge of the countries surrounding that sea. Theopompus also is cited as authority for various details concerning the manners of the Tyrrhenians (Etruscans),8 which undoubtedly present a very exaggerated picture of the luxury and profii- gacy of their habits: but may nevertheless be taken as showing increased familiarity with that people. But the imperfect character of his purely geographical knowledge is sufficiently shown by his maintaining that the Danube communicated by one of its mouths directly with the Adriatic.9 It is singular that this erroneous notion, which is not found in any earlier writer, had at this time become a fixed idea among the Greeks, of which, as we shall see, they were very slow to divest themselves. §8. To the same period belongs the Periplus still extant under the name of SOYLAX, which is of interest as the first pro- fessed geographical treatise that has been preserved to us. It is a specimen of a class of works that seems to have been common in antiquity, professing to describe in regular order the coasts of the countries known to the Greeks, without any general geographical survey of the countries themselves, but merely giving very succinctly the names of the maritime cities and towns, the headlands and rivers, in the order of their occur- rence, as they would present themselves to any one sailing along the coast, with the addition, in more or less detail, of the’ distances from point to point. The deficiencies of such a work, as compared with a regular treatise on geography, would be much less glaring, in the case of Greece and the adjoining lands than in any other instance, and as the Greek colonies were, almost without exception, situated- on the sea-coast, all 8 Fragm. 222 M.; ap. Athenaeum, xii. p. 517. 9 Theopomp. ap. Strab. vii. 5, p. 317 ; Fragm. 140 M. Snow. 1. ' PERIPLUS or serum. 385 that was most interesting to the ordinary Greek mind would find its place in a Periplus such as we are describing. Towns’ and tribes in the interior were occasionally noticed, as it were in passing; in connexion with those of each district on the sea-coast. But this is rarely done except in the case of Greek cities. § 9. The date of the Periplus now extant has been a matter of much controversy. It is ascribed in the extant MSS. to Scylax of Caryanda, and was on that account supposed by the earlier editors to be the work of the navigator of that name, whose voyage down the Indus and from thence to the Arabian Gulf is recorded by Herodotus.1L But the internal evidence of its belonging to a period long after the reign of Darius Hys- taspis is conclusive: and N iebuhr was the first to show that it could not have been written before the time of Philip of Macedon. Recent editors have attempted to fix its date within more precise limits: but at all events it may fairly be con- sidered as contemporary with the histories of Theopompus and Ephorus.2 § 10. From its plan and arrangement it is not calculated to throw much light upon the extent or limits of geographical knowledge among the Greeks, being confined for the most part to the regions best known to them—those bordering on the Mediterranean Sea. The author begins from the Pillars of Hercules, and follows the northern coast of the Mediterranean from thence (including the Adriatic and the Euxine) as far as the mouth of the Tanai's, which he regards as the boundary between Europe and Asia: and from thence he returns along the coasts of Asia and Africa to the point from which he started: adding however a brief notice of the western or Atlantic coast of Africa, as far as the island of Cerne. Of the western shores of Europe, on the contrary, he seems to have known almost nothing. After briefly mentioning Gadeira (Gades) he adds that, outside the Pillars of Hercules, there 1 Herodot. iv. 44. See above, Chap. VII. p. 227. 2 Note A, p. 404. VOL. I. 2 c 386 HISTORY or ANCIENT GEOGRAPHY. CHAP. XI. are “ many trading stations of the Carthaginians, and much mud, and high tides and open seas.” 3 It is evident that these seas were never. at this time visited by Greek traders, while the confused notions of the obstacles to their navigation, pur- posely diffused by the Carthaginians, were all that had reached our author’s ears. Even of the eastern coast of Spain his information is ex- tremely imperfect, though this defect perhaps arises from the fault of our manuscripts. In its present form the Periplus only begins a regular description from the mouth of the 'Iberus (Ebro), though the author distinctly tells us that the whole extent of the coast of Spain (Iberia) was a voyage of seven days and seven nights: a very fair approximation to the truth.4 His account of the other nations that bordered on this part of the Mediterranean is very brief and summary, nor is it easy to reconcile with what we know from other sources of the real history of the nations mentioned. Thus he describes the tract from the confines of Iberia proper 5 to the mouth of the Rhone as occupied by mixed races of Iberians and Ligu- rians; and the Ligurians alone as inhabiting the region from the mouth of the Rhone as far as Antipolis (Antibes); in which were the Greek cities of Massilia and its colonies ; while he assigns the whole coast from thence to the confines of -3 ’A1rb ‘HpcucAeiwv O'TnMIW 7631/ £11 'rfi Ez’Jpcé-lrp e’pnrdpza ROAM}. Kapxndoviwv Ital 'mfltbs real aflmuuvpides Kai 'neM'ryn. Peripl. § 1. . His imperfect acquaintance with these regions is further shown by his statement (1.0.) that the strait sepa— rating the Pillars of Hercules was a day’s voyage across: an error which appears to result from his confounding the actual straits with the entrance to them between Gades and the opposite point of Africa—a distance which is estimated by several writers at a day’s voyage. 4 He reckons (as we shall presently see) a day’s voyage as equal to 500 stadia (50 G. miles), and a night as equal to a day. Hence his estimate of the paraplus of the coast of Spain I would give 700 G. miles; it really amounts to about 660, as far as Em- poriaa, or near 680, if measured to the Pyren aean Promontory (Capo di Creus). 5 He nowhere defines exactly the point which he considers as the boun- dary of Iberia, but he evidently seems to regard Emporiae (Ampurias) as the last Greek city in Spain: and there can therefore be no doubt of his adopt— ing the natural boundary, recognized by all subsequent geographers, in the headland where the Pyrenees descend to the sea. But his omission of all mention of the Greek colony of Rhoda (Rosas) is singular. SECT. l. PERIPLUS OF SCYLAX. 387 Latium to the Tyrrhenians. The Latins held the sea-coast from the borders of Tyrrhenia to the Circeian promontory: then came the Olsi (Volscians) for a short distance: next in order, the Campanians, Samnites and Lucanians, the last people extending as far as Rhegium and the Sicilian Straits.6 It is remarkable that he does not notice either the Arno or the Tiber: nor does he give the name of a single city, till he comes to Campania, with the exception of Massilia and its colonies, and the incidental mention of Rome:7 the first pas- sage in any extant author in which the name of the rising city is found. On the other hand he enumerates in detail the Greek colonies on the coast of Lucania, including even such comparatively unimportant towns as Pandosia, Hipponium and Mesma. In like manner his knowledge of Sicily, as might be expected, is detailed and accurate, while of Sardinia and Corsica he tells us nothing more than their names and their geo- graphical position, with regard to which he was well informed.8 § 11. He is also the first writer who has left us a detailed account of the shores of the Adriatic: and here his mention of the Greek cities of Ancona, Pharus, and Issa, which were not founded till the reign of the elder Dionysius, affords a clear proof that the Periplus could not have been written earlier than no. 380. His enumeration of the nations along the western shore of the Adriatic is clear, and consistent with all we know of their history from other sources. He assigns the coast as far as the mountain Orion (a name not found elsewhere, but which must clearly designate the mountain promontory of Garganus) to the Iapygians: next to whom came the Sam- nites: then the Umbrians, in whose territory was the Greek city of Ancona: next to these the Tyrrhenians, whose dominion 6 Periplus, § 12. No mention is very close approximations to the truth, found of the Bruttians, as a people distinct from the Lucanians. 7 ’A1rb 5% ’Ax/'riov (Antipolis) Tufipnvol é'dz/os uéxpz ‘Poi/ms ‘II-(Mews, § 5. 8 Thus he describes Sardinia as dis- tant from Africa a day and a night’s voyage, and two days and a night from the nearest point of Sicily. Both are reckoning the day’s voyage at 50 G. miles. From the Tyrrhenian coast to Corsica he calls it a day and a half, which considerably exceeds the truth; the distance between the nearest points of the island and the mainland being less than 50 G. miles. 202 388 r-ns'roav or ANCIENT GEOGRAPHY. CHAP. xr. extended from sea to sea. After them came the Celts or Gauls, “ who were left behind on the expedition of their brethren ”-—a remarkable passage, as the first mention of Gauls in this part of Italy—and then the Veneti, who adjoined the Istrians on the other side of the Adriatic.9 On the other hand, that of the tribes, as well as of the islands and headlands on the opposite shore, is confused, and often unintelligible : a circumstance which may be owing in some measure to the peculiarly complicated and intricate character of the coast in question, as well as to the shifting and unsettled nature of the semi-barbarous Illyrian tribes by which it was peopled. But the detail in which he gives the nomenclature of these tribes, as well as the minuteness with which he describes the con- figuration of the coast, proves how much it was frequented, and how well its details were known to the navigators from the Greek colonies of Gorcyra, Apollonia and Epidamnus.1 At the same time it is a remarkable instance of the per- sistence of a popular error once established, that in describing the coast of lstria, at the head of the Adriatic, he places there a river Istrus, and adds that it has another mouth by which it ‘discharges itself into the Euxine.2 This error—of which no trace is found in Herodotus —appears to have originated merely in the circumstance of the Greeks having found at the head of the Adriatic a people bearing the name of Istri or Istrians, which they at once connected with the celebrated river Ister. The confusion is the more singular as there is in fact no river subsequent writers—of supposing the 9 1449. 0111' author, however, (Ka'mucoiio'i 5% Xeppdvna'ou r’mi'yqo €7wimrw fell into the error—shared by many 'r‘ris H5A01r0w'fi0'0v,§22). Such a state- I. ment is not only absurd in itself, but innermost recess or head of the Adri- ! utterly at variance with our author’s atic (5 avxbs 'rofi ‘A6piou miMrov) to i own account of the distances along this have been situated in the territory of ? part of the coast; yet we strangely find the Celts, near the city of Atria, instead 1 the very same error repeated in the of in that of the Veneti. (See on this - geographical treatise ascribed to Scym- point the remarks of Letronne, p. 196.) nus Chius (vv. 405, 406), more than 1 There is, however, according to two centuries after our author. our present text, one enormous error, 2 Me'ra 8% ’Ez/é'rovs eialv "low-poi £61103 Where he speaks of the Hylli, an Illy- Ical 1ro'rapbs "Imrpos. 0570s 6 flora/1.65‘ Kai rian tribe, as occupying a peninsula “a sis 761/ Ildv'rov e’xBoiAAei. (§ 20.) little smaller than the Peloponnese ” (l) Snow. 1. PERIPLUS OF SOYLAX.~ 389 of any considerable size or importance flowing through the land of the Istrians, or indeed into the upper part of the Adriatic at all. But the fable, once admitted, was seized upon by the poets and logographers with a view to explaining the supposed passage of the Argonauts by this route: hence geo- graphers were determined to find a river that would answer their purpose, and gave the name of Ister or Istrus to the first stream that they could pitch upon.3 Scylax is also the first extant author who distinctly applies the name of Eridanus to the Padus or Po, the great river of northern Italy. But he has no mention of amber in connection with it: and he places the islands called Electrides, on the opposite or Liburnian coast of the Adriatic, apparently giving the name to some of the numerous islands in the Gulf of Quarnerofi" \Vith the geography of Greece itself, as well as the northern and eastern shores of the Egean, our author was naturally well acquainted : but these were of course familiar to all, and while his account of them is valuable to the student of ancient geography for its accuracy in detail, it cannot be regarded as having contributed anything to the enlargement of geo- graphical knowledge. § 12. Of the Euxine also his account is very full and de- tailed, giving the names of many of the numerous petty tribes which occupied its eastern shores, from the entrance of the Palus Maeotis to the neighbourhood of Trebizond,5 as well as of all the Greek settlements which studded its southern shore, many of these being inconsiderable places, which disappeared 3 The river Quieto, which is sup- posed by Mannert and C. Miiller to represent the Adriatic Istrus of the ancients, is in reality a very trifling stream, that could never have been seriously supposed to be one of the outlets of a great river. * § 21. The name of these Electrides or Amber Islands, as has been already pointed out (p. 14), was a more fiction. 5 Among these tribes, however, we are surprised to meet with the names of the Melanchlaeni and the Gelones, both of which are placed by Herodotus far in the interior of Scythia and on the European side of the Tana'is (Herodot. iv. 20, 101, 107,108). There is probably some error here, but the ethnography of the various tribes in- habiting the Caucasus has been in all ages so obscure that it is impossible to pronounce with certainty. 390 ‘ HISTORY OF ANCIENT GEOGRAPHY. CI-IAP. XIL in later times, and whose names are consequently not found in other geographers. The most easterly of these colonies were Dioscurias and Phasis, of which the latter was situated at the mouth of the river of that name, and may thus have commanded an important trade with the interior.6 On the'other hand the Greek settlement of Tanai's, at the mouth of the river of that name, which became in later times an important emporium of commerce, is not mentioned by Scylax, and probably was not founded till after his time.7 His account of the northern shores of the Euxine is indeed much less precise and full than that of the southern :8 and he shared in the very exaggerated notion prevalent in his day as to the size of the Palus Maeotis, which he regarded as about half the size of the Euxine.9 § 13. The description of the coasts of Asia Minor, from the Bosphorus to the Gulf of Issus, though copious and accurate, “ Xenophon, when he found himself i know from Demosthenes (Or. adv. Lep- at Cotyora, at the head of the Ten ! tinem, § 33) that the emporium of Thousand Greeks, entertained for a Theodosia, on the southern shore of time the project of proceeding with the Tauric peninsula, had at that time his whole force to Phasis to found a only been lately established by Leucon, colony there (Anal). v. 6, §§ 15, 36). king of Bosporus. Hence the mention Mr. Grote justly points out that this of this place in the Periplus (§ 68) may means the town of that name, and not be added to the numerous proofs of its the river (Hist. of Greece, vol. ix. late period which have been collected p. 185, note); but he considers Phasis by the diligence of its editors. to have been at this time “a native 8 It can hardly be otherwise than a city,” and not a Greek colony (Ib. mere accident that he omits all mention p. 180). As however we find it recog- of the Borysthenes, as well as of the nised as such by Scylax, it is probable Hypanis and the city of Olbia, passing that it was already such in the time of across the northern gulf of the Euxine Xenophon. Thatleader may, however, direct to the Tauric Chersonese. But well have thought that there was room it is more inexcusable that where he is for a larger and more important colony enumerating the greatest rivers of in so favourable a situation. Europe (§ 69), while he mentions the 7 Strabo, who describes it as having Tana'is in conjunction with the Ister been a flourishing place of trade, calls and the Rhodanus, he omits altogether it a colony of the Greeks ot' the Bos- the far more important Borysthenes. porus (xi. p. 493), meaning probably 9 703 finiaeos r017 Hdzl'rov b’m-os 100v from Pantioapaaum. It was likely to 'r'fis Maw'mSos Aim/m. Peripl. § 69. be one of the latest. as it was one of See Chapter VI. p. 178. the most remote, of their settlements; The statement here refers indeed and may very well not have been (strictly speaking) only to the length founded till after the date of our Peri- of the Palus Maaotis; but even this is plus. It was just about this time that in reality 16:8 than a fourth part of the the Greek kings of the Bosporus were paraplus of the Euxine with which our at the height of their power; and we : author compares it. Snow. 1. PERIPLUS or 'soYLAx. 391 presents nothing of any special interest. That of Phoenicia and Syria, on the contrary, as well as a part of that of Egypt, are in so corrupt and mutilated a condition as to be almost worthless. But the account of the African coast from the frontiers of Egypt to Carthage is one of the most ‘complete and satisfactory portions of the Periplus, and its details are in general extremely accurate. The clear information that the author possessed concerning the two Syrtes, the island of the Lotophagi, and the Tritonian Lake,1 contrasts in a striking manner with the vague and confused notions of these regions transmitted by Herodotus. Beyond Carthage on the contrary, his knowledge appears to have been comparatively meagre: but few names are mentioned, and the distances are not given in detail. The author contents himself with the general esti- mate that from Carthage to the Columns of Hercules was a voyage of seven days and seven nights under the most favour- able circumstances.2 § 14. 7While our author’s knowledge of Grades and the coast of Spain beyond the Pillars of Hercules was, as we have seen, extremely imperfect, he gives us a brief but interesting account of the west coast of Africa as far as the island of Cerne, which he describes as situated twelve days’ voyage from the Straits, and seven days beyond Cape Soloeis, with which he was evidently well acquainted, as he describes it as a promontory standing out boldly into the sea, and having an altar sacred to Poseidon on its summit.3 Intermediate between Cape Soloeis and Cerne he places a river called Xion, which can evidently be no other than the Lixus of Hanno!‘ Cerne was the seat of a considerable trade, carried on by the Phoenician 1 §§ 109, 110. 2 § 111. l 4 The Lixus of our author on the 3 It is clear that the headland t0 contrary, at the mouth of which he which he gives the name is the same 1 places a Phoenician (zle. Punic) city of with the Soloeis of Hanno: and his 1 the same name, with a Libyan town on mention of the altar 0n the summit is 1 the other side of the river (Peripl. probably derived from the voyage of l § 112), can certainly be no other than that commander. At all events this the river of Laraiche, to which the part of his Periplus is evidently taken ' name was given by Pliny and later from Carthaginian sources. 1 geographers. HISTORY OF ANCIENT GEOGRAPHY. OHAP. XI: merchants with the Ethiopian natives of the opposite coast, from whom they obtained ivory and the skins of lions, panthers, &c., in exchange for pottery, ornaments of stone, sweet ointments, and other wares.5 Beyond Oerne (says our author) the sea was no longer navigable on account of shoals and mud and sea-weed. But it was vaguely reported that there was continuous sea all along the southern coast of the Ethiopians, round to the shores of Egypt, so that Africa was in fact a great peninsula.6 § 15. Of the Indian Ocean, or even the Red Sea, no mention is found in the Periplus, a circumstance which would appear the more remarkable in the supposed work of an author, who had himself navigated those seas, and been indeed their first explorer. But it seems to have been assumed by the earlier editors that he had described these portions of the world in another work, specially devoted to that object: a supposition in itself not improbable, but unsupported by any evidence. § 16. Of the general configuration, or geographical relations, of the countries and seas that he describes, our author gives us very little information. He in one passage indeed repeats the statement of Herodotus that the width of Asia Minor from sea to sea (measured from Sinope across to Soli in Oilicia) was five days’ journey :7 and in another states the breadth of Italy from the city of Spina on the Adriatic across to Pisa on the Tyrrhenian shore at three days’ journey :8 but these vague estimates are almost the only statements of the kind which he has furnished us.9 His distances by sea are almost uniformly 5 § 112. The notice in such acon- duc‘ripl 8% eTuou Thu AiBiinv, i§112. He nection as this of Attic pottery (rcépa- ,uoz/ ’A'r'rmbv) is a curious proof of the established reputation of that manu- facture. 6 Tijs Ke'px/ns 5% 117'70‘011 ‘rd é'zre'icewa oi’mé'rz e’o'rl who're‘: as Bpaxu'rn'ra OaAoi'r- 'rns Kal 'nnAbx/ Kal ¢3Kos . . '. Aé'youo'z fo‘é 'rwes 'roz'rrovs 'roi‘zs Aieio'lras (those on the mainland adjoining Cerne)1rapv'7lcew .avvexsias, oilcofiv'raséz/Tefiflev sis Ai'yvvr'rov, xai ell/a: rau'rnv 1'2‘71/ OoiM'z'r'raV ovvexfi, here uses the term Euc'rh precisely in the same sense that it is employed by Herodotus. (See above, p. 161.) 7 § 102. He has, however, omitted to add the important words agar,’ Ewfipl annexed to his statement by Herodotus (ii. 34). 8 § 17. 9 The two statements are indeed widely discordant if compared with the reality; The actual distance from the Sizer. 1. PERIPLUS or SCYLAX. 393 given in days’ voyages :1 but he tells us in one place that he reckons a day’s voyage as corresponding to 500 stadia, and a night’s voyage (where he mentions nights and days consecu- tively) as equal to that of a day.2 Of course this can only be taken as an average computation: but it shows clearly what was his mode of estimating distances: he ascertained as best he could the length of the voyage by the number of days or nights actually occupied, without any attempt to compute the distance as a matter of measurement, or even estimate, except by the rough process of rendering the number of days’ voyage into stadia on the general average thus assumed. It is to be remarked that while the author, in more than one passage, sums up the whole amount of the distances, so as to give the total length of the paraplas or voyage along the coast, with all its windings in detail, from the Pillars of Hercules to the Tanais, and again from thence to the Nile, and back again to the Straits, we find no attempt to estimate the length of the Mediterranean, or even of the Euxine, as deduced from the number of days’ voyages in the most direct line. Probably our author did not conceive such a calculation to fall pro- perly within the province of his little work. But it is scarcely conceivable that some such estimate should not have been made even before his time by more systematic geographers. His other statements of the more considerable distances will be found on the whole very accurate, according to his own Euxine to the Mediterranean, on the which in this respect differ from all line indicated, exceeds 300 G. miles in ! the rest of the work. Probably this a straight line, while that across Italy l part of the Periplus was derived from from sea to sea is less than 100 G. miles, i a dilferent authority. without allowing for the distance inland It is worth notice that this practice of the two cities named. : of giving the distances only in days’ 1 Of course this applies only to the voyages, and not in stades, is mentioned longer distances. The shorter distances 1 as one of the peculiarities of the Peri- (of a few miles only) are naturally ‘ plus of Scylax of Caryanda by Marci- givcn in stades; and for this reason 1 anus of Heraclea (Epit.Artemiclor. § 2, this mode of measurementis frequently p. 63). It is therefore probable that employed in describing the coasts of the Periplus known to that author (in Greece, where the subdivisions are the fourth century A.D.) under the given in detail. But the same mode name of Scylax was the same which of computation is applied also to the we now possess. ‘ coasts of Syria, Phoenicia, and Egypt, 5 2 §69. 394 HISTORY or ANCIENT GEOGRAPHY. onip. XI. mode of computation. Thus as we have seen his estimates of the distances from Sardinia to Africa, and from Sardinia to Sicily, both of which must have been measured in a direct line,3 are very close approximations to the truth, at the rate of 50 G. miles for a day’s voyage. Onthe other hand he calls it a voyage of seven days and seven nights under the most favour- able circumstances 4 from Carthage to the Pillars of Hercules, where we must therefore suppose him to mean a rate above the average; and accordingly we find that the distance is not less than 800 G. miles, or 8000 stadia, by the shortest course that a ship could follow. § 17. At the end of the extant Periplus is found an enu- meration of the principal islands known to the author in the order of magnitude, beginning with 1. Sardinia as the largest of all: then 2. Sicily. 3. Crete. 4. Cyprus. 5. Euboea. 6. Cyrnus (Corsica). 7. Lesbos. 8. Rhodes. 9. Chios. 10. Sa- mos. 11. Corcyra. 12. Casos. 13. Cephallenia. 14. N axos. 15. Cos. 16. Zacynthus. I7. Lemnos. 18. ZEgina. 19. Im- bros. 20. Thasos. The most remarkable point in this list is the superiority given to Sardinia over Sicily, in which respect the author is at variance with almost all other ancient writers on geography, as well as with the received opinion of his own time.5 It is not worth while to discuss the correctness of the order of the smaller islands: but the total omission of the Balearic Islands, the largest of which is so superior to the greater part of those mentioned, shows how imperfect was the author’s acquaintance with the western parts of the Medi- terranean. 3 Peripl. § 7. i lowing all its sinuosities, but little 4 7017 Kaaaic'rov 1mm’), §111. This is exceeds 620 G. miles. But here it was confirmed by the fact that he elsewhere impossible to hold a direct course, and calls it seven days and seven nights’ ,' his estimate of the whole was merely voyage from the Columns to the limit , made by adding together the separate of Spain (§ 2), a distance which, if I distances from point to point. measured along the coast, without fol- j 5 See Note B, p. 406. SE01‘. 2. ARISTOTLE. 395. SEcTIoN 2.—Am'st0tle. § 1. It is much to be regretted that among the multifarious works of the great philosopher ARIsToTLE, there is none specially devoted to geography. The loss is the greater, as such a treatise would have afforded us a clear general view of the knowledge possessed by the Greeks upon this subject, before the great and sudden extension of their geographical horizon, arising from the conquests of Alexander. It is true that two of his extant works, the Meteorologica and the treatise On the Heavens,6 contain numerous notices connected with physical geography, as well as with those general views of the cosmical relations of the earth to the other heavenly bodies, and its figure and position in the system, which must always form the basis of scientific geography. "(But he has nowhere attempted to connect these with the descriptive geography of the earth’s surface, or to give any details with regard to that habitable world which was throughout the sole object of investigation with the Greek geographers; After passing in review the opinions entertained by earlier philosophers, from Thales to Anaxagoras,—which we have already briefly examined—he establishes the position that the earth is a sphere, and that it is situated in the centre of the universe, where it remains at rest, while all the other celestial bodies, including the fixed stars, revolve around it.7 The spherical form of the earth had become, in the time of Aristotle, a generally received tenet among philosophers. But he demonstrates it in two different methods; first, because this is the form which matter gravitating to a centre must necessarily assume, and it is a fact that all things do gravitate 6 Hep) Oz’ipowofi. There is no doubt and belongs to a much later age. (Sec of the genuineness of this work. The treatise “ On the Universe ” wepl mimuov (De Mundo) on the contrary, which is found in our collections of l Aristotle’s works, appended to . the Mcteorologica, is undoubtedly spurious, the Dissertation prefixed to Barthélemy de St. Hilaire’s translation of the ills- teo'rologz'ca (8V0, Paris, 1863); and Sir G. Lewis's History of Ancient Astro- nomy, p. 218.) ' 7 De Carlo, ii. 13, H. 396 HISTORY or ANCIENT GEOGRAPHY. 01m. x11 from all sides to the centre of the earth; secondly, by the circular appearance of the shadow of the earth during eclipses of the moon, which he rightly considered to be produced by the interposition of the earth between the sun and moon.8 Both these luminaries he considered to be spheres—an obvious con- clusion; but ‘he proceeded to infer that the planets and fixed stars9 were also spherical bodies, which of course in his day there was no means of proving. It is more remarkable that he arrived at‘ the conclusion that the earth itself was of small dimensions, as compared with the vast distance of the stars.1 This he inferred from the changes which took place in the fixed stars, as viewed from different points, so that even a moderate change of place, as for instance from Greece to Egypt, was sufficient to produce a notable change in the aspect of the heavens. With regard to its positive magnitude he was content to acquiesce in the conclusion of certain mathe- maticians, to whom he refers without naming them, that it was calculated to be 400,000 stadia in circumference.2 §2. In the above views of the cosmical relations of the earth, Aristotle may be regarded as following, or rather con- curring with, those of Eudoxus of Onidus, which had been already put forth nearly a generation before; but he was not content with adopting them as already established, but con- sidered himself bound to demonstrate them afresh. Hence he is not only the first extant writer to whom we are indebted for their statement, but is entitled to be considered as in great measure the original author of the system thus presented to us. The works of Eudoxus were for the most part of so distinctly astronomical a character that they scarcely claim- 8 Ibid. ii. 14,§13. 'rmc'iv 5001 Th ,ué'yefios (iwako'yfgecrem 9 Ibid. ii. I]. wezpéiv'rai 'rfis ‘ll'éPUPEPGL’GS‘, sis 're'r'rapof- 1 Ibid. ii. 14, § 14. The same thing ROI/‘Ta Aé'yovo'w ell/cu twprddas a'ra'o‘fwll. is still more strongly asserted in the It is singular that this reference to Mcteorologica, i. 14, §19, where he says ; certain unknown mathematicians, has that the bulk of the earth is “ nothing,” 1 been cited by many modern writers, as in comparison with the heavens that if it were astatement made by Aristotle surround it. himself, and entitled to implicit rever- 2 Ibid. ii. 14, § 16. real 'rz'éu pafimua- | ence as such. Seer. 2. ARISTOTLE. 397 any place in a review of the historical progress of geo- graphy.3 §3. Aristotle regarded the whole “habitable world,” as it was always termed by Greek geographers, as confined to the temperate zone between the tropics and the arctic regions. All beyond the tropic to the south he asserts to be unin- habitable from heat, while the portions of the earth beneath the Great Bear were equally uninhabitable from cold. Hence the habitable world had natural limits in breadth, or latitude; while it had none such in length, or longitude; the tract between the Columns of Hercules and India enjoying a tem- perature which would allow of its being inhabited or visited, had not the extreme points been separated by so great an extent of sea!‘ He adds moreover that there must certainly be in the southern hemisphere a temperate tract, correspond- ing to that in the northern; though he refrains from making any suggestion as to whether or not it was inhabited.5 The length of the known world, from India to the Columns of Hercules, he considers to be in the proportion of five to three to its breadth, from Ethiopia to the extreme parts of Scythia.6 Hence he justly treats with ridicule those who represented the inhabited world as circular; an idea which seems to have been still prevalent in his day, as it had been in that of Herodotus.7 § 4. The Meteorologica of Aristotle contain numerous specu- 3 It appears, however, that Eudoxus left a work of a distinctly geographical character, called Pfis wepiooos, to which Strabo frequently refers, though he gives us very little information con- cerning it. But it would seem to have included a detailed description of Greece. (Strabo, viii. p. 379; ix. p. 300; x. p. 465.) 4 Meteorologica, ii. 5, §§ 13, 15. 5 Ibid. §16. He points out, how- ever, that for this region there would be a set of winds corresponding to those in the northern hemisphere; but with relation to their pole; the south wind 3 answering to the north WIIlCl in our ‘ portion of the world, 850. 6 Ibid.§ 14. 7 Ibid. 9' 13. It is a curious instance of the persistence of once-established errors, that this idea should have still maintained its ground. But Aris- totle’s expression certainly points to it as the customary mode of drawing maps in his day. Au‘) and 'yekoiws 7pci¢ov¢n V1711 'rc‘zs weplofiovs r715‘ 'y'fis. Perhaps the notion that the earth was spherical, which had become established among philosophers, tended by a confusion of ideas to fortify the supposition that the world in the popular sense (if oiicovuéun) was also round. 398 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XI. lations upon subjects closely connected with physical geography, such as the winds, the changes of weather, with their accom-_ panying phenomena of rain, hail, &c., earthquakes and their causes; and the more slow and gradual changes in the con— formation of lands and seas. Among these last he correctly points ‘out the process of alluvial deposition by which in so many places the land is gaining on the sea, and especially in the Palus Maeotis, which, as he affirms, was continually becoming shallower, and would no doubt be one day entirely filled up, and converted into dry land.8 He adverts also to the currents by which that sea flowed into the Pontus, and that again into the ZEgean; but erroneously considers them as connected with the difference in depth of the several seas, supposing, or rather assuming, that the Palus Meeotis was shallower9 than the Euxine, that again than the Z'Egean, and the fEgean than the Sicilian Sea, which with the Tyrrhenian, was the deepest of all. The sea outside the Columns on the contrary he supposed, in accordance with a notion generally prevalent among the Greeks, to be muddy and shallow, and little agitated by winds.1 It is remarkable that no other notice of the external Ocean, or its tides, a phenomenon so striking to the Greeks in general, is found in this treatise. Indeed the very name of the Ocean occurs only in one passage, in reference to the notions of “ the ancients ” concerning it.2 § 5. Almost the only passage in the Meteorologica in which Aristotle has entered into any detailed statements of a strictly 8 Meteorologica, i. 14, § 29. The same 'rucbv naae'i'ral' ml 6 mum/cs, weplpfie'wv speculation, as we shall see, was re- finals. 0. 3, §8). The whole of this vived by Polybius. section, in which the author notices 9 Ibid. ii. 1, §§ 12, 13. “the two very large islands called the 1 Ibid. § 14. The notion that this Britannic, Albion and Ierne” (év Torin,» outer sea was undisturbed by winds 76 pxhv vfio'ol pé'ym'rai 're 'rv'yxolvovo'w (Gm/ca) shows how little it was known 050a: 3150, Bpe'ravmal Ae'yduex/ai, "AXBwv to Greek mariners. Kai ’Ie'pv'n, r6311 1rpol'o'ropnjue'vwu uel§ovs, 2 Ibid. i. 9, § 6. In the spurious z‘nrép 'roils Keivroils Kelueval,§13) bears treatise De Manda on the contrary the the unquestionable stamp of a much Ocean is distinctly described as sur- more advanced stage of geographical rounding the whole earth, and as iden- knowledge than that of the age of tical with “ the Atlantic Sea” (He'Aa- Aristotle. 709 3% Th new 550: 'rfis ollcovpe'u'ns ’A'r)\av- SE01‘. 2. ARISTOTLE. 399 geographical character is that in which, after pointing out that almost all great rivers took their rise in great mountain ranges, he proceeds to add examples in proof of his assertion. “ It is thus that in Asia most of the rivers and those of largest size descend from the mountain which is called Parnasos; and this is generally considered to be the largest of all the mountains that are situated towards the winter sunrise” (the south-east).3 “For directly after you have crossed this, you come in sight of the outer sea, the limit of which is not known with any certainty to the inhabitants of these parts. Now from this there flow, among other .rivers, the Bactru's, the Choaspes and the Araxes; of which last the Tanai's is a branch, which parts 0H’ and flows into the Palus Maeotis. The Indus also takes its rise in the same chain, which has the largest stream of all rivers. From the Caucasus in like manner there flow many rivers, of great size, as well as surpassing numbers, among which is the Phasis. Now the Caucasus is the largest mountain towards the summer rising (the north-east), both in extent and height. . . . . Again it is from the Pyrene, which is a mountain towards the equinoctial setting (ale. due west) in Celtica, that both the Istrus and the Tartessus flow: the latter towards the sea beyond the Columns, while the Istrus, after having traversed the whole of Europe, discharges itself into the Euxine. Of the other rivers towards the north, most take their rise in the Arkynian mountains, which are both in extent and height the largest in this part of the world. But under the Great Bear itself (in the extreme north) beyond the farthest limit of Scythia, are the mountains called Rhi- paean, concerningfthe magnitude of which many tales are told that are clearly fabulous.” 4 It would be diflicult to conceive a more utterly confused notion than this passage presents of the geographical relations of the mountain chains and rivers both of the east and the 3 The phrase of the winter sunrise intermediate quarters of the heavens and sunset, and conversely, the summer (the N.E., S.E., &c.), though of course rising and setting, are here used,as they not corresponding to them correctly. always are by Strabo, to designate the ‘ Meteorologica, i. 13, §§ 15-22. 400 HISTORY OF- ANCIENT GEOGRAPHY. can». XII west. Mount Parnasus is in all probability the same as was called by the Greeks in later times Paropamisus or Paropanis us : the name being wilfully assimilated to the familiar form of Parnassus: and it may be fairly supposed that the Araxes is confused with the Iaxartes, the connection of which with the Tanais long continued to be one of the prevailing errors of Greek geography. But the mention of the Choaspes in con- nection with this system is strangely erroneous, if the river of that name so familiar to the Greeks be intended.5 Nor is it true that the Caucasus gives rise to many great rivers. Aristotle himself indeed was only able to mention the Phasis, which notwithstanding its mythical celebrity is but an unim- portant stream, as compared with the great rivers of Asia, such as the Euphrates and the Tigris.6 The whole passage indeed clearly shows how little progress had been made from the days of Herodotus to those of Aristotle in any definite knowledge of the geography of Asia. § 6. In regard to the west and north it will be seen at a glance that his notions were still more vague and erroneous. He had indeed heard the name of the Pyrenees, as a mountain chain in the west of Gaul ;7 but preposterously makes them the source of the Ister or Danube, while he ignores altogether the far greater chain of the Alps. His Arkynian mountains in the north are evidently an exaggerated notion of the Herky- nian forest in- Germany, a name which we shall find again appearing in a vague manner among the Greek geographers. before they had any definite knowledge concerning it. But 5 It is however more probable that the flanks; an entirely erroneous idea—the Choaspes here mentioned may be in- absence of lakes being, in fact, one of tended for the river of Cabul, which is the leading peculiarities of that great called by the Greeks in general Cophen, mountain chain. but according to Strabo received a 7 At the same time the name of the tributary named the Choaspes (Strab. river Tartessus would show some idea xv. 1, p 697). But even in this case of the (onnection of the Pyrenees with Aristotle would seem to have con- Spain. It is impossible to determine founded it with the far more celebrated 5 what river is here intended; but the river of the same name. name was always connected with the 6 He tells us also that the Caucasus ; southern regions of the peninsula. was said to have many lakes on its ; SE01‘. 2. ARISTOTLE. 40 I at the same time he could not divest his mind from the fiction of the Rhipaean mountains in the far north, “from which (he adds) most of the other rivers of the largest size, next to the Ister, are said to derive their sources.”8 In regard to Africa it is not surprising that his information should be still more imperfect: but it is unintelligible that, while mentioning the Nile as rising in the Silver Mountains (Argyrus Mons), the only other names that he notices are those of the ZEgon, and the Nyses—both of them otherwise wholly unknown, but which he describes as rising in the Ethiopian mountains—and the Chremetes, which flowed into the outer sea, a name equally unknown to later geographers.9 §7. We see therefore that, while Aristotle’s views of the physical and meteorological conditions connected with geo- graphy, were decidedly in advance of those of earlier writers, his knowledge of the positive geography of the inhabited world was still very imperfect, as well as limited. His Meteo- rologica were certainly published before the expedition of Alexander into Asia had given a sudden extension to the horizon of Greek knowledge in this direction.1 In one respect indeed his adherence to the old ideas, generally received in his time, preserved him from an error, which obtained general acceptance among the Greek geo- graphers for the next three centuries. He states distinctly that the Caspian Sea is entirely isolated, and is inhabited all round.2 \Ve have already seen that this fact was known to Herodotus ; but as we shall presently see, a contrary view pre- 8 fie'ovcn 5’ 051/ of 1rAei‘a'ror Kai ,ue'yro'ror : note.) ,ue'ra 'rbv "Io'rp011 7631/ ZiAAwv werapaiv 1 We have no positive evidence con- e’u'refieev, 56s ¢ao'w. Ibid. §20. I cerning the date of the Meteorologica, These words doubtless refer to the but it may safely be assigned to about Borysthenes, and the other great rivers I 13.0. 341-330. of Scythia, which are not otherwise 2 ii. 1,§10. At the same time he mentioned. ; commits the mistake of supposing the 9 The suggestion of some modern Caspian and Hyrcanian to be two pd8a.) It is true that the distance given by Eratosthenes (ap. Strab. xi. p. 514) of 1600 stadia from Alexandria in Aria to Prophthasia considerably exceeds the actual distance from Herat to Furrah: but he himself adds that others gave it as only 1500 stadia. Prof. Wilson was led by this discrepancy to identify Prophthasia with the ruins of a city called Peshawarun, near the shores of the shallow lake which occupies the north of Seistan, about 70 miles south of Furrah. These ruins were first discovered by Captain Christie, but it appears from the observations of recent travellers, that they are not of ancient date. (Bellew’s Journey, p. 245.) Similar ruins are scattered through the whole of Seistan, which is a district of great fertility, though of limited extent, and appears to have at one time supported a large population. The capital at the time of the Arab conquest in A.D. 652 still retained the ancient name of Zaranj, which obviously represents the Drangiana of the Greeks, or Zarangiana as Isidore writes the word. It was situated between the Helmund and the lake, near the ruins of the later city of J ellalabad. But there is no evidence from ancient writers of the existence of a city on this site, to which the capital was removed according to Arab tradition shortly before the Mahometan conquest. Se'istan has of late years been repeatedly visited by modern travellers, but it is still imperfectly known. The results of recent explorations have been brought together by Sir H. Rawlinson in the Journal of the Geographical Society, vol. xliii. p. 27 2. NOTE U, p. 426. INDIAN TRIBES ‘VEST OF THE INDUS. It is a point of some ethnographical interest that Alexander is described as encountering in this part of his advance, from Can- NOTE V. HERODOTUS : AFRICA. 489 dahar to Cabul, tribes who are designated as “ Indians ” (ré‘w ’Iv3&3v 1'01‘); n'poo'Xu'ipovs ’Apaxérots. Arrian, iii. 28.) This cmfirms the traditions of the Hindoos themselves, that at this period tribes of true Indian origin occupied the valleys of the Paropainisus, and a considerable extent of country west of the Indus, from which they were gradually driven out by the pressure of invading tribes from the north. (Wilson’s Ariana, p. 125 ; Cunningham, pp. 125, 133.) The Gandarians also, who occupied a part of the territory to the west of the Indus (see note to Herodotus, chap. VII. p. 238), were clearly an Indian tribe. Their name is, however, not found in the historians of Alexander. It may not be useless to observe that the resemblance to that of Candahar is purely accidental, the latter, which is that of the city, not of a people, being probably a corrup- tion of Alexandria. NOTE V, p. 426. CLIMATE OF ARACI-IOSIA. The great hardships and sufferings endured by the army of Alexander in this part of their march became a favourite topic of exaggerated declamation with the rhetorical writers of later days, and sometimes led to the misconception that they must have occurred during the actual passage of the Hindoo Koosh. The route from Candahar by Ghizni to Cabul presents indeed no serious difliculty to the advance of an army during the greater part of the year: but in winter the cold is intense, and the snow is such as fully to justify the Greek historians in their accounts. Accord- ing to Elphinstone: “ In proceeding east from Candahar, the cold of the winter increases at every stage. . . . Even at Kelat-i-Ghilzi snow falls often and lies long, and the Turnuk is often frozen so as to bear a man. . . . Ascending the valley of the Turnuk, We attlast reach the level of Ghizni, which is generally mentioned as the coldest part of the plain country in the Caubul dominions. The cold of Ghizni is spoken of as excessive, even by the inhabit- ants of the cold countries in its neighbourhood. For the greater part of the winter the inhabitants seldom quit their houses; and oven in the city of Ghizni the snow has been known to lie deep for some time after the vernal equinox. Traditions prevail of the city having been twice destroyed by falls of snow, in which all the inhabitants were buried.” Elphinstone’s Caubul, vol. i. p. 182. 490 HISTORY OF ANGIENT GEOGRAPHY. CIIAI’. XII. The city of Ghizni, in fact, is situated at a height of not less than 7730 feet above the level of the sea, while the pass between that and Oabul rises to 8700 feet. Lieut. Oonolly also speaks of the road from O'andahar to Oabul as practically closed to travellers in the winter, on account of the depth of the snow-drifts, and the severity of the cold. (Oonolly’s Journey, vol. ii. p. 44.) NOTE x, p. 4.27. SITE OF ALEXANDRIA AD OAUOASUM. Professor Wilson, (writing in 1841) rem'arks that the exact site of Alexandria ad Caucasum is “ exceedingly difiicult to determine, though it is not possible now to stray so widely from the spot, as geographers of the first merit deviated some few years ago, in fixing it at Ghizni or Oandahar.” (Ariana, p. 179.) Other writers, in- cluding Sir A. Burnes, were disposed to place it at Bamian, but this is wholly irreconcilable with the statement of Pliny concerning the distance from Ortospana (assuming that to be rightly placed at Oabul), and is moreover at variance with the explicit statement of Arrian (confirmed as it is by the nature of the case) that Alexander halted at the foot of the Hindoo Koosh, waiting till spring before he undertook the passage of that great mountain-chain, Now Bamian is situated in the midst of the mountains, after crossing the main ridge of the Hindoo Koosh (Burnes’s Travels, vol. ii. p. 163), where there could be no reason for making any halt of long duration. The plain or broad valley of Koh Daman on the contrary is a very rich and fertile district, of the beauty of which travellers speak in terms of great admiration, and extending up to the very foot of the great mountain barrier of the Hindoo Koosh. Such a site would be admirably adapted for the foundation of a permanent settlement; and in this valley, near the modern village of Charikar, are found ruins indicating the former existence of an important city. These ruins, which were first discovered by Mr. Masson, are regarded by Prof. Wilson as those of Alexandria ad Caucasum, and the same view has been adopted by M. Vivien de St. Martin, who has since investigated the subject with much care, as well as by Gen. Oun- ningham. The choice must be considered as lying between this spot and Beghram, about seven miles to the east of it, where a multitude of coins and other ancient relics have also been found, NOTE X. EXPEDITION OF ALEXANDER. 49I indicating beyond a doubt the site of an ancient city. The distance between Charikar and Cabul is indeed considerably less than that indicated by Pliny; but its position on the direct route to the passes which lead by the valleys of Ghorbund across the central range is a strong argument in its favour. At the present day “ the caravans that proceed from Cabul across the Hindoo Koosh to Khulm (in the valley of the Oxus), pass through Charikar, a long straggling village, near the foot of the Koosh.” (Vigne’s Visit to Ghuzni, Kabul, eta, p. 215.) The position of Beghram vis decidedly less favourable, and the abundance of ancient remains there may be explained by supposing that locality to have been the site of Nicaea, a city which must have been founded by Alexander about the same time with Alexandria, as we find it mentioned as already in existence at the time of his return from Bactria. (Arrian, iv. 22, § 6.) 1 The argument derived by Gen. Cunningham, as well as by M. Vivien de St. Martin, from the proximity of a village named Houpian or Opian, which they regard as directly derived from the ancient appellation of ’Qvrtamj, is undoubtedly entitled to some weight, though by no means so conclusive, as it is considered by the latter author. The name of ’Qvrtam§ is found only in Stephanus of Byzantium, who in his enumeration of the different cities which bore the appellation of Alexandria, mentions one £11 '71) ,Q'mavfi Karo‘. 'n‘yv ’Iv8u<fiv. The name being otherwise unknown, several of the editors have proposed conjectural emendations : thus Freinshemius would read ’O$tam§ and Salmasius ’Aptav1j. But the correctness of the reading is confirmed by the mention of the ’Qvrt’at as an Indian tribe cited by Stephanus from Hecataeus (v. Dm’at), and the MSS. vary only between Klmavfi and Throw/vi. On the other hand, the supposed mention of the same name in Pliny (vi. 0. 21, § 62), on which Gen. Cunningham lays much stress, is certainly an error, the reading “ Alexandriam Opianes” being a mere conjecture of Pintianus (derived from the passage in Stephanus), while that of “Alexandri oppidum,” which is found in all the MSS. is quite unobjectionable, and is justly retained by the recent editors. (See Sillig’s edition of Pliny, l. c.) 1 General Cunningham, however, but Whose name is found in Pliny and claims the site of Beghram for Car- Ptolemy (Ancient Geogr. of India, pp. tana, a city not mentioned by the his- 26-29). torians or geographers of Alexander, 492 nIs‘roRY or ANCIENT GEOGRAPHY. can». x11. Both passages of Stephanus, however, point to the name as that of a district or territory: hence the evidence of the modern appel- lation has comparatively little weight in determining the exact position of the city. (See Wilson’s Ariana, pp. 179-182; Vivien de St. Martin, Ge'o- graphz'e Greeqne et Latina de Z’Inde, 1858, pp. 23-26 ; Cunningham’s Ancient Geography ofIndz'a, pp. 20—24.) '‘ NOTE Y,‘p. 428. PASSES OF THE HINDOO KOOSH. Our knowledge of these passes was derived in the first instance from Lieut. Wood, who states that all the three most direct passes. from Cabul into Turkestan lead through the plain of the Koh-Daman, “ where diverging as they enter among the mountains at its head, they wind up the course of the diiferent streams, from which the several passes take the names of Ghorbund, Parwan, and Pancbshir.” (Wood’s Journey to the Scarce of the Oxus, p. 118, 2nd edit. Lond.. 187 0.) He himself attempted the passage of the Parwan Pass, but was driven back by snow storms, and compelled to take the more open road by Bamian. This was in the month of November: in the following April he recrossed the Hindoo Koosh by the Panch- shir Pass with comparatively little difficulty. This last is the route which Alexander is supposed by de St. Martin to have followed (Ge’ogr. de Z’Inde, p. 23): but when he adds “11 n’y a pas deux routes possibles,” this positive assertion is based upon the assumption that Adrapsa or Drapsaca is identical with Anderab or Inderab at the foot of the Panchshir Pass, on the northern side of the great mountain chain. But this identification rests mainly on the supposed resemblance of name, which is certainly not close enough to be conclusive : and it seems more probable that Drapsaca -—-where Alexander halted some time to recruit his troops after their fatigues—was situated quite in the plain or fertile valley of the Oxus, in the same manner as the modern town of Kunduz. General Cunningham has adopted the same view with M. de St. Martin, which appears on the whole the most probable. We learn that Alexander on his return from Bactria took a difl'erent and shorter route, descending to the same point—his newly-founded city of Alexandria. And this is easily explained, if we suppose him to NOTE Z. EXPEDITION OF ALEXANDER. 493 have followed the route by the Panchshir Pass in the first instance, and to have recrossed the mountains by the more direct route known as the Kushan Pass, which leads from Ghori to Kushan at the entrance of the Ghorbund valley. The latter route, which is evidently the same that is called by Lieut. Wood the Ghorbund Pass, is frequented by travellers and caravans, and was even suc- cessfully crossed in 1840 by a troop of horse artillery. It could therefore offer no extraordinary difliculties to an army such as that of Alexander. (Cunningham’s Ancient Geography of India, pp. .24, 25.) Mr. Grote inclines to the pass of Bamian, because “ it seems the only one among the four passes open to an army in the winter” (vol. xii. p. 271, note). But it is clear that the supposition of Alexander’s having crossed the mighty range of the Hindoo Koosh “ towards the close of winter ” (Ibid.) is not only uncalled for, but at variance with the distinct statements of ancient authors. Arrian indeed conveys no definite information upon the subject, and the narrative of Curtius is very confused, but Strabo, whose narrative of this part of Alexander’s movements derived from Aristobulus is remarkably distinct and clear (xv. p. 725), says that he tra- versed the land of the Paropamisadae (Le. from Candahar to Cabul) at the time of the setting of the Pleiades(1‘m-o IIAewiSos Sv'o'w) z'.e. the beginning of the winter: and suifered much from snow and hardships. He was still at the foot of the mountains on the south side: but having wintered there and founded a city (Btaxetlado-as 8’ afirédt, Kat miAw Kn’cras) he crossed the mountain range into Bactria (tivrepwjKpw-ev at; riyv BaKrptai/fiv). It was not therefore till the spring (of 13.0. 329) that he crossed the Hindoo Koosh, at which time all the passes are open, though still covered with snow, so that an army would suffer severely in crossing them, as we know in fact that Alexander’s army did. (See Lieut. Wood’s account of his passage of the Panchshir Pass in April. Journey to the Oxus, chap. xxiv.) It is said however that the Pass of Kushan is open all the year through. NOTE Z, p. 437. COMPARATIVE VALUE OF THE HISTORIANS OF ALEXANDER. In all geographical inquiries and discussions of the difficulties that occur in attempting to trace the campaigns of Alexander, it is 494 HISTORY OF ANGIENT GEOGRAPHY. OHAP. XII. especially important to bear in mind the different character and value of the authorities from which our information is derived. Of these Arrian, who occupies beyond all comparison the foremost place, though writing centuries after the events which he records, bases his narrative, as he himself tells us, mainly, if not exclusively, upon those of Aristobulus and Ptolemy the son of Lagus, both of whom were companions in arms of Alexander, and accompanied him throughout his expedition. So far therefore as his narrative represents these authorities, it may be received as thoroughly trustworthy, or at least incomparably superior in this respect to any other that we possess. But moreover Arrian was himself a man who had received a military training, who had commanded armies and governed provinces, and was therefore the better able to understand and appreciate the true merit of such authors as those mentioned. To this it must be added, that we learn from his report to the emperor Hadrian concerning the shores of the Euxine (commonly known as the Periplus of the Euxine Sea) as well as from his having prepared and published the elaborate ab- stract of the voyage of Nearchus, which will be examined in the next chapter, that Arrian had a special turn for geography, and was therefore disposed to pay as much attention to the geographical, as to the historical, statements of his original authorities. It is entirely in accordance with this that we find the notices in Strabo connected with the campaigns of Alexander (which are very numerous) almost always in agreement with those of Arrian. Strabo indeed appears to have written these portions of his work with that of Aristobulus constantly before him, and there is little doubt that many of his statements are derived from that author, even where he is not quoted by name. But the case is quite otherwise with regard to the three other extant historians of Alexander—Diodorus, Plutarch and Quintus Curtius. All three of these writers appear to have chiefly followed the authority of Clitarchus, who, though a contemporary of the great king, was certainly not a writer of judgment and discretion, and had in a great degree the turn, unfortunately so common with the Greeks, of converting history into a rhetorical exercise, and looking far more to the opportunities afforded him for the display of his oratorical powers than to the accuracy of his facts or the truth of his historical details. (See Geier, Alewandrz' Magni His- toriaram Scriptores, 8vo. Lips. 1844, pp. 154—159; Heyne, dc Font'i- bus Diodorz', p. 115.) Unfortunately the same defect is found in his NorE Z. EXPEDITION or ALEXANDER. 495 Roman follower Quintus Curtius. It is impossible to read his history through without being struck with the fact that his main object was evidently to imitate, and if possible rival, Livy; and that the style, and not truth or accuracy was what he regarded as the main end of history. Unfortunately his success in this respect was such as to secure for him in modern times a degree of popu- larity that has tended greatly to vitiate the prevailing notions concerning the history of Alexander.2 At the same time he has written with that careless indifference to geography which was so common among Roman writers; and though he has sometimes preserved to us names and details which would otherwise have been wanting, his geographical statements must always be received with caution, and are by no means implicitly to be relied on, in the absence of other evidence. , The same remark applies with equal force to Diodorus. Though his dry and prosaic narrative presents the strongest contrast with the turgid eloquence of Quintus Curtius, it is almost equally unworthy of confidence in regard to any details whether military or geographical, while the close agreement between the two shows that they are unquestionably derived from the same source. But we have abundant evidences from other parts of the history of Diodorus, where we have the opportunity of comparing him with better authorities, of ‘his carelessness and inaccuracy in regard to geographical matters. Nor is the case otherwise with Plutarch. If that writer has been justly termed the prince of biographers, the very qualities which entitle him to that praise are unfavourable to his value as a historian. To him everything is subordinate to the representa- tion of his hero. It is the man Alexander with whom alone he is concerned; not the conquest of Persia and India, still less the geographical details of his campaigns, that he is desirous to bring before his readers; and if we find at times an incidental notice of value in regard to these subjects, it is rather accidental than intro- duced with a set purpose. 2 It is remarkable that a writer so ' It may, however, probably be inferred popular and well known in modern that he was subsequent to Quintilian, times as Quintus Curtius should not be who could otherwise have hardly failed mentioned by any ancient author, and to notice a writer of such oratorical we are consequently left in the dark as pretensions. to the period at which he flourished. 496 HISTORY OF ANCIENT GEOGRAPHY. can). XII. NOTE A3, p. 440. LEGENDS CONCERNING BACCHUS.AND HERCULES The Greeks found, as they fancied, a confirmation of these legends concerning Bacchus, not only in the occurrence of names such as Nysa and Meron,——the resemblance of which was doubtless purely accidental,—but in the presence of ivy, as well as of wild vines, and the festive habits and bacchanalian processions of the natives. This account of their manners is confirmed by recent researches. “Ivy as far as has yet been ascertained, does not grow in the Cabul valley, but the grape flourishes and abounds in all the valleys of the Hindoo Koosh. The Kafirs, or people who dwell on the north of the Afghans, make wine, and old and young of both sexes amongst them drink it. They are very fond of drinking- parties, and of music and dancing.” (Wilson’s Ariana, p. 193.) It is not improbable that these so-called Kafirs are the remains of the aboriginal tribes, which in the time of Alexander were more widely spread, and inhabited many of the valleys now occupied by the Afghans. Even Strabo treats all these fancied resemblances as pure fictions, invented for the purpose of flattering Alexander, and rejects the supposed expeditions of Hercules and Bacchus to India as wholly fabulous (xv. i. § 8, pp. 687, 688). NOTE Bb, p. 4410. THE ROCK AORNUS The position of the celebrated rock fortress of Aornus, which holds so prominent a place in the narratives of Alexander’s histo- ‘ rians, has especially exercised the ingenuity of modern writers, but it cannot yet be said to be determined on satisfactory grounds. In fact there are so many sites in this rugged. tract which suit with the general description of this impregnable fortress, that mere local resemblance has little weight, and we are so little able to follow the movements of Alexander through these regions that they afford us almost no assistance in the matter. If we can trust the expressions of Quintus Curtius and Diodorus, it rose immediately above the Indus, so that that river actually washed its base (Q. Curt. viii. ii. § 7 ; Diodor. xvii. 85), and this is confirmed by NOTE BB. EXPEDITION OF ALEXANDER. Strabo (xv. p. 688), though not mentioned by Arrian. It appears also from Arrian’s narrative to have been situated near to (ééveyyvs) a town called Embolima, which he describes as some distance above the city of Peucelaotis, in the upper valley of the Indus ; a position which explains Strabo's expression that it was near the sources of the Indus, a term by which he undoubtedly meant to designate the place where that river first issues from the gorges of the Himalaya. (See p. 449.) It seems probable therefore that the site must be sought in the neighbourhood of Derbend (about 50 miles above Attock), where the river issues out from the deep mountain gorges, through which its course has been confined for some time past, and emerges into the plains. According to Major Abbott, whose view has been adopted by M. Vivien de St. Martin, the rock Aornus is represented by a pro- jecting rock on the right bank of the Indus, opposite to the village of Torbela, about twenty miles below the gorge of Derbend ; while the site of Embolima is marked by the names of Arab and Balimah, still retained by two old castles in the neighbourhood of Derbend. But such resemblances of name are worth very little; and Arrian’s narrative would certainly lead us to suppose that Embolima was situated below Aornus. The name also is certainly Greek (at least in the form transmitted to us), and was probably applied to a place situated at the confluence of some other river with the Indus. The whole subject is ably discussed by Mr. Grote (History of Greece, vol. xii. p. 304, note), by M. Vivien de St. Martin (Ge'o- graphie Grecque et Latine cle l’Inde, p. 40-44, and by Mr. Long in Smith’s Diet. of Ancient Geography, art. Aornus). It has been since investigated anew by General Cunningham, who had the great advantage of personal acquaintance with the localities. (See his Ancient Geography of India, pp. 58—78.) But the site on which he has fixed—an isolated mountain called Ranigat, 16 miles N. of the town of Ohind on the Indus—though answering in some respects well to the description of the fortress of Aornus (he himself admits that the resemblance is incomplete), is liable to the insuperable objection that it is so far from the Indus as to be wholly at variance with the statements of Curtius, Dio- dorus and Strabo, as to its proximity to that river. It is singular that General Cunningham does not even allude to this difficulty, which must be considered as conclusive against the site in question. The authority of either Curtius or Diodorus ‘alone might well be von 1. 2 K 498 HISTORY or ANCIENT GEOGRAPHY. crap. XII. rejected, especially if opposed to Arrian, but that is not the case in this instance, while their joint statement is confirmed by that of Strabo, who certainly derived his information concerning the campaigns of Alexander in these regions from very good sources. In accordance with this view General Cunningham would identify Embolima with Ohind, a town on the right bank of the Indus, about 18 miles above Attock; but his reasons for so doing are quite inconclusive in themselves. The one conclusion must stand or fall with the other. The name Aornus is doubtless a Hellenised form of some native appellation—distorted so as to suit the fanciful etymology from d and iipvts, as inaccessible even to birds. It is suggested by Pro- fessor Wilson that this was merely the Sanscrit term Awara or Awarana, signifying “ an enclosure” or “ stockade ;” so that Aornus was in reality nothing more than a stockaded enclosure, situated on a detached hill or mountain of difficult approach. (Wilson’s Ariana, p. 192.) In this case we cannot hope for any assistance in determining the site from resemblance of name. NOTE Oc, p. 440. PEUOELAOTIS. The position of Peucelaotis, which appears sometimes as the name of a city, sometimes as that of a district, is very imperfectly marked by the Greek and Roman historians, though they all agree in placing the latter in the lower valley of the Oophen or Oabul river, corresponding to the district of Peshawer. The expressions of Strabo (xv. 1, § 27) would indeed seem to imply that the city was on the Indus, and close to the point at which Alexander crossed the river; but Arrian repeatedly uses the expression that it was not far from the Indus, which is doubtless the more correct. The district of the name probably extended quite to that river, while the city was at some distance from its banks. The form Peucelaotis is so peculiarly Greek, as applied to a district (as in the case of Pelasgiotis, Histiaeotis, &c.) as to lead to the suspicion that it was of purely Greek formation; but it appears probable that it was really a corruption of the Sanscrit name Pushkalavati, which occurs in the Mahabharata. The position of this is considered by oriental scholars to be fixed at a place called Hashtnagar, on the NOTES D D, E E. EXPEDITION 0F ALEXANDER. 499 north bank of the Cabul river, near its junction with the river of Swat. (St. Martin, Geographic de Z’Inde, p. 37; Cunningham’s Geography of Ancient India, pp. 49, 50.) This situation is certainly well in accordance with the expressions of Arrian. NOTE Dd, p. 443. TAXILA. The situation of Taxila between the Indus and the Hydaspes is unquestionable, but its precise site was, until very lately, the sub- ject of much doubt. It was placed by M. Court, by Sir A. Burnes, and by Professor ‘Wilson at Manikyala, celebrated for its tape, and where there are many other ruins. But the distances given by Pliny (vi. 17 , § 62), on the authority of Diognetus and Baeton (the so-called mensores of Alexander) certainly seemed to coincide better with the suggestion of Major Abbott, adopted by M. de St. Martin, that we should look for it in the neighbourhood of Hassan Abdul, about 25 miles to the N.W. I of Bawul Pindee. (See Burnes’s Travels, vol. ii. p. 58; Wilson’s Ariana, p. 196; St. Martin, Geogra- phic de Z’Inde, pp. 92-98.) General Cunningham was the first to point out the existence of very extensive ruins in the neighbourhood of a place called Shah Dheri, about 8 miles SE. of Hassan Abdul, which from their cha- racter and extent there seems every reason to believe to be those of Taxila. That city, the Indian name of which was Takshasila, con-. tinued to be a flourishing and important place for many centuries, and was still in existence in the seventh century after Christ. (Cunningham’s Ancient Geography of India, pp. 104-120.) NOTE E e, p. 443. PASSAGE OF THE HYDASPES. Besides the interest which naturally attaches to the scene of the great battle between Alexander and Porus, the site in question is of importance as being the point from which the king afterwards set out on his memorable voyage down the rivers to the Indian Ocean. All ancient writers agree in stating that the two cities, 2K2 500 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. Bucephala and N icaea, were founded in commemoration of his victory, and were situated in the immediate neighbourhood of the spot where he crossed the river. Hence we might naturally look for some_remains by which to identify the locality. But the difficulty arises rather from the abundance than the absence of such indications. Numerous ruins are found on both banks of the river for a space of about 40 miles below the modern town of Jhelum, extending as far as J elalpoor and Harriah. But the question may be considered prac- tically to lie between J helum, at which point one line of high road has in all ages crossed the Hydaspes, and the neighbourhood of J elalpoor, about 30 miles lower down the river. It was at this latter point, to which there is also a frequented high road, that Mr. 'Elphinstone and his suite, on their return from Caubul, crossed the Hydaspes, and the spot was thought by them to correspond precisely with the description given by Quintus Curtius of Alex- ander’s battle with Porus. (Elphinstone’s Caubul, vol. i. p. 109, 8vo. edit.) Sir A. Burnes however inclined in favour of the neigh- bourhood of J helum, and the same view was adopted by General Court, who had investigated the subject on the spot, as well as at a later period by General Abbot. The question has been examined again with great care by General Cunningham, and the result of his researches seems to show that there is a strong predominance of argument in favour of J elalpoor. In this case the town of that name probably occupies the site of Bucephala, while N icaea may be placed on the same site with the modern town of Mong—only about six miles distant from the recent battle-field of Chilianwalla. (Cunningham, Ane. Geogr. of India, pp. 159-178. See also Burnes’s Travels, vol. ii. pp. 419-51 ; and St. Martin, Geographic de Z’Inde, pp. 99-102.) mm F f, p. 4.44. GEOGRAPHY OF THE PUNJAB. At the present day we are so familiar with the geography of the ' Punjab, that it appears strange to recall how lately it is that we have acquired anything like a competent acquaintance even with its leading features. While in ancient times it was the first por- tion of India with which the Greeks and Romans became acquainted -indeed it may be said to be the only part of which they ever NOTE F F. EXPEDITION OF ALEXANDER. SOI obtained any true geographical notion—it has in modern days on the contrary vbeen the last region that has come within the domain of European knowledge. As late as 1775 when D’Anville pub- lished his Antiguite' Ge’ographique de l’Inde, the information possessed by that eminent geographer concerning this part of India was so imperfect that he was led into the grossest errors, and this portion of his work is a mass of confusion. Major Rennell was the first who was able, in part by the assistance of Oriental materials, to rectify these errors, and introduce a clear and intelligible view of the subject. (Memoir of a Map of Hindostan, p. 94-103, 3rd ed. Lond. 1793.) He himself observes that, as his own researches advanced, he was continually confirmed in his opinion of the accuracy of the statements transmitted to us by the historians of Alexander. Indeed there is no portion of the campaigns of that monarch in which the leading geographical features are more clearly marked, and can more readily be identified. That the case should be otherwise with regard to the various tribes and nations he encountered, can excite no surprise, when we consider how fluctuating are both the boundaries and appellations of such tribes, and how many successive waves of conquest have swept over the land since the time of Alexander. Moreover the operations against these different nations are generally indicated in the vaguest and most general manner, affording us little clue to their geographical position. (See the remarks of Major Rennell, p. 124.) Arrian himself had doubtless a very imperfect idea of the geography of the countries in question, and even if the writers whom he fol- lowed had supplied him with fuller details—which may well be doubted—he did not think fit to preserve them. As for Curtius and Diodorus their narratives are characterized, throughout the Indian campaigns, by even greater carelessness and inattention to geographical accuracy than that which generally marks their works. The Punjab is (as its name imports) the “Land of the five Rivers,” and all these rivers can be recognized and identified with— out difficulty. Even their names, which at first appear so totally . different in their modern and ancient forms, will be found to pre- sent‘ more points of resemblance than would be at first suspected, or else the change can be otherwise accounted for. 1. The Hydaspes is unquestionably the river commonly called in modern times the Jhelum (from a city of that name on its banks) but by Sanscrit writers the Bitastha or Vitastha, of which 502 HISTORY OF ANGIENT GEOGRAPHY. OIIAP. XII. the modern name of Behut (by which it is also known) is a mere corruption. The name is written by Ptolemy Bidaspes, a form that approximates more nearly to the Sanscrit than that usually adopted by Greek and Latin writers. 2. The Acesines is the Chenab, a name derived from the more ancient Sanscrit form Ohandrabhaga, which is traceable in the name Sandabala by which the river is designated by Ptolemy. The more usual form Acesines is said to have been an arbitrary change introduced by the Greeks with a view of avoiding a name supposed to be of ill omen. - 8. The Hydraotes is the modern Ravee, an undoubted corrup- tion of the Sanscrit Iravati, in which we trace without difficulty the origin of the Greek appellation. 4. The Hyphasis or Hypasis is unquestionably the modern Beas or Beyah, a name which is derived from the Sanscrit Vipasa. Here also the form preserved by Ptolemy, Bibasis, is the more correct, ‘ while that of Hypanis, adopted by Strabo and Diodorus, is clearly erroneous. 5. The Sutledge, which at the present day is reckoned the last of the five rivers, is not mentioned by the historians of Alexander, that conqueror having stopped short at the Hyphasis—but it ap- pears in Pliny under the name of Hesidrus, while it is termed by Ptolemy Zaradrus. The Sanscrit form is Satadrus. But while the principal rivers of the Punjab can thus be iden- fied with certainty, it must not be too hastily assumed that they - followed in the time of Alexander exactly the same course as at present. Since the country has been better known, abundant evidence has been brought forward to show that great changes have taken place even in recent times. Thus we know that the junction of the Sutledge and Beas, which now takes place about 40 miles above Ferozepoor, was formerly in the immediate neigh- bourhood of that city, and it was not till the year 1796 that the Sutledge suddenly changed its course, and joined the Beas at their present point of confluence. (Cunningham, pp. 217, 222.) Almost exactly the same thing occurred with regard to the Ravee (Hy- draotes), which formerly flowed under the walls of Mooltan, and - fell- into the Ohenab (Acesines) about 15 miles lower down. At the present day the junction takes place near Diwana Sinand, more than 30 miles above Mooltan. (Id. p. 221.) It is probable also that the Beas had in former days a wholly distinct course of its own, parallel with that of the Sutledge, and General Gunning- NOTE GG. EXPEDITION OF ALEXANDER. 503 ham supposes it to have held this independent channel until it ultimately fell into the Chenab, and did not join the Sutledge at all. (Id. p. 223.) Still more remarkable is the change in the junction of the Indus itself with the Chenab, which in the time of Timour and Akbar took place opposite Ooch, though their streams now unite at Mittunkote, 60 miles lower down. This change has taken place since the commencement of the pre- sent century. (Id. p. 220.) There is great reason, as we shall hereafter see, to suppose that still more extensive changes have taken place in the lower course of the Indus since the period of the Mahometan conquest of Sinde. NOTE Gg, p. 444. ALTARS ON THE HYPHASIS. \Vhen we read the description of the twelve altars erected by Alexander on the western bank of the Hyphasis, which appear to have been massive constructions, destined by him to remain as monuments of the point to which he had advanced, we are almost tempted to hope that some vestiges of them may still be discovered. This hope would be strongly confirmed if we could believe the statement of Philostratus, that Apollonius of Tyana on his journey into India (in the 2nd century after the Christian era) found the altars still subsisting; and even their inscriptions still legible. (Philostr. Vit. Apollon. ii. 43)? But it is certain that no reliance can be placed upon that fabulous narrative; and the researches of modern travellers have failed to discover any trace of such monuments. If indeed they are to be sought (as is held by many modern writers) below the present confluence of the Beas with the Sutledge, the shifting character of the stream and its frequent changes leave little probability that the site can ever be ascertained. (This character of the Sutledge is sufficiently marked. by its original Sanscrit name of Satadrus—-“ the hundred-channeled river.”) But if the point where Alexander came to the banks of the Beas was situated (as appears most probable) at some distance above the confluence of that river with the Sutledge, the right 3 Plutarch also speaks of the altars as still subsisting in his time (Alex. (32), but this is doubtless mere vague hearsay. 504 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. bank is throughout considerably more elevated than the left, and hence any considerable changes in the channel are much less likely. Yet it appears that even in this part of its course the river carries away villages and even towns, and it is said that ancient ruins have been thus destroyed in quite modern days. Very little hope can therefore be entertained of the discovery of any monumental remains calculated to throw light upon this in- teresting geographical question. (See the descriptions of the rivers Beas and Sutledge in Elphinstone’s Caubul, vol. ii. p. 426, and Burnes’s Travels, vol. i. pp. 153, 157, vol. ii. 4-7 The altars were undoubtedly situated on the right bank of the Hyphasis. Pliny alone places them on the opposite or eastern bank. He says of the Hyphasis “ qui fuit Alexandri itinerum terminus, exsuperato tamen amne, arisque in adversa ripa dicatis.” (Hist. Nat. vi. 17, § 62.) But this is opposed to the concurrent testimony of the historians of Alexander, as well as to the proba- bilities of the case. (Arrian, Anab. v. 29; Curt. Diodor. xvii. 95.) It appears to be well ascertained (as has been already mentioned) that the Sutledge, at a comparatively recent period, did not join the Beas till near Ferozepore, about 4.0 miles below the present junction (Burnes, vol. ii. p. 4, 5 ; Cunningham, p. 222) and this, of course, greatly increases the probability that Alexander reached the banks of the latter river above the confluence. Though the Sutledge is in many respects the more important stream and has much the longest course, it is little, if at all, the larger river at the point of junction, and the united streams are known for some distance below the confluence by the name of Beas, as they appear to have been in ancient times by that of Hyphasis. But the combined stream in the lower part of its course is now generally known as the Garra. (Elphinstone, l. 0.) General Cunningham, in his recent work, places the site in question below the present junction of the two streams, but above the ancient confluence near Ferozepoor. This is liable to the objection that the interval between the two rivers could in this case have been only a few miles, while the measurements given by Pliny assign a distance of not less than 168 Roman miles from the Hyphasis to the Hesidrus or Sutledge, and the same from thence to the Jumna. (Plin. l. 0. § 63.) But General Cunningham sup- poses Pliny to have misunderstood his authorities, and that these really gave only one distance—the interval between the Hyphasis and Sutledge being disregarded as practically of no account (p. 217 NOTE H H. EXPEDITION OF ALEXANDER. 505 This is cutting the Gordian knot with a vengeance! and is the less excusable in this case, as, if we suppose Alexander to have followed a more northerly route, keeping nearer to the mountains, the interval between the Beas and the Sutledge really becomes almost exactly equal to that from the Sutledge to the J umna. NOTE Hh, p. 4.45. SANGALA AND THE CATHZEL These arguments have been very fully urged by Gen. Cunningham (Ancient Geography of India, pp. 17 9—190), who had himself visited the site which he would identify with Sangala. It is situated in the midst of the plain called the Rechna Dooab, between the Ravee (Hydraotes) and the Chenab (Acesines) about 60 miles west of Lahore, and the local circumstances certainly seem to correspond very well with the description of the siege by Alexander. But if this be really the position of Sangala, we must suppose that after crossing the Hydraotes (which he is distinctly stated to have done) he recrossed that river in order to attack the Cathaeans, and cap— ture their chief city. That he should have turned aside from his general line of march for this purpose would be quite in accordance with his practice on other occasions, and is not inconsistent with the expressions of Arrian in regard to this particular expedition (Anal). v. 22) : but if he recrossed the Hydraotes and returned so far Westward, without our finding any indication of it in our existing historians, it will certainly prove that their narrative is even more vague and untrustworthy in a geographical point of view than we had been accustomed to suppose. The successive passages of the great rivers of the Punjab appear to form the distinct steps that mark the conqueror's progress; and if we cannot rely even upon these, the whole subject is thrown into hopeless confusion. It is remarkable however that Strabo tells us that the Cathaeans were situated, according to some authorities, between the Acesines and the Hydaspes, while others placed them beyond eastward of) both the Acesines and the Hydraotes (xv. c. 1, § 30, p. 699). The name of Sagala, which is probably identical with the Sangala of Arrian, is found also in Ptolemy (vii. 1, § 46), who however places it in the neighbourhood of the Hydaspes, thus apparently following the anonymous authorities cited by Strabo. 506 HISTORY or ANCIENT GEOGRAPHY. can». XII. The Cathaei are described by ancient authors as one of the most powerful nations of India, and Strabo has preserved to us some curious particulars concerning their manners and customs (l. 0.). They appear to have been closely connected with the Oxydracae and the Malli, with whom they were probably conterminous, as the three nations made common cause against Alexander. (Arrian, l. 0.) But this affords us little assistance in determining their precise position; that of the Oxydracae being almost equally obscure. The resemblance of name might easily lead us to suppose that there existed some connection between these Cathaeans and the well-known mediaeval appellation of Cathay, as applied to a region of the far east. But it is certain that the similarity is purely fortuitous; an instructive warning against too hastily building any conclusions upon such resemblances. Sir A. Burnes, on the other hand, has suggested, very plausibly, that the name is connected with that of the Kattia, a predatory and warlike race who are found scattered at intervals through the plains of the Punjab, and even across the deserts to Delhi. They now live an erratic life, but are supposed by Burnes to be the aborigines of the country. (Travels, vol. i. p. 112.) NOTE Ii, p. 446. BOATS ON THE INDUS. The statement in the text is that given by Arrian in his Anabasis (vi. 2, § 7), and rests on the authority of Ptolemy. In his Indica however (0. 19, § 7), where he is apparently following Nearchus, Arrian states the whole number of ships at only eight hundred, including both ships of war and transports. Schmieder and some other editors would correct this to eighteen hundred : but it seems more probable that the basis of the two calculations was different. Ptolemy distinctly includes the ordinary river-boats, which would doubtless have been collected in large numbers to assist in trans- porting so great an army and its supplies; while the terms of ~Nearchus would seem to imply only ships of war or regular transports. At the present day the Indus is navigated by a large number of boats or vessels of the country, some of them of large size. Near its mouth indeed it is navigable only for large flat-bottomed boats, NOTE KR. EXPEDITION OF ALEXANDER. 507 called “ doondees,” which, though large and unwieldy, never exceed fifty tons in burthen. (Burnes’s Travels, vol. i. p. 219.) But from Bukkur upwards it is navigated by a different description of boat called “shruk,” “ which is admirably adapted to the transport of troops, both horse and foot, from being as roomy before as astern.” (lb. p. 261.) The Chenab or Acesines is also navigable for boats of a similar description. (Ib. p. 276.) There can be no doubt that numerous vessels of this character would be found in the Indus and its tributaries as early as the time of Alexander, and Ptolemy ex- pressly mentions that they were turned to account as part of his fleet. But the transports for his cavalry appear to have been especially constructed for the purpose ; such a sight as a body of horses on board ship had never been before seen in these parts. (Arrian, Anab. vi. 3, § 4.) _ The Mahomedan historians, with their customary exaggeration, speak of forty thousand vessels as employed in the navigation of the Indus under the Mogul dynasty. (Abul Fazil, quoted by Vincent, Voyage of Nearchus, p. 88.) Alexander had, on a former occasion, when he first came to the banks of the Hydaspes and found himself opposed by Porus, trans- ported the vessels of which he had previously made use for the passage of the Indus overland to the Hydaspes. (Arrian, Anab. v. 8, § 4.) But there is certainly no reason to suppose, as Dr. Vincent has done, that this was the case with the fleet with which he de- scended the river. It was in fact much more easy to construct a fleet on the Hydaspes than on the Indus. (See note to p. 446.) NOTE Kk, p. 44.7. DESCENT OF THE INDUS. Sir A. Burnes, who in 1831 ascended the Indus and its tribu- taries the Chenab and Ravee, to Lahore, took exactly sixty days on the ascending voyage at the most favourable season; but he considers it possible for a boat to “drop down from Lahore to the sea in fifteen days ” at the very quickest, six of which would be occupied in the descent to Mooltan, and nine from thence to the sea. (Travels, vol. 1'. p. 196, 197; Journal of Geographical Society, vol. iii. p. 113.) He estimates the distance from Lahore to the sea, by the course of the river, at about 1000 British miles (Ib. p. 195). The distance 508 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. traversed by the fleet of Alexander must have been considerably _ less, as although the point from which it set out was (probably) at least as near the mountains as Lahore, the course of the J helum and Chenab, above their junction with the Ravee, is much less tortuous than that of the latter river. Pliny absurdly asserts that Alexander took five months and some days to descend the Indus, though he never proceeded less than 600 stadia a day. (Proditur Alexandrum nullo die minus stadia sexcenta navigasse in Indo, nec potuisse ante menses quinque enavigare, adjectis paucis diebus. Plin. H.N. vi. 17, § 60.) He would thus have navigated the river for more than 9000 G. miles! It would be curious to know from what source Pliny derived this extra- ordinary statement, which he repeats without in the least perceiving its absurdity. But a descent of sixty miles in a day, as an occasional rate of progress, is by no means improbable. NoTE L1, p. 447. CONFLUENCE OF THE HYDASPES WITH THE AOESINES. The confluence of the Hydaspes with the Acesines appears as a point of much importance in all the accounts of the voyage of Alexander. According to the historians the junction of the two streams gave rise to dangerous rapids, accompanied with violent eddies and tumultuous waves, which not only caused great alarm to the Macedonian sailors, but occasioned the loss of several ships. (Arrian, Anal). vi. 4, 5; Q. Curt. ix. 4, §9-14; Diodor. xvii. 97.) It is a ‘strong instance of the carelessness of Diodorus in geographical matters that he represents these rapids as occurring at the confluence of the two rivers with the Indus. The circum- stances are as usual amplified by Quintus Curtius with much turgid eloquence, but even in the more sober narrative of Arrian the dangers appear sufficiently formidable. Cherefeddin also, the his- torian of Timour, says that “the waves dashing against each other, made it appear like a troubled ocean ” (cited by Rennell, p. 118). But it appears from the description of Sir A. Burnes, the first European who visited the spot, that these accounts, though not without foundation, are greatly exaggerated. “ The river (he says) joins the Acesines with a murmuring noise, but the velocity of the current is inconsiderable, and vessels pass it without danger, NOTES MM, N N. EXPEDITION OF ALEXANDER. except in July and August. There are no eddies or rocks, nor is the channel confined, but the ancient character is supported by the noise of the confluence, which is greater than that of any of the other rivers.” (Travels, vol. i. p. 109.) The boatmt-n of the river however still regard the passage as a perilous one, during the season when the river is swollen. (Ibid.) NOTE Mm, p. 4.47. COURSE OF THE HYPHASIS. Arrian expressly says that the Hyphasis (meaning undoubtedly the combined stream formed by the Beas and Sutledge, now known as the Garra) falls into the Acesines (Anal). vi. 14, § 5) below its junction with the Hydraotes. He adds that the Acesines continues to retain its name, notwithstanding its junction with so many im- portant rivers, until it finally discharged their united waters into the Indus. This still continues to be the case with- the Chenab. (Burnes, vol. i. p. 78.) It is singular that notwithstanding this distinct statement of Arrian, both Major Rennell and Dr. Vincent doubted whether the Hyphasis really fell into the Acesines, and the former even goes so far as to say “the truth is, that the Hyphasis (or Beyah) does not join the Chenaub, but after uniting its waters with those of the Sutlege, falls into the Indus a great way further down ” (p. 129), and his map is constructed in accordance with this assumption. As has been already mentioned (Note F f) it is probable that the Indus and Acesines in the time of Alexander met in the neighbour— hood of Ooch, considerably above their present point of junction, but in this case there is no reason to suppose that the Hyphasis or Sutledge pursued a separate course till it joined the Indus. The contrary is distinctly asserted by Arrian, and on this point it is difiicult to believe that the companions of Alexander could have been in error. NOTE N n, p. 448. THE INDUS IDENTIFIED WITH THE NILE. So imperfect were the geographical ideas commonly entertained previous to this voyage of Alexander, that the king himself when 510 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. he first came to the river and saw crocodiles in it, was convinced that it was the same with the Nile, and wrote a letter to Olympias containing this statement, which appears to have been extant at a later period. (Arrian, Anal). vi. 1 ; Strab. xv. 1, p. 696.) It was only on his march through the Punjab that he obtained more accurate information from the natives, and became convinced that it ulti- mately flowed into the Southern Ocean. It is remarkable to see how in this respect the geographical information of the Greeks seems to have retrograded since the time of Herodotus. No allu- sion is found to the voyage of Scylax related by that historian, which must either have been disbelieved or forgotten, while the just conclusions derived from it by Herodotus had fallen, into the same oblivion. But absurd as was this identification, the general resemblance between the Indus and the Nile, which are constantly brought into comparison by the Greek geographers (Strabo, xv. p. 692, &0.), is certainly such as to justify their observations. The resemblance of the lower valley of the Indus, from the time it has received the waters of the Punjab, with Egypt, is dwelt upon by modern tra- vellers: “ One description (says Mr. Elphinstone) might indeed serve for both. A smooth and fertile plain is bounded on one side by mountains, and on the other by a desert. It is divided by a large river which forms a Delta as it approaches the sea, and annually inundates and enriches the country near its banks. The climate of both is hot and dry, and rain is of rare occurrence in either country.” (Elphinstone’s Oaubul, vol. p. 225.) NOTE ()0, p. 448. WIDTH OF THE INDUS. According to Sir A. Burnes the Indus where it has been joined by the rivers of the Punjab “never shallows, even in the dry season, to less than fifteen feet and seldom preserves so great a breadth as half a mile.” (Travels, vol.i. p. 195.) “ Near Hydrabad it is but 830 yards, at Tatta less than 7 00, and below the village of Hilaya, 15 miles from that town, it does not exceed 600.” (lb. p. 242.) But in one part of its course above Bukkur, and below Mittun Kote (at which place it receives the united waters of the Punjab), it is more widely spread, so that it “ often exceeds a NOTE P P. EXPEDITION OF ALEXANDER. 51 I thousand yards in breadth, and at Mittun was found to be even double that width.” (Ibid. p. 260.) Similarly exaggerated statements were current among the ancients with regard to the principal tributaries of the Indus: but we meet with others of a much more authentic character. Ptolemy the son of Lagus, as cited by Arrian (v. 20, § 8) stated that the Acesines, at the point where Alexander crossed it, was fifteen stadia in width, with a strong and violent current. This was the only one of the rivers of the Punjab concerning the size of which that author had left any definite information: and his accuracy is confirmed by Mr. Elphinstone, who says that the Chenab at the Wuzeerabad Ghat, where he crossed it on the 31st of July, “ lneasured one mile three furlongs and twenty perches, from edge to edge of the water. The soundings were the same as the J elum, fourteen feet the greatest, but the current was more rapid by a knot and a half.” (Elphinstone’s Caubul, vol. ii. p. 423.) The Jhelum (Hydaspes) was found by the same authority to measure at J elalpoor one mile one furlong and thirty-five perches, though it had not then attained its full height (Ib. p. 421). NOTE Pp, p. 449. SOURCE OF THE INDUS. The Afghans, even at the present day, regard the place where the Indus first issues from the mountains as very near the source of the river. (St. Martin, Géographie de l’Inde, p. 44.) The part of its course above Derbend is indeed the most imperfectly known of its whole stream, and it is a remarkable fact that down to the present time no European traveller has ever followed the valley between Acho and Derbend. In the last century modern geographers only knew vaguely that it must rise at some distance from a point where it appeared as a large stream; but even Major Rennell, as late as 1793, supposed it to flow from the north, and that it must have its sources on the western side of the range that runs northward from the Hindoo Koosh, and in which the Oxus also takes its rise. He was thus led to place the sources of the Indus nearly due north from Attock, While he supposed the river that flowed by Ladak, of the existence of which he had heard, to be a branch of the Ganges, and that another branch of the same river took its rise in the Lake 512 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. Mansarowar. (See the map annexed to his Memoir, p. 201.) D’Anville, about twenty years earlier, took much the same view of the sources of the Indus and the Gan ges, while be supposed the Brah- maputra, which he rightly conceived to have its origin in the same mountain group, to be the same with the Irawaddy, and to flow through Pegu on its ‘way to the Gulf of Bengal. (See the map annexed to his Antiguité Géographigue ole l’Inde, published in 1775.) This last error was first corrected by Major Rennell, in 1781. In criticising ancient geographers it is important to bear in mind how imperfect was our knowledge of many countries of the old world down to a very recent period, and how readily even the ablest modern writers have been led into false geographical combinations by imperfect information. NOTE Qq, p. 450. DELTA OF THE INDUS. The statements of ancient writers concerning the ‘width of the Delta, or the extent comprised between its two arms, are equally irreconcilable with the supposition that the existing Delta was meant. Aristobulus estimated the base of the triangle, or the inter- valfbetween the two mouths, at 1000 stadia, but N earchus reckoned it not less than 1800 stadia (Strab. xv. p. 701). Now, according to Burnes, the land embraced by the two actual arms of the Indus extends, at the junction of the rivers with the sea, to about 70 British miles, “ and this, correctly speaking, is the existing Delta of the river.” (Travels, vol. i. p. 208 ; Journal of Geographical Society, vol. iii. p. 115.) But he adds that “ the Indus covers with its waters a wider space than is thus described, and has two other mouths to the eastward of these, the Seer and Koree ;” and with the addition of those (now forsaken) branches, the river presents a face to the sea of about 125 British miles. (Ib.) But these two arms have nothing to do with Tattah, and it is difficult to understand how Sir A. Burnes reconciled the statement which he repeatedly makes, that the Koree was formerly the eastern branch of the Indus, with his adoption of the view (generally followed in his, day) which iden- tified Pattala with Tattah. The two main arms of the Delta of Egypt are more than 90 G. miles in length (as measured on the map, without following their NOTE R R. . EXPEDITION OF ALEXANDER. 513 windings in detail), while the sea front of the Delta, from the Pelusian to the Canopic mouth, is not less than 140 G. miles, or 1400 stadia. NOTE Rr, p. 450. SITE OF PATTALA. This view, which appears to have been first suggested by Pottinger, and adopted by Droysen and Benfey, has been worked out very fully by M. de St. Martin, Geographic de l’Inde, pp. 169- 172, and has certainly great probabilities in its favour. The suggestion first thrown out by Capt. McMurdo, and which Prof. Wilson was inclined to adopt, that the bifurcation in the time of Alexander took place at a point much farther up the river, above Bukkur, where a dry channel or river-bed may still be traced, part- ing off from the main stream, and holding a direction towards the estuary of Koree, is liable to the great objection that the extent of the two arms would in this case enormously exceed those of the Nile, the excess being in this case as great as the deficiency if we place Pattala at Tattah; and it would also render it impossible to find room for the different tribes and cities reduced by Alexander between the confluence of the Indus with the Acesines, and the head of its Delta at Pattala. But the point which both these views have in common, that the eastern mouth of the Indus was in ancient times by the estuary of Koree, close to the peninsula of Cutch, has certainly much to recommend it; and appears to be in accordance with all that we know of the changes that have taken place in this part of the country. Sir A. Burnes, though he continued to identify Pattala with Tattah, speaks of the Koree as “ the eastern, though forsaken branch of the Indus,” and even terms it “ the largest of all the mouths of the river, having become a branch of the sea as’ the fresh water has been turned from the channel.” (Travels, vol. i. p. 7 There appears, moreover, to be authentic evidence of the waters in this instance having been artificially diverted from their course and prevented from flowing in their former channel, with a view to injure the in- habitants of Cutch. (Ib. pp. 215, 309.) At the same time we know that all this neighbourhood has been visited by repeated earth- quakes, and it is not unlikely that changes of level resulting from this cause may also have contributed to alter the geographical VOL. I. 2 L 514 HISTORY OF ANCIENT GEOGRAPHY. GHAP. XII. features of the country. (See Lyell’s Principles of Geology, vol. ii. pp. 98-102, 10th edition; and the valuable paper of Sir Bartle Frere on the Runn of Cutch, in the Journ. Geogr. Society, vol..xl. p. 121.) Gen. Qunningham, who has most recently investigated the subject, concurs in placing Pattala on the same site with Hyder- abad, which occupies a long, flat-topped hill that must have in all ages been favourably adapted forv the site of a city. The name indeed is modern, but it is still known to the people as Neeruncote, by which name it is mentioned in the Arab historians and geo- graphers. (Cunningham’s Ancient Geography of India, pp. 279-285.) NOTE Ss, p. 451. CITIES OF THE MALLI. ‘It must be borne in mind in discussing this question that the Ravee or Hydraotes, which at present joins the Ohenab more than 30 miles above Mooltan, formerly held a separate course much lower down, and passed under the walls of Mooltan, completely encircling that fortress, before it joined the Ohenab. (Cunningham, p. 221.) General Cunningham, who has himself explored this part of the country, has endeavoured to trace the successive marches of Alexander in his expedition against the Malli, and to identify the towns and strongholds that he reduced. He considers the first city taken by him (Arrian, vi. 6, § 2) to be represented by Kot Kamolia, “ a small but ancient town,” about 44 miles to the SE. of the junction of the Hydaspes and Acesines (the point from which Alexander set out), and only a few miles from the west bank of the Ravee. Harapa, a place where there are extensive ruins on the opposite side of the Ravee, about 16 miles from Kot Kamalia, he supposes to be “ the other city ” against which Perdiccas was dis- patched with the cavalry, while Alexander himself attacked the former town (Ibid. § 4). The next city, taken by Alexander after crossing the Hydraotes (Ibid. 7, § 2, 3), he identifies with Tulamba, “ a. remarkably strong fortress ” on the left bank of the Ravee, about 52 miles to the N.E. of Mooltan: while the fourth city, called by Arrian “a certain city of the Brachmans ” (Ibid. § 4) he considers to be represented by the old ruined fort and town of Atari, 20 miles from Tulamba on the road to Mooltan. All these NOTE 'T T. EXPEDITION OF ALEXANDER. 515 sites are marked by the presence of mounds, which, together with the large size of the bricks employed in the ruins, are considered to afford unquestionable evidence of remote antiquity. (Cunningham, pp. 208-229.) The above identifications appear plausible enough. But there is great difficulty in adopting the supposition that the capital city of the Malli ,aeyio-rvy 763v MaAAGw 'miAtg, Arrian, vi. 8, § 4) occupied the site of the modern Mooltan, the strongest fortress in this part of the country. and the natural strength of which must have been still greater when it was surrounded by the waters of the Ravee. But the capital was abandoned by the Malli, without an attempt at resistance, when they heard of the approach of Alexander: they withdrew to the north bank of the Hydraotes and being there attacked again by the Macedonians, fled for refuge to a neighbour- ing city that was strongly fortified (Ibid. § 7). It was in the attack of this last city that Alexander himself was severely wounded and narrowly escaped with his life. General Cunningham has con- founded this city with the capital of the Malli, or combined the two together, which is certainly at variance with the narrative of Arrian. Moreover that author clearly represents the place where the king was wounded as at some distance from the Hydraotes: and his description of the voyage down that river to its confluence with the Acesines, where the main army was encamped, implies one of longer duration than the few miles from Mooltan to the Chenab. Q. Curtius indeed represents it as a four days’ voyage (ix. 23, § 3), but as he is evidently following a different set of authorities from Arrian, no reliance can be placed upon this. NOTE Tt, p. 451. THE OXYDRACZE. The position of the Oxydracae is much more obscurely indicated than that of the Malli. They are mentioned on more than one occasion as having combined with the Malli, and also with the Cathaei, to oppose the progress of Alexander. After the reduction of the several cities of the Malli, the Oxydracae joined with them in sending deputies to tender their submission. But we have no account in Arrian of any military operations against them, or of the capture of any of their cities: and we cannot infer with cer- 2L2 516 HISTORY or ANCIENT GEOGRAPHY. CRAP. xII. tainty from the manner in which he speaks of Alexander’s de- scending the river to attack the Malli and Oxydracae (vi. 4, § 3), that he actually traversed the territories of the latter people. All that we learn concerning them is quite consistent with the suppo- sition that they lay on the east or left bank of the Sutledge—the province of Bahawalpoor—though they may very well have extended as far as its junction with the Indus, and the neighbourhood of Ooch. General Cunningham alone places the Oxydracae to the north of the Malli, which is certainly contrary to the natural inference from the narrative of Arrian. His principal reason seems to be that Pliny speaks of the turning-point of Alexander’s career (meaning probably the spot on the bank of the Hyphasis where he erected his altars) as in the territory of the Sydracae, which is probably the same name with that of the Oxydracae. But the passage in question (in Sydracis Empeditionum Alewandri termino, xii. 6, § 24) is merely one of those passing notices hastily thrown in by the author, and which are of very little value in a geographical point of view. We learn from Arrian that there was considerable confusion among Greek authors with regard to the Oxydracae and the Malli, many writers asserting that the fortress in the attack of which Alexander was so "severely wounded lay in the country of the Oxydracae (a statement which we find repeated by Q. Curtius, ix. 18, § 26), but this he unequivocally rejects as a mistake. He evidently is here writing on the authority of Aristobulus and Ptolemy. The name of the Oxydracse (’O$v8pd|526 : .. Zulu-then am’ A 6 NOTE X X. EXPEDITION OF ALEXANDER. 5I9 excellent dates, though little or no grain. Moreover, time would necessarily be lost in the repeated attempts made to communicate with the fleet under Nearchus. The statement of Strabo that the days’ marches amounted to two hundred, four hundred, and even six hundred stadia (70 miles) a day, is manifestly a great exaggera- tion, but forced marches would no doubt be required in order to pass from one station of wells to another. Such forced marches would however necessitate corresponding intervals of repose. NOTE Xx, p. 456. ROUTE THROUGH GEDROSIA AND CARMANIA. The description given by Sir H. Pottinger (Travels in Beloochi- stan, p. 131-134) of the desert which he traversed between Sarawan and Kullugan so closely resembles those‘ preserved by Arrian and Strabo from the historians of Alexander, that it might almost seem to be translated from them. But the desert in question is situated in the northern part of Beloochistan, to the north of the range of mountains (the Mushti or W ashati) which traverses the country from east to west, and apparently formed the northern limit of Gedrosia, while the route of Alexander unquestionably lay to the south of the same range. Strabo’s testimony to his continued proximity to the sea is precise: and Sir H. Pottinger remarks that “had the Greek historians been even less explicit, the nature of the country alone must have decided any question that might have arisen upon this point” (Travels, p. 264). The interior of the Mekran is indeed still very imperfectly known: though the coast has been of late years thoroughly examined, for the purpose of laying down the line of electric telegraph from Gwadur to Kur- rachee, and several lines of route have been followed, crossing it towards the interior of Beloochistan. But no traveller has as yet traversed its length from one end to the other, in the direction followed by Alexander. So far as we can judge he appears to have kept along a kind of plain or valley, which is found to run nearly parallel to the coast, between the interior range of the Mushti hills and the lower rugged hills that bound the immediate neighbour- hood of the sea-coast. This line of route has been followed in very recent times by Major Ross from Kedj to Bela, and seems to form a natural line of communication, keeping throughout about the required distance (60 to 70 miles) from the coast. (See Proceedings 520 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. of Geogr. Society, vol. xvi. pp. 139—141 and 219.) The arid and impracticable character of the coast district itself, which deterred Alexander from following that route (Arrian, vi. 23) is fully con- firmed byrecent accounts: but the more inland line of march, so far as is yet known, does not appear to traverse any such frightful vdeserts of sand as those described by the historians of Alexander. Nor can the site of Pura, the place which formed the terminus of his toilsome march‘ through the desert, be' determined with cer- tainty. It has been generally identified with Bunpoor, the most important place in Western Beloochistan, or with Puhra, a village in the same neighbourhood, but the resemblance of name is in this case of little value—Poor signifying merely a town—while the remoteness of Bunpoor from the sea, and its position to the north of the central chain of mountains, which Alexander must there- fore have traversed in order to reach it, present considerable difficulties in the way of this view. Unfortunately the subsequent march of Alexander through Carmania is as obscure as that through Gedrosia. The country being comparatively fertile, and his passage being unopposed, there was little to attract the attention of historians, and we are briefly told that the king proceeded into Carmania; and after that into Persia, sending Hephaestion with the bulk of the army by the coast read, while he himself, with a light-armed force, both of horse and foot, took the direct road to Pasargada. Arrian, Anab. vi. 27, §§ 2, 3, 28, § 7, 29, § 1.) But it seems clear that, until this separation, his line of march was throughout not very remote from the sea. When Nearchus arrived at the river Anamis, he learned that Alexander was “ not far off,” an expression subsequently explained to mean that he was at a distance of “five days’ journey.” (Indica, c. 33.) This at once excludes the supposition that Alexander was anywhere in the neighbourhod of Kerman, or the fertile district of northern Carmania. For the city of Kerman is at a distance of eighteen “ menzils” or days’ marches from Bunder Abbas on the Persian Gulf. (Pottinger, p. 227.) The same distance is estimated by Kinneir (Geography of Persia, p. 200) on the authority of a native traveller, at 177 farsangs. It must be somewhat more from the mouth of the Anamis. If indeed he had struck up so far to the north as Bunpoor, his natural line of route from thence to Pasar- gada and Persepolis would have lain through the fertile district of Nurmansheer, and Kerman, which was probably then (as we NOTE Y Y. EXPEDITION OF ALEXANDER. 52 I know it to have been in the times of Ptolemy and Ammianus) the capital of the province. But if Alexander had been following this inland route, his detaching Hephaestion to the sea would be inex- plicable. If, on the contrary, he was still, as we find him when N earchus first rejoined him, within easy reach of the sea-coast, it would be natural for him to continue his march with the whole army through the maritime districts of Carmania and Persia to ' Susa. Instead of so doing, he left his army to follow this route under the command of Hephaestion, while he himself struck up into the mountains to Pasargada. The ancient accounts of Carmania in general difl'er much from the impressions of modern travellers, from which it is clear that a large part of the province is very arid and mountainous. But in such cases it often happens that the character of a particular district is transferred to the whole region. In some portions of Carmania there are tracts of great fertility. The district near the sea-coast around the river Minab (the Anamis 0f Arrian) and that of Giroft, in the interior—supposed by Dr. Vincent to be the place of meeting of Alexander and Nearchus—are both of them fertile and well-peopled districts. The former especially is called by the natives the Paradise of Persia. (Kempthorne in Geo-_ graphical Journal, vol. v. p. 274.) The district of Nurmansheer in the interior is also a fertile and productive region, but this, like Kerman, seems too remote from the sea to agree ‘with the accounts of the meeting of Nearchus and Alexander. It took Sir F. Gold- smid 16 days travelling to reach Bumm, the capital of this district, from Bunder Abbas on the Persian Gulf. (Journal of Geographical Society, vol. xliii. p. 65.) It is curious that a large part of the province of Carmania is still virtually unknown, and just in the direction where Alexander would proceed from Gedrosia to the frontiers of Persia proper, large tracts of country are still marked on the latest map (that of Captain St. John in 1877) as “unexplored.” NorE Yy, p. 4.57. MARCH OF CRATERUS. Dr. Vincent justly observes that “ by a view of the map and a reference to the geographers, we can hardly discover the means of his [Craterus] avoiding some part of that desert on the east of 522 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. Carmania, which the Nubian geographer says is the largest 'in the world ” (p. 340). Yet he assumes that Craterus had experienced none of the difliculties which Alexander had encountered in Ge- drosia. This statement is not warranted by the ancient authorities, who, indeed, ignore all difliculties encountered by Craterus, but do not say that he met with none. The great desert of Kerman, which occupies the northern part of that province, and extends from thence to the confines of Yezd, Khorasan, and Seistan, is a vast tract of the most unmitigated barrenness, and a considerable portion of this, interposed between the fertile districts of Nur- mansheer, in northern Carmania, and the Lake Zurrah in Seistan, must of necessity have been traversed by Craterus with his army. According to Sir H. Pottinger an Afghan army, which invaded Persia in 1719, suffered the most dreadful hardships in this waste, and after one-third of the whole had perished, the remainder reached Nurmansheer with the loss of all their equipage and baggage. (Pottinger’s Travels, p. 229. See also Abbott in Geographical Journal, vol. xxv. p. 34, 37 It has been more recently traversed by Sir F. Gcldsmid 'on his way from Bunder Abbas to Se'istan (Journal of the Geographical Society, vol. xliii. pp. 65-74), who, however, did not encounter any serious difliculties for an ordinary traveller. The line of route he followed must probably be much the same as that of Craterus, though in an inverse direction. NOTE Zz, p. 460. EMBASSY OF THE ROMANS TO ALEXANDER. Both Niebuhr (Roman Hist. vol. iii. p. 169) and Mr. Grote (vol. xii. p. 335) attach much weight, and undoubtedly with good reason, to this argument. Clitarchus appears to have published his his- tory some time before the death of Ptolemy Soter (B.0. 283), and therefore before the war of Pyrrhus in Italy. It is difficult to believe that, until after that event, the Romans could have attracted so much attention among the Macedonian Greeks as to lead to the introduction of such a fiction. On the other hand, the opinion expressed by Livy that the Romans had at this time not even heard of the name of Alexander, is evidently absurd: as there is no doubt of the fact (stated by Arrian from his best authorities) that embassies were sent by other nations of Italy—the Bruttians, Lucanians, and Tyrrhenians. If these nations were familiar with NOTE Z Z. EXPEDITION OF ALEXANDER. the name of the great conqueror, and aware of the commanding position that he occupied, it is incredible that the Romans should not be so likewise. Equally futile is the argument of Arrian that it was not consistent with the republican spirit of the early Romans to have sent such an embassy to a foreign king. He must have forgotten that they had already—some six years before—. sought the alliance of Alexander, king of Epirus, the uncle of the great conqueror, and actually concluded a treaty with him. This circumstance may have naturally led to a more familiar acquaint- ance with the affairs of Macedonia and the neighbouring nations, and adds much to the probability of their having subsequently sent an embassy to the nephew. . Could we feel certain that the statement cited by Pliny was really made by Clitarchus, the probabilities would, as it appears to me, be greatly in its favour : but it is remarkable that Arrian, who also notices the story (Anab. vii. 15, §§ 5, 6), speaks of it as mentioned only by Aristus and Asclepiades, two authors of very little note, and certainly of a later period, and though he discusses at some. length the probability of the incident, never alludes to Clitarchus, though a contemporary, and one of the most popular of the his- torians of Alexander. Still more important is it that Diodorus, who undoubtedly wrote with Clitarchus before him, and followed him as one of his chief authorities (see the remarks of Geier, Scriptores Alexandri Magni, p. 155; and of C. Muller, Fragmenta Scriptorum. Alex. M., p. 75), while mentioning the names of many of the nations that sent embassies on this occasion to Alexander (xvii. 113), says not a word about the Romans, though he at least must have been fully aware of the interest of such an incident, had it really occurred. This portion of the history of Q. Curtius, who was also to a great extent a follower of Clitarchus, is unfortunately lost. The authority of Justin is not worth much, but his silence on such a point is not without its importance. (Justin. xii. 13.) The statement of Strabo (v. 3, p. 232), referred to by Niebuhr (l. c.) that Alexander sent back some pirates from Antium that he had cap- . tured to the Romans, may much more probably refer to Alexander of Epirus than to the great conqueror: but the whole story has a, very doubtful aspect. 524 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XII. NOTE AA, p. 463. THE PALLACOPAS. It is impossible to identify the Pallacopas with any reasonable certainty. ' The same circumstances that had rendered it defective in the time of Alexander for the purposes it was intended to serve, would almost certainly operate again to render useless the new cut made by order of the king: and this would in its turn be super- seded by some later canal. The name is not found in the other historians of Alexander, or in the geographers. All that can be said is that it was a cut on the right (or western) bank of the Euphrates for the purpose of diverting the surplus waters of the river, during times of inundation, into the sandy tracts of Arabia, just as is done at the present day by the canal called Hindiyeh, which quits the Euphrates about 15 miles above Babylon, and forms the extensive marshes and meres on the west and south-west of Hillah. But it is clearly a mistake to suppose, as Col. Chesney did, that the one cut represented the other. (The same view is taken by Mr. Loftus (Chaldoea and Susiana, p. 42) who mistranslates the passage of Arrian to suit it.) The Pallacopas, according to Arrian, quitted the Euphrates 800 stadia below Babylon (vii. 21): and though the distance may be overstated, it must certainly be sought between Babylon and the sea. The suggestion of Capt. Felix Jones that an ancient dry water-course, called the Shat al Roumayieh, from its passing 'by a town of that name, represents the Pallacopas, is more plausible, though the distance from Babylon is much less than that given by Arrian : but no great reliance can be placed on any such identification. The mode in which the Pallacopas is represented on several recent maps, as a canal having a course nearly parallel with the Euphrates and extending all the way to the sea, appears to me wholly at variance with the account given by Arrian of its character and purpose. There is no other authority: but the mention of it by Appian (B. Cir. ii. 153) shows at least that he understood its object as above explained. (525) CHAPTER XIII. VOYAGE OF NEARCHUS. §_1. WE must now return to examine somewhat more in detail the voyage of N earchus, who, as We have already seen, had conducted the fleet of Alexander in safety from the mouth of the Indus to that of the Euphrates. Such an enterprise was undoubtedly in those days a task of difficulty and danger, requiring great energy as well as prudence on the part of the commander. It moreover derived, even in ancient times, great additional interest from the circumstance of its being the first—or at all events generally believed to be the first-— navigation of the Indian Ocean. The much more arduous voyage of Scylax in the same seas was either forgotten or dis- believed; and it seems certain that no full or authentic record of it had been preserved.1 But the voyage of N earchus has attracted an increased, and in some degree a disproportionate amount of attention, in modern times, from its having been preserved to us in a form so much more complete and au- thentic than any other record that we possess of a similar character. A detailed narrative of his voyage was written by N earchus himself, of which Arrian has transmitted to us an abstract so full that it may to a great extent—for geographical purposes at least,——stand in the place of the original work. And it is interesting to see how accurately this account is found to tally, both in the geographical details, and in the particulars ‘transmitted concerning the country and its in- habitants, with the results of recent observations. In proportion as we have in modern times become better acquainted with 1 See the observations in Chapter VII. p. 227, Note B. 526 HISTORY 0E ANcIENT GEOGRAPHY. can». xIII. the wild and barren shores explored on this occasion, do we find the agreement with the statements of N earchus more complete.2 Much labour has been bestowed by several modern writers, especially by the learned Dr. Vincent,3 upon the examination and verification of all the details transmitted to us by Arrian, who has given us, with very few exceptions, a complete journal of the voyage, day by day, recording both the stations at which they brought to for the night, and the distances traversed.‘ Of the mode in which these last were computed we have un- fortunately no indication.5 It would be wholly foreign to the plan of this work to follow these inquirers into the minute details of their investigations, especially as the greater part of the coasts which were thus explored are of singularly little interest in themselves.6 The really important results of the voyage of Nearchus are comparatively few; and lie within small compass; it will be therefore sufficient to give enough .of the details of his narrative to show their close agreement with the geographical conditions as they are now known to us. 2 Note A, p. 542. 3 His elaborate dissertation on the voyage of Nearchus was originally published in a separate form in 1797 ; and subsequently republished, together with that on the Periplus of the Ery- threan Sea, as the first volume of his Commerce and Navigation of the An.- cients in the Indian Ocean (2 vols. 4to, Lond. 1807). It is from this later edition that my citations are made. Dr. Vincent's investigations were based in some degree upon imperfect information as to the actual geograph- ical details of the countries in question, and for this reason have in some in- stances been superseded by later and more accurate knowledge. These more recent sources of information have been well turned to account by Dr. C. Miiller, in his valuable commentary on Arrian’s Indica, in the first volume of his Geo- graphi Graaci Minores (Paris, ed. Didot, 1855), which now constitutes the most complete and satisfactory authority upon the subject. The earlier dissertations by D’Anville and Gosselin are now of little value. 4 Arrian, Indica, c. 21, foll. The citations from this work in the present chapter all refer to the edition by C. Muller. 5 Note B, p. 544. 6 Lieutenant Kempthorne in his Notes on the Eastern Shores of the Per- sian Gulf (published in the Journal of the Geographical Society, vol. v.), remarks that “ the whole of this coast from the Indus to Bussora, a distance of more than twelve hundred miles, is, with few exceptions, ‘ one vast arid and sterile waste, with high mountains rising at the back, wholly destitute of both trees and vegetation,’ ” p. 270. This memoir of Lieut. Kempthorne contains much valuable information as to the coasts and localities he visited, but his attempts to connect these with the voyage of Nearchus are for the most part hasty and ill considered. . OHAP. XIII. vovAcE or NEAROHUS. '527 § 2. It is unfortunate that Arrian has not preserved to us any statement of the number or description of the ships employed on this occasion. They formed without doubt but a small proportion of the numerous fleet with which Alexander had himself descended the Indus; and it may safely be assumed that they would consist principally of ships of war, or “long ships,” as the Greeks termed their galleys for distinction’s sake. But they were undoubtedly accompanied also by transports, apparently vessels of no great size, which had already formed part of the fleet on the Indus.7 But we have no means of forming any estimate of the number of these, or of the troops and seamen on board. It is still more to be regretted that we have no distinct account of the point from which the fleet took its departure.8 It is certain that this was considerably lower down than Pattala, as we learn from the journal of the voyage that it was only 150 stadia from the sea. Alexander had indeed con- structed naval stations (N austathmi) at several points in the Delta of the Indus, and it appears that the one at which the fleet of N earchus was assembled, and the last preparations made for the voyage, was situated on the western arm of the river, probably on a large’ island called Killouta. But in a locality where the channels and mouths of the river are per- 7 In the few passages in which they are mentioned these vessels are called rcépicovpoi, a term usually applied to small vessels; and the same that is employed by Arrian (Anab. vi. 2, § 7) in describing the fleet that was col- lected by Alexander for the descent of the Indus. 8 Pliny, following Onesicritus, says that the fleet took~its departure from a place which he calls Xylenepolis; but he complains that his author does not more accurately describe its position. “ I’rimumque Xylenepolis ab Alexandro condita, unde ceperunt exordium, juxta quod flumen, aut ubi fuerit, non satis explanatur ” (Nat. Hist. vi. 23, § 96). The words "juxta quod flumen ” can only mean, “which of the branches of the Indus.” N earchus, as reported by Arrian (Indica, c. 21, § 2), says only that they took their departure from the naval station (ea r06 vavo'ra'ieuov), and Alex- ander had caused so many of these to be constructed in the delta of the Indus, that the expression conveys no information. The suggestion of Dr. C. Mtiller, that it was situated on the island of Killouta (Kumov'rc‘z), men- tioned by Arrian in his other work (Anab. vi. 19) as the place where Alex- ander stationed his fleet in descending the Indus, is plausible enough, but rests on no definite foundation. 528 HISTORY or ANCIENT GEOGRAPHY. carp. x111. petually shifting, it is impossible to fix its position with any certainty. Three days were employed in dropping down the river, the bar at the entrance of which presented so formidable an obstacle as to induce the Greeks to cut a channel through a narrow bank of sand in order to reach the open sea.9 Having thus entered the ocean they proceeded as far as a sandy island called Crocala, and the next day reached a large and con- venient harbour, to which Nearchus gave the name of the Port of Alexander. §3. Here they remained for some time. N earchus had originally intended not to commence his voyage until a later period of the year, when the monsoon, concerning which the Greeks had obtained accurate information, should be fully past; but the attacks of the natives had become so trouble- some, after the departure of Alexander, that he had set out earlier than he designed. The consequence was that he found the south-west monsoon still blowing with great violence, and having found a secure station in the Port of Alexander, he remained there twenty-four days until the wind abated. It is probable that the harbour which afforded him this opportune shelter was no other than the port of Kurrachee, now one of the most frequented sea-ports in this part of India.1 During their stay here the troops subsisted to a great extent on shell- fish, the enormous size of which, as compared to those in the European seas, naturally struck them with astonishment. When the fleet was able to put to sea again (about the beginning of November), it proceeded along the coast towards the west, and in five days reached the mouth of the river Arabis, having accomplished a distance (according to their reckoning) of a thousand stadia (100 G. miles) from the mouth of the Indus.2 This distance is undoubtedly over-stated and the details of this part of the voyage cannot be made out with any distinctness; but the Arabis, at the entrance of which ‘ they found a commodious port, is undoubtedly the Poorally, 9 Arrian, Indica, c. 21, § 6. 1 Note C, p. 546. 2 Ibid. c. 25, § 3. CHAP. XIII. VOYAGE OF NEARCHUS. 529 the only considerable stream on this part of the coast, near the mouth of which is a small town called Somneanee, that is a place of some trade.3 § 4. The coast thus far had been held by the Arabies, whom Arrian calls the last of the Indian tribes. From hence the navigators proceeded along the coast of the Oritae, and after three days’ voyage—during which they encountered a severe gale, in which‘ they lost three of their ships—arrived at a place called Cocala, where, though there was no harbour, but merely an open roadstead, N earchus landed his crews in order to refresh them after the fatigues they had undergone. Here he was met by Leonnatus, who (as we have seen) had been left behind by Alexander for that purpose, and was able to take on board provisions for ten days, as well as to repair his ships, and recruit his crews out of the forces of Leonnatus.4 Sailing hence with a fair wind he made good a course of 500 stadia to the mouth of a river, or rather torrent, called the Tomerus, where his landing was opposed by the natives in arms, and had to be effected by force, under circum- stances strongly resembling those to be found in many modern voyages of discovery. The victory was indeed an easy one: the natives appear to have been mere savages, unacquainted with the use of iron or any other metal, and using sharp stones for knives, and spears with the points hardened in the fire.5 Here Nearchus again remained for six days, and then made a short day’s voyage to a place called Malana, which is termed by Arrian the limit of the Oritae. This part of the voyage presents no geographical difficulties. The site of Cocala cannot indeed be identified, because there is nothing to mark it, the locality being a mere open beach, off which it was possible for the ships to anchor. But the river Tomerus corresponds with that now called the Muklow, or Hingul: and Malana may be clearly recognized in a head- 3 Kempthorne, in Geogr. Journal, 4 Arrian, Indica, c. 23. vol. v. p. 264. Pottinger’s Travels 5 lbid. c. 24. in Beloochistan, p. 11. von 1. 2 M 530 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIII. land which still preserves the name of Cape Malan or Malin. The distances however, as usual, are greatly overstated, the whole length of the voyage along the coast of the Oritae being estimated at 1600 stadia (160 G. miles), while the real dis- tance does not exceed 900 stadia in a direct line, and the windings of the coast are very trifling. § 5. From this point began the navigation along the barren and desolate coast of the Ichthyophagi, the name given by the Greeks to the poor miserable savages who inhabited the sea- coast of the Mekran, or Gedrosia. This sterile tract extends for a space of more than 400 G. miles from Cape Malan to Cape J ask, preserving throughout a very uniform direction from E. to vv., and presenting but few prominent geographical features. It is well described by Nearchus, in much the same terms as have been repeated by modern voyagers, as a barren and inhospitable coast, though presenting from distance to distance spots where palm-trees grew, and water was to be procured. Even where the coast was apparently a mere tract of barren sand, water was generally to be obtained (though often of bad quality) by digging wells of moderate depth near the shore.6 The difficulties of the fleet on this account were consequently not so great as had been encountered by the army under Alexander. The time of year also was favour- able, the winds at this season blowing generally prettily steadily from the land, and we do not hear of their encoun- tering any gales, or being detained by contrary winds, through- out this part of their passage. Both the dangers and the hardships of this part of the voyage appear indeed to have been greatly exaggerated. Nor can we wonder at this. They were proceeding along an unknown coast, with a vast and unknown ocean beside them: and such exploring expeditions were almost entirely new to the Greeks. It was doubtless the same feeling that led them vastly to overrate the distances 6 Indica, c. 26. The same thing is (See Vincent, p. 234, and the authorities found to be the case at the present day. quoted by him.) . CHAP. XIII. VOYAGE OF NEARCHUS. 531 actually traversed. The voyage along the coast of the Ich- thyophagi, which had occupied them twenty days,7 was esti- mated by N earchus at “a little more than 10,000 stadia” (1000 G. miles), which, after making reasonable allowance for the windings of their course, along a coast offering few considerable sinuosities, is nearly, if not quite, double the truth.8 The natives of these sterile shores subsisted, as they do at the present day, almost entirely upon fish, which they fre- quently devoured raw, at other times drying it in the sun, and pounding it into a sort of meal, of which they formed a kind of cakes. Even the few horses and cattle they possessed were compelled to subsist on the same food.9 They had very few boats and relied principally upon the supply of fish left stranded on the beach, or caught in stake-nets set up near the shore. Even their huts were constructed in great part of fishes’ bones: in those of a superior kind they employed the bones of whales by way of beams or door-posts. This descrip- tion of the natives and their habits of life coincides precisely with the observations of modern voyagers.1 So strongly marked are the natural conditions, by which those habits have been inexorably prescribed.2 ’ See the careful computation by Dr. Vincent. Pliny, however, says that it took them thirty days, and his authority on a point of this sort is not wholly worthless. 8 It may be observed that 10,000 stadia in 20 days would give 500 stadia a day, which is just the average of a day’s voyage, as estimated by Scylax. This gives rise to a suspicion that the statement of the distance as given by Nearchus was merely based upon this rough mode of computation. 9 Indica, c. 29; Strabo, xv. 2, § 2. The same thing takes place at the present day. Lieut. Kempthorne says : “ The inhabitants still live entirely on fish, the cattle having much the same diet as their masters, for the country is wholly destitute and barren, and yields no sort of grass. Vast stores of oysters, crabs, and all kinds of shell-fish are found on the coast, of which Nearchus’s description is generally very accurate. In many places, both here and in Arabia, the cattle are fed entirely on dried fish and dates, mixed together, on account of the great scarcity of grass in these sunburnt and sandy regions ” (Journal of Geogr. Soc. vol. v. p. 270). The strange assertion of Pliny (H. N. vi. 23, § 95), that Alexander “ forbade the Ichthyophagi to subsist on fish,” would have been equivalent to con- demning them all to starvation. 1 Kempthorne, l. 0. pp. 270—273. The authority of two earlier voyagers, Capt. Blair and Lieut. Porter, is quoted by Dr. Vincent to the same effect. 2 The modern name for this country, the Mekran or Mekran. is in all proba- bility derived from Mahi-Khoran, “fish- eaters,” and thus exactly corresponds 2M2 S32 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIII. § 6. The geographical details of this part of the Paraplus offer no special interest, but they may for the most part be identified without difficulty, after making due allowance for the exaggeration of the distances. The first day’s voyage from Malana brought them to a place called Bagisara, where there was a commodious port, after which they had to round a lofty promontory standing far out into the sea. This is clearly the headland now called Cape Arabah, on the east side of which is a bay, forming a deep and commodious port. Two days’ voyage from thence brought them to a place called Calama, where there was a village and a few palm-trees. This name is still retained by a small river called the Kalami: and - the island mentioned by Nearchus as about 100 stadia distant from the mainland is readily recognized in that now called Ashtola, one of the few islands along this coast, from which it is really about twelve miles distant.3 The next point of importance was a lofty headland called Mosarna, projecting 150 stadia into the sea, on the west side of which they found a well-sheltered port.“ This can be no other than the promon- tory now called Passenee, or Pasni, with a town of the same name, but the town and port are on the east side of the headland, instead of the west. Here they found C only a small village of fishermen, but obtained an important assistance by taking on board a pilot who was acquainted with the coast from thence to the con- fines of Carmania, and under his guidance and direction the remainder of the voyage presented comparatively little diffi- culty. From this time we find them sailing much at night in order to take advantage of the land breeze. Meanwhile their provisions were beginning to fail them, especially corn, of which they do not appear to have been able to get any addi- tional supply after leaving Cocala. But, on the sixth day after leaving Mosarna, having found a town on a hill, with to the appellation of Ichthyophagi 3 Kempthorne, p. 266. See Note D, given them by the Greeks. (See p. 547. Hughes’s Baloochistan, p. 152.) 4 Indica, c. 26. CHAP. XIII. VOYAGE OF NEARCHUS. some signs of cultivation round it, N earchus landed his troops, and by making a hostile demonstration, compelled the inha- bitants to furnish him with such stores as they possessed. These were however for the most part limited to cakes of meal made from dried fish, in the manner already described, with a slight mixture only of wheat or barley-flour. This town, the name of which is not given by Arrian, appears to have been situated on the bay of Gwettur, but the exact site cannot be identified.5 § 7. From this point their voyage was conducted with much greater expedition than before. They had evidently acquired greater confidence in their navigation, besides having the advantage of a native pilot; and the crews suffered so much from scarcity of provisions, especially from the total want of flour or grain of any kind, that N earchus was obliged to press the voyage by every means in his power. They thus accom- plished a distance according to their own computation of 3750 stadia in six days—a rate of progress far exceeding what they had performed in the earlier part of their voyage. In the last instance they sailed on through a whole night and day continuously, in order to reach a long low headland, which formed the boundary between the Ichthyophagi and Karmania, and the next day, after rounding this point, they found them- selves at a place called Badis, in a land of comparative fertility, where they were able to procure both dates and grain in abundance. The site of Badis may be fixed at, or in the imme- diate neighbourhood of the modern town of J ask; and the low promontory is undoubtedly Cape J ask, which forms a pro- minent feature on this line of coast.6 5 Indica, cc. 27, 28. 6 There is considerable confusion (as Dr. Vincent has pointed out) in the earlier descriptions of this coast, be- tween Cape J ask and Cape Bombareek, the next headland towards the N.W., which is also a long low point, though marked by a singular detached rock. This is the Carpella of Ptolemy, but is not mentioned by Nearchus, while on the other hand no mention of any promontory answering to Cape J ask is found in the Alexandrian geographer. There can be no doubt as to the cor- rectness of the identification of Cape Jask (Vincent, vol. i. pp. 280, 285): the uncertainty of the modern names may serve as a warning in similar cases which occur so frequently in comparing the accounts of ancient geographers. 534 HISTORY or ANCIENT GEOGRAPHY. can». XIII. § 8. Before we proceed to follow the fleet on its subsequent progress along the coasts of Carmania and Persia, we must advert to two other points connected with the earlier part of the voyage. The first of these is the curious account given by N earchus of their battle—the expression is not exagger- ated—with a swarm of whales, which took place off a town called Cyiza. The presence of these sea-monsters was first announced by the columns of water shot up into the air by their “ blowing ;” a phenomenon which was at first taken for water- spouts, but when the sailors were told by their native guides that they were produced by whales, “ they were so terrified that the cars fell from their hands.” N earchus, however, encouraged them, and having drawn up his ships “in order as if for battle,” commanded them all on a given signal to row rapidly forward towards the animals, and as they drew near to raise loud shouts, while the trumpets sounded for the attack, and the rowers made as much noise as possible with their oars. The astonished whales naturally plunged into the depths of the sea, and though they afterwards reappeared in another spot, the appre- hensions of the sailors were removed, and Nearchus was hailed with loud applause as the saviour of the fleet.7 Such was the terror of the crews on this occasion that it appears to have produced as much effect on their minds as all their sufferings from hunger and other hardships. At the present day whales are still frequently met with in this part of the Indian Ocean, and it is not uncommon for a steamer bound from Aden to Bombay to encounter “ a school” (as it is termed) of whales similar to that which caused such alarm to the fleet of N earchus. They however rarely approach so near the coast. § 9. A much more startling assertion is, that while the navigators were following the coast of India (under which head N earchus included the territory of the Arabies and Oritee, but not that of the Ichthyophagi), they found the shadows not to follow the same rule as in other countries, but either the sun 7 Arrian, Indian, 0. 30. Strabo, xv. 2, § 12, p. 725. \- CHAP. XIII. VOYAGE OF N EARCHUS. 5 was vertical at noon, or the shadow was cast to the south. He added, that the constellations and stars, which had usually been high in the heavens, now rose only just above the horizon, and those that were elsewhere constantly visible, rose and set again after a short interval.8 Both these phenomena would of course be really observed by navigators in the Indian Ocean who advanced far enough to the south to be well within the tropic : but it is certain that no such effect as that first stated could be witnessed by N earchus and his comrades, who at no period of their voyage were in a lower latitude than 24° 7 0', or more than a degree north of the tropic. Moreover, the time of year (November) was one in which the sun would be far to the southward of the equator, and therefore no such appearance could be seen, even had the voyagers penetrated—which assuredly they did not—far within the tropic. The only solu- tion of this difficulty that seems to present itself is, that Arrian, though in general reporting his authority with great clearness, has in this instance misunderstood his author, and represented N earchus as stating, as a matter of his own empe- rience and observation, that which he had only reported as a fact witnessed by those who continued the navigation of the Indian coasts farther to the south.9 § 10. After having refreshed his crews at Badis, N earchus continued his voyage along the coast of Carmania, which, as he correctly observed, had from this point a general direction towards the north-west, and after proceeding 800 stadia they came in sight of the lofty promontory, called by the natives Maceta, which stands on the opposite side of the strait at the entrance of the Persian Gulf, and being only about 30 G. miles distant would of course be readily seen from the Persian coast.1 Here Onesicritus, who appears to have acted as second 8 Arrian, Indica, c. 25. termination of a craggy ridge of moun- 9 Note E, p. 548. ‘ tains running through the whole pro- 1 Cape Maceta is undoubtedly the vince of Oman in Arabia, and here same with the conspicuous headland ‘ abutting on the sea. It is so high as now called Cape Mussendom: it is i to be visible, not only from the coast lofty and rugged, and forms in fact the immediately opposite, but all the way HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIII. in command, urged N earchus to cross the straits to this head- land, and prosecute their exploration along the coast of Arabia. Fortunately for the fleet N earchus opposed this suggestion, and determined to continue the voyage along the eastern coast. Here two days more brought them to the mouth of the river Anamis, in the midst of a fertile district called Harmozia, where N earchus halted to repose his forces after all their toils. It was here that they unexpectedly fell in with a Greek, who had strayed from the army of Alexander, and learned to their great satisfaction that the king himself was encamped within a dis- tance of only five days’ journey in the interior. N earchus himself, having provided for the safety of the ships by drawing them up on the shore, and defending them with a rampart and trench, hastened to report to Alexander the safe arrival of the fleet ; an announcement that was received with every demonstration of joy.2 The river Anamis, which from this circumstance assumed so important a position in the narrative of N earchus, is clearly the same with the Minah or Minab, a considerable stream that flows into the northern angle or bight of the bay formed by the Persian coast opposite to Cape Mussendom. It flows through a very fertile district ;3 and the name of Harmozia, mentioned by N earchus, was perpetuated down to a recent period in that of Ormuz, which became in the middle ages the centre of a flourishing and opulent monarchy.4 The position of the camp of Alexander, where that monarch was evidently making a pro- longed halt, is too vaguely indicated to be determined with ' any certainty : Dr. Vincent would place it conjecturally in the district of Giroft or Jeruft, a fertile tract in the interior, about from Carpella (Cape Bombareek). See Vincent, vol. i. pp. 318, 321 ; and Kempthorne, in Geogr. Joarn. vol. v. p. 272. 2 Arrian, Indica, cc. 33—35. 3 The country adjoining the Minab is one of the most fertile in Persia, and is termed by the natives “ the Paradise of Persia” (Kempthorne, p. 274). 4 The name of Ormuz was originally given to a town or district on the main- land, and was afterwards transferred to the island more generally known by that name, when that became the centre of trade, and consequently the capital of all the surrounding coasts. CHAP. XIII. VOYAGE OF NEARCHUS. 85 G. miles from the mouth of the Minab.5 The suggestion is at all events a plausible one, and would suit well with the statement that it was from this point that Hephaastion was appointed to lead the main body of the army by the lower road, through Laristan, to the sea.6 After a few days spent in rejoicings and festivities, N earchus returned to the fleet and resumed his voyage along the coast of Carmania, with the view of conducting the fleet to Susa. But his progress from thence along the eastern shores of the Per- sian Gulf presents comparatively little interest. Though it still partook in some degree of the character of an exploring voyage, as these coasts had never been navigated by Greeks, and their details were doubtless imperfectly known, there was a certain amount of trade carried on along them by native vessels, and the general direction and character of the coast were undoubtedly known to the Persians. From this time therefore the voyage became rather what we should term in modern days a surveying voyage than one of actual discovery. § 11. Immediately after leaving the mouth of the Anamis N earchus mentions their passing by a small island, barren and rugged, after which they came to a much larger one that was fertile and inhabited, which he calls Oaracta. This is clearly the large island of Kishm, at the entrance of the Persian Gulf, while the barren islet, to which he gives the name of Organa, was destined at a later period to become one of the most important centres of commerce in the East, and obtained a world-wide celebrity under its mediaeval name of Ormuz.7 The succeeding points in the voyage have no particular interest. After touching at several other islands, as well as at a small town on the mainland called Sisidone (probably the modern Duan) they visited a small island, the name of which is not given, but which is noted by Nearchus for its pearl 5 Vincent, p. 338. The district of 6 Arrian, Anab. vi. 28, § 7. Sec Jeruft is described by Mr. Abbott in Chapter XII. p. 457. g the Journal of Gcogr. Soc. vol. xxv. l 7 Arrian, Indica, c. 37. See Note F, p. 46. I p. 549. 538 HISTORY or ANcIENT GEOGRAPHY. can». xIII. fishery, which was carried on in the same manner as in the Indian Ocean.8 It is remarkable that this is the only notice found of the pearl fisheries for which the Persian Gulf is now so celebrated.9 As they proceeded up the sea-coast of Persia Proper, the navigation became more intricate and difficult, the coast being described as abounding in shoals, rocks, and reefs, a character fully confirmed by modern observations. On one occasion three of the ships got aground on a sand-bank at low water, but were got off in safety,1 and N earchus had the satis- faction of accomplishing this difficult part of the voyage without the loss of a single vessel. They were however com- pelled to halt at the mouth of a river named Sitacus (the modern Jayrah) for not less than twenty-one days to refit the ships: and here also they laid in a fresh supply of provisions. From hence to the head of the Gulf the rivers and headlands may be identified with little difficulty: but the only marked geographical point is a peninsular headland named Mesembria, which is clearly the same occupied by the modern town of Abu-Shehr, or Bushire, now the principal trading-place on the Gulf. N o towns of importance were met with; a circumstance readily accounted for by the barren character of the shores. N earchus indeed well describes the whole province of Persia as divided into three natural zones or belts, of which that along the sea-coast is sandy and barren from the extreme heat: farther inland comes a tract of great fertility, which enjoys a temperate climate, and produces all kinds of fruit, vines, &c., in great abundance, while it contains also extensive pastures and forests, and is traversed by abundant streams. North of this again is a cold and snowy region of a rugged and moun- tainous character.2 This description is in substance repeated 8 Arrian, Indica, c. 38, § 3. called the Bassadore Bank, where 9 See Chapter XII. p. 461, note. Lieut. Kempthorne himself got aground 1 It'appears probable that the place with his surveying vessel, the ‘ Clive ’ where this accident occurred is a long (Journal of Geogr. Soc. vol. v. p. 280). shoal or sandbank near the western 2 Arrian, Indica, c. 40. extremity of the island of Kishm, CHAP. XIII. VOYAGE OF NEARCHUS. by Strabo,3 and entirely agrees with the observations of modern travellers. § 12. The boundary between Persia Proper and Susiana was marked by a river called by N earchus the Arosis (by Strabo and other writers the Oreatis), which, as he remarks, was the largest stream he had seen since leaving the Indus. This is " clearly the modern Tab, sometimes also called the Endian, or Hindian (from a city of that name on its banks), which is decidedly the most important stream that flows into the Persian Gulf on its eastern side. From hence the shores became so shoal and muddy, that the ships could no longer approach the .‘land at night, as they were accustomed to do; but held their course along the outskirts of the mud banks till they reached a place called Diridotis, at the mouth of the Euphrates, which, though a mere village, was a place of considerable trade in the aromatic productions of Arabia, which were brought thither by merchants from that country} It is singular that though we are distinctly told in the first instance that N earchus had been commissioned by Alexander to bring the fleet on to Susa,5 he had thus carried it to the mouth of the Euphrates, as if with the object of ascending that river to Babylon. But having learnt at Diridotis that Alexander himself was on his way to Susa, he turned back, and after retracing his course for some distance, and passing the marshy lake which was formed by the waters of the Tigris, he entered the river Pasitigris, and ascended it with his whole fleet as far as the point Where Alexander had thrown across it a bridge of boats, for the passage of his main army to Susa. Here the land and sea forces were once more united, and the voyage of N earchus was at an end.6 § 13. The voyage had occupied almost exactly five months7 3 Strab. xv. p. 727. 7 See the careful computation of Dr. 4 Arrian, Indica, c. 41, § 7. See Vincent, p. 495. Pliny says that the Note G, p. 550. actual voyage had lasted less than 5 Id. ib. e. 36. three months; but it was nearly seven 6 Id. ib. c. 42. Concerning the Pasi- since Alexander had left them at Pat- tigris of Nearchus, see p. 458, note. tala. Neither statement is accurate. 0 540 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIII. from the time the fleet left the Indus: but of this period con- siderable intervals had been taken up by long halts; especially that at first starting of 24 days in the port of Alexander, and more recently of 21 days in the mouth of the Sitacus. But even allowing for all such deductions, their progress had been unusually slow, a circumstance which was owing in great part to the nature of the voyage, the object of which, as N earchus distinctly urged, was not merely to accomplish the navigation from point to point, but to examine the coasts as they went along, observe the nature of the country, and ascer- tain what towns or harbours were to be found.8 The fulfilment of these purposes necessitated a very different rate of progress from that of an ordinary voyage, while under any circum- stances the progress of a large fleet will be much slower than that of a single well-appointed vessel. Hence it is altogether delusive to refer to the voyage of Nearchus as a specimen of the rate of sailing of ancient navigators, and the distances that they could accomplish in a given time. The success with which N earchus had completed this enter- prise, was a source of additional satisfaction to Alexander, as giving him favourable prospects for the scheme which he undoubtedly entertained of following it up by a similar voyage along the coasts of Arabia from the Persian Gulf to the Red Sea; a much longer and more perilous navigation, yet one which would probably have presented no insuperable obstacles to so able and cautious a commander as N earchus. Had Alex- ander lived to see the completion of this greater design, he would not only have made a most important addition to geo- graphical knowledge, but would have opened up the way for direct trade between Egypt and India, and have thus laid the first foundation of that regular intercourse with the latter country which assumed so much importance at a later period. As it was, the voyage of N earchus was destined to remain an isolated effort, attended with very little result that had any 8 Arrian, Inclz'ca, c. 32. CHAP. XIII. VOYAGE OF NEARCHUS. 54.1 immediate bearing upon the commerce or civilization of the ancient world. It was a link in a chain of which the others were still wanting. It is indeed of peculiar interest to us as the first voyage of discovery of which we have any detailed narrative ;9 and it is certain that the commander displayed no common amount of ability in its execution. But while doing justice to the really great qualities of which N earchus gave proof—to his energetic perseverance and courage, combined with prudence and caution—it is idle to compare him, as has been done by Dr. Vincent, to such navigators as Columbus and Vasco de Gama, whose exploits have exercised an enduring influence upon all succeeding ages. 9 That of Hanno is, as we have seen, so imperfectly known to us as to admit of no comparison in this respect. 542 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIII. NoTE A, p. 526. COMPARISON WITH PLINY. WE cannot better appreciate the value and importance of the unusually authentic form in which the voyage of Nearchus has been recorded to us by Arrian, than by comparing it with the account of the same voyage as given by Pliny. (Nat. Hist. vi. 23, § 96-100.) That writer appears to have followed exclusively the authority of Onesicritus, without comparing it with the more authentic narrative of Nearchus: but even that of Onesicritus he in reality quotes only at second hand, from the work of J uba the Mauritanian, who had doubtless given a mere compendium or abridgement of the original. (This appears to be clearly the mean- ing of the words “ indicare convenit quos prodit Onesicritus, classe Alexandri circumvectus in mediterranea Persidis ex India, narrata proxime a Juba,” § 96.)1 The words “Onesicriti et Nearchi navigatio” correctly describe the voyage, but certainly do not imply that he had consulted the work of Nearchus himself: while the statement that follows, that it contained no regular enumeration of the halting-places or distances (nec nomina habet mansionum, nec spatia) is glaringly false, if applied to the methodical and regular journal that we have before us. In the following extract he gives indeed a certain number of geographical names of rivers, head- lands, &c.; but none that were mere halting-places (mansiones), so many of which occur in the work of Nearchus, nor has he in any instance given the distances from point to point (spatia). The information he has furnished us is in this respect perfectly in accordance with what he tells us of the authority from which he wrote. Such an enumeration however—though very meagre and 1 This is denied by M. Urlichs in his Vindiciaz Plinianaz (p. 95); but his view that the words “ narrata proxime a J uba ” refer to the inter- mediate stages between the voyage of Onesicritus and that which Pliny de- scribes as practised in his own day appears to me entirely untenable. Nor can I at all concur with him in thinking that Pliny must have seen the original work of Nearchus, because he includes him in the list of his authorities in the first book. It is abundantly evident that Pliny cites many of these autho- rities only at second hand. NOTE A. VOYAGE OF NEARGHUS. 543 unsatisfactory as compared with the narrative preserved to us by Arrian—would not be without its value if the selection were made with method and accuracy. But unfortunately, whether from the fault of his authorities or his own, the abstract given by Pliny is so utterly confused and inaccurate that it may be safely pronounced altogether worthless. It will be sufficient here to give that por- tion of it which relates to the voyage from the mouth of the Indus to the entrance of the Gulf of Persia. After having justly censured his author for omitting to indicate clearly the point from which the fleet took its departure, he proceeds: “ Heec tamen digna memoratu prodentur: Arbis oppidum a Nearcho conditum in navigatione ea, flumen Nabrum navium capax, contra insula distans LXX stadia, Alexandria condita a Leonnato jussu Alexandri in finibus gentis, Argenus portu salubri, flumen Tomberon navigabile, circa quod Pasirae; deinde Ichthyo- phagi tam longo tractu ut XXX dierum spatio praenavigaverint; insula qum Solis appellatur et eadem Nympharum cubile, rubens, in qua. nullum non animal absumitur, incertis causis. Ori gens, flumen Carmaniae Hyctanis portuosum et auro fertile; ab eo primum septentriones apparuisse adnotavere. Arcturum neque omnibus cerni noctibus, nec totis unquam. Achaemenidas usque illo tenuisse. Aeris et ferri metalla et arsenici et minii exerceri. Inde Promontorium Carmaniee est, ex quo in adversa ora ad gentem Arabiae Macas trajectus distat L M. p. Insulae tres, quarum Oracla tantum habitatur aquosa, a continenti XXV M p. lnsulae quatuor jam in sinu ante Persida. Circa has hydri marini vicenum cubi- torum adnatantes terruere classem.” (§§ 97, 98, ed. Sillig, whose text has been followed.) It is hardly worth while to examine this extract in detail. This has already been done by Dodwell in his Dissertation on the Voyage of Nearchus; by Dr. Vincent (vol. i. p. 7 0-7 6), and by Geier (Alex. Magm' Histor. Scriptores, pp. 80, 81, 104). The slightest comparison with the authentic narrative as preserved to us by Arrian will be sufficient to show that while Pliny has retained some names correctly, or in a form that can be easily recognized, and has preserved some interesting detached facts, the whole has been jumbled together in so confused a form, that we should be unable to derive from it any geographical conclusions at all, if we were not provided with the means of doing so by the assistance of other and more authentic sources. Fortunate is it for us that we 54.4 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIII. possess a journal, of which (as Dr. Vincent remarks, p. 76) “the accuracy is as conspicuous as the inaccuracy of Pliny is demon- strable.” ' It is remarkable that Pliny in two or three other passages (vi. 107, 109, 124) cites Nearchus as his authority for distances along the coast, though he appears to have found none in the narrative which he had before him, when he wrote the passage which we are now considering. But in these cases also there can be very little doubt that Pliny quotes him only at second-hand, and has taken the statement of distances from some intermediate compiler. The numbers given are in every instance erroneous, but this may arise merely from the coruption of the text. NOTE B, p. 526. ESTIMATE OF DISTANCES BY SEA— SUPPOSED DIFFERENCE OF STADIA. No ancient writer (as far as I am aware) has preserved to us any account of the mode in which ancient navigators computed or estimated the distances traversed by them at sea. N o allusion is found to any process analogous to that of the modern log : a simple device, but by no means so simple that we are entitled to assume it to have been in use in all ages. . In regard to the voyage of Nearchus it will be found that the distances are uniformly over-rated : in most cases very far exceed- ing the truth. To such an extent indeed is this the case that it induced Dr. Vincent to adopt the suggestion of M. D’Anville, and assume that Nearchus had throughout reckoned by a different stadium from that employed in estimating distances on land. This was indeed, in the time of D’Anville, the ordinary mode of solving such difficulties, and attempting to save the credit of ancient writers for accuracy. But even if we overlook the extreme im- probability that Arrian should have continued reckoning the dis- tances in this part of his work by stadia of fifteen to the Roman mile, while he elsewhere uniformly employs the ordinary stade of eight to a mile; or that there should have existed a nautical stadium, not much more than half the common itinerary stade, without any hint of such a difference being found in any of the NOTE B. VOYAGE or NEARCHUS. 545 ancient geographers—it will be found on examination that even this bold hypothesis is far from removing the difiiculty. Though Dr. Vincent considers himself compelled, by the wide divergence of the measures of Nearchus from the truth, to adopt the supposition that they are reckoned in these short stadia, he nevertheless admits that even with this resource “it is not possible to establish a [correct] proportion of part to part, or perhaps to measure five hundred stadia in any detached portion of the course with satisfaction” (p. 55). Moreover when he comes to the part of the voyage along the coast of the Ichthyophagi, where the fleet encountered the greatest hardships and difficulties, he finds that “ the stadium of D’Anville is less applicable to this coast,” which measures only 450 miles in a straight line, and 625 (according to his own calculation) by the course of the fleet (p. 2249). A pretty notable difference, which he explains principally by supposing that “ their distresses caused the distances to appear longer, at the same time that they engaged the mind too much to allow of accurate calcula- tion ” (p. 280). What means of accurate calculation they possessed in any case he has omitted to tell us. But when the fleet reached the coasts of Carmania and proceeded up the eastern shores of the Persian Gulf, where they were comparatively free from anxiety and encountered no serious difiiculties, it is found that the error is in the contrary direction, “ the measures upon this coast proving as erroneous from deficiency as those on the coast of the Ichthyophagi from excess” : a fault for which (adds Dr. Vincent) “ no better ex- cuse can be given than the situation of Nearchus in both instances. If distress magnified the length of his former measures, ease and security appear to have diminished those on'the coast of Kar- mania ” (p. 865). But such an explanation at once excludes the sup- position of any real measures or calculation whatsoever, and if we are compelled to fall back upon mere vague estimates (which is un- doubtedly the truth) it is surely much more simple to admit those estimates to be for a part of the voyage more widely erroneous than Dr. Vincent would allow, rather than have recourse to the forced assumption of a different mode of reckoning, in stadia of a kind of which we have no other mention. The mode in which M. D’Anville arrived at his conclusion is indeed sufificiently extraordinary, and is justly rejected by Dr. Vincent, though he adopted his result as applied to the voyage of Nearchus. In a passage of Aristotle (De Cwlo, ii. 14) already VOL. I. 2 N 546 HISTORY or ANGIENT GEOGRAPHY. can». xIII. adverted to, that philosopher estimates the circumference of the earth at 400,000 stadia, while it is well known that the same cir- cumference was" calculated by Eratosthenes at 252,000 stadia. Therefore, concludes the French geographer, the stadium of Aristotle must have been to that of Eratosthenes as 252 to 400 ; and it is reasonable to suppose that this was the stadium in use among the Macedonians, and consequently that employed by Nearchus. But in fact the attempt of Eratosthenes to measure a portion of a great circle on the earth’s surface was, so far as we know, the first real attempt made by any Greek to measure or calculate the circum- ference of the earth at all. All previous measurements, or rather estimates, including that of Aristotle, were merely conjectural: and the true inference from the passage in question is, not that Aristotle employed a different stadium, but that he supposed the world to be vastly larger than it really is. We find accordingly that his estimate of the earth’s circumference was universally re- jected by later Greek writers, after the more accurate investigations of Eratosthenes. Major Rennell also justly rejected the Aristotelian stade of M. D’Anville, of which no trace is found in any ancient author. Yet he himself fell into a somewhat similar error, and was induced by a comparison of numerous distances, as stated by ancient authors, to assume “an average of 718 to the degree,” for the Olympic or common stadium of Greece, giving 505 English feet to the stade, instead of 606 feet, which correspond with the 600 Greek feet universally recognized by the ancients as the length of the ordinary stadium. (Rennell’s Geography of Herodotus, p. 31.) NOTE 0, p. 528. PORT OF ALEXANDER. This suggestion was first made by Major Rennell Memoir of a Map of Hindostan, p. 187), and has been adopted both by Gossellin and C. Miiller. The importance of the port of Kurrachee is such as to leave little doubt of its correctness. Dr. Vincent, in whose days Kurrachee was little known, placed the Port of Alexander just beyond Cape Monze, a view that is open to the unanswerable NOTE D. VOYAGE or NEAROHUS. 547 objection that there is no port of the kind to be found there; and he has thus thrown his whole geography of this part of the coast into confusion. It must be admitted that on any view the topography of this portion of the voyage (from the mouth of the Indus to that of the Arabis) remains very obscure: it is difficult to understand how Nearchus should have rounded such a headland as Cape Monze without making any mention of it : and the harbour called Moron- tobara or the Port of Women, described by Arrian as large, roomy, deep, and well sheltered, though having but a narrow entrance, (MW/iv ,ae'yas Kai €6KUK)\09 Kai [3.16%]; Kai dK/\vo"ros, 5 8’ E'ovrhovs e’; airrov ora/6;, Indica, c. 22, § 5) has not yet been identified. This point is the more important as the name is still found in Ptolemy (vi. 21, § 2), and it must therefore probably have continued to be a place frequented by navigators. NOTE D, p. 532. THE ISLAND 0F ASHTOLA. Ashtola being the only island of any size along this coast, there can be little doubt that it is the same which is referred to by Nearchus in another passage (Arrian, Indica, c. 31) though he there says it was called by the natives Nosala. It was regarded by them with superstitious dread, believing that whoever landed there im- mediately disappeared. One of the transports was indeed actually lost, but Nearchus touched at the island with his own ship, and compelled his sailors to go ashore, without any ill effect. It is singular that a similar superstition should still linger about the island in modern times ; Captain Blair, who was one of the first to explore these coasts (in 17 74) having been warned by the natives at Passenee, that it would be dangerous to approach the island of Ashtola, as it was enchanted, and that a ship had been turned into a rock. (Vincent, vol. i. p. 299.) It abounds in turtle of large size and excellent quality; but no mention is made of these by Nearchus, nor does it appear ever to have suggested itself to the minds of the Greek voyagers to avail themselves of this important 2N2 548 HISTORY or ANGIENT GEOGRAPHY. can». x111. resource, though suffering so severely from want of provisions. It is remarkable indeed that no indication of the use of turtle as an ' article of food is found in either Greek or Roman writers. NOTE E, p.535. SUPPOSED ASTRONOMICAL CHANGES. This suggestion, which agrees in substance with the views of Gossellin and Schmieder, appears to me much more plausible than that proposed by Dr. C. Muller, that the statement in question was interpolated from Onesicritus, and did not really proceed from Nearchus at all. Apart from the improbability that Arrian, who has throughout followed the narrative of Nearchus with the utmost faithfulness, should, in one or two instances only, introduce state- ments from other sources without any hint that they were so derived, the particular fact in question is expressly cited as being stated by Nearchus (waparrhwdv'rwv 3% 'ri'p/ ’IV8(IW yijv . . . Ae'yet Ne'apxos {in ai o'Kt'at afiroio'tv of; 75.575 e’n-ot'eov, C. 25). It seems therefore impossible to doubt that it was found in his narrative. But two circumstances seem to me to concur in favour of the other hypothesis: the one, that this passage occurs at the point where Nearchus, according to his own view, had arrived at the western limit of India,-—-the Arabies and Oritae being regarded by him as Indian tribes: where he therefore seems to have paused, as we find him doing on other occasions, (see c. 29, 38, 40) to make some general remarks on the Indian coasts and people. On such an occasion he might easily have introduced the mention of so remarkable a phenomenon (to the Greeks of those days) which he had learnt from Indian navigators, who were doubtless even at that period in the habit of descending the western coast of the peninsula as far as Malabar. Even the WOI‘dS waparrltwdv'rwv 'rfiv ’Iv3v yr'jv may perhaps have in fact referred to such navigators, though, as they are intro- duced by Arrian, they undoubtedly can have reference only to Nearchus and his comrades. But another circumstance that appears to me unintelligible on the supposition that the statement was really made by Nearchus as a matter of his own experience, is, that, as the passage stands in Arrian, these appearances were only visible NOTE F. VOYAGE or‘ NEARCHUS. 549 when they stood far out to sea, towards the south (5mm ,uév e’n-Z n'oMov 'roi) wov'rov (he 'n'pos lueo-n/ifipt'nv n'poxwpfio'etav). NOW their voyage was throughout a coasting voyage in the strictest sense of the term ; they were never driven out to sea by violent gales; in fact they anchored off shore every night. It is certain therefore that they could never have been on any occasion more than a few miles to the south of the coast line; and of course no appreciable differ- ence in the appearance of the sun and stars could have been pro- duced by such a deviation. On the other hand, the words just cited would naturally find a place, if Nearchus were recounting the observations of others, and the general experience of navigators in the Indian seas. There seems indeed little doubt that such perverted statements as to the astronomical phenomena observed were really made by Onesicritus (cited by Pliny, vi. 23, § 98 ; see above, p. 543), and are quoted from him by other writers, but it is incredible that Nearchus, Whom we find throughout a careful and, trustworthy authority, should have been guilty of so flagrant a departure from the truth. NOTE F, p. 537. ISLAND : OF ORMUZ. The fate of Ormuz has indeed been a singular one. The island itself is not only, as it is termed by Nearchus, barren and rugged in a supreme degree, but wholly destitute of fresh water. Yet the advantages of its opportune situation at the entrance of the Persian Gulf, and of its secure harbour, rendered it in the middle ages under its Arabian rulers, and still more after it had fallen into the hands of the Portuguese, one of the greatest emporiums of trade and con- sequently one of the wealthiest cities of the East. Barbosa, who describes it as it existed before its conquest by the Portuguese in 1514, has left a curious account of the variety and extent of its commerce at this period, when it served as the chief entrepo’t whence the spices, gems and other productions of India and the silk of China, were carried to Alexandria, Cairo, and the chief cities of the West. (Barbosa’s Description of the Coasts of East Africa and Malabar, published by the Hakluyt Society in 1866, pp. 41-45.) At a later period “the wealth of Ormuz or of Ind” had become proverbial 550 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIII. for the riches of the East. (Milton, Paradise Lost, book ii. v. 2.) The island continued in the hands of the Portuguese till 1622, when it was wrested from them by the Persian monarch- Shah Abbas, who demolished the city and transferred the inhabitants, as well as what survived of the trade, to the town of Bunder Abbas or Gombroon, on the mainland, nearly opposite to Ormuz. The island itself is described as “ a mere barren rock, formed of rock salt and sulphur, and entirely destitute of vegetation. Its appearance is thus altogether the most desolate that can be imagined.” (Kemp- thorne in Geogr. Journal, vol. v. p. 274.) It is inhabited only by a few fishermen, but a small garrison is maintained there by the Imam of Muscat. (For a more recent account of Ormuz, as well as of the neighbouring island of Kishm, see a paper by Col. Pelly, in the Journal of Geogr. Soc. vol. xxxiv. pp. 251-258.) The Arabic name of the island, before it assumed that of Ormuz from the neigh- bouring mainland, was J erun or Djerun, in which some resem- blance may perhaps be traced to the ancient appellation, which is written Organa by Arrian, but appears under the form Ogyris in Strabo and Pliny. NoTE G, p. 539. DIRIDOTIS. The name of Diridotis is not found in any ancient author. But Strabo and other authors place a city called Teredon at the mouth of the Euphrates, which appears to have been a frequented port under the Macedonian kings. (Abydenus ap. Euseb. Chron. p. 21.) Hence the two have been generally identified. But this seems to me very doubtful‘. Teredon enjoyed the reputation of being an ancient city, founded by Nebuchadnezzar (Strabo, ii. p. 80, xvi. 3, p. 765 ; Dionys. Perieg. v. 980), while Diridotis is described as a mere village, though resorted to by the merchants of Arabia. It was probably therefore a mere temporary mart, at the actual en- trance of the Euphrates, while Teredon may be sought for a little higher up. If there was any foundation for the tale of its ancient origin, it could not have been situated really at the mouth of the river in the days of N earchus, on account of the rapid advance of ‘the ‘coast line. The suggestion of Col. Chesney (vol. ii. p. 355) NOTE G. VOYAGE or NEARCHUS. 551 that the site of Teredon is marked by a gigantic mound called J ebel Sanam, near the (supposed) Pallacopas branch of the Euphrates, and about 23 miles S.S.W. of Bassorah, is plausible enough. It is more interesting to observe that the manner in which Nearchus speaks of Diridotis as situated at the mouth of the Euphrates, which was navigable for 3300 stadia from thence to Babylon, appears to show conclusively that the Euphrates in his time still communi- cated with the sea by an independent channel of its own. CHAPTER XIV. SUccEssoEs or ALEXANDER. SECTION 1.—Megasthenes.——Increased knowledge of India. § 1. THE death of Alexander the Great, with the confusion which followed it, and the internecine struggles among his generals, gave a severe check to the progress of geographical discovery, as well as to the general advance of civilization in the East. It is true that, for a time, the death of the great conqueror did not seem likely to be followed by the dis- solution of his empire. While the rival candidates for power were contending among themselves for supremacy, the provinces, even the remotest of them, continued to acquiesce in Macedonian rule, and to be governed by Macedonian satraps. When at last a new order of things began to emerge out of the chaos, and it was evident that the great Macedonian empire would be permanently divided into several separate and independent monarchies, still the whole, or almost the whole, of the Asiatic conquests of Alexander remained united under one head. From the time that Seleucus N icator esta- blished himself at Babylon in B.0. 312, he became virtually the lord and master of all the vast regions of Upper Asia, to which, after the fall of Antigonus (in B.0. 301) were added also Syria and great part of Lower Asia, or what we now call Asia Minor. Ptolemy meanwhile had not only fortified himself in the undisputed possession of Egypt, but held also the im- portant island of Cyprus, with the maritime cities of Phoenicia, still the seat of an extensive and flourishing commerce. Of the steps by which Seleucus established and secured his authority over the extensive satrapies from the Euphrates to the Indus we have unfortunately no information; the meagre SE01‘. 1. SUGCESSORS OF ALEXANDER. 553 historians of this period confining their attention almost ex- clusively to the contests between the rival competitors for power. But there appears no doubt that he was called upon to make a series of successive campaigns into different provinces, the governors of which had probably taken advantage of the con- fusion at head-quarters to establish themselves in virtual independence, trusting for safety to the remoteness of their situation, and the difficulty of approaching them. The details of these operations, had they been preserved to us, would doubtless have thrown much light upon the geography of Upper Asia,'but as it is, they are a mere blank. All that we learn is that by degrees Seleucus consolidated his dominion over the whole of the eastern provinces, which had been comprised in the empire of Alexander, including the remote regions of Bactriana, Sogdiana, and the Paropamisus ; and that he transmitted this extensive empire undiminished to his son Antiochus. That prince had indeed been already established during the lifetime of his father (B.0. 293) in the government of all the provinces east of the Euphrates, with the title of king. The statement that no less than seventy-two satrapies were subject to the authority of Seleucus,1 if it be indeed founded on fact, would prove that he must have instituted a new administrative partition of his dominions, subdividing the previously existing provinces into much smaller governments; a measure that would indeed be dictated by sound policy, as tending to diminish the excess of power confided to former satraps.2 But of such a system we find no other indication. § 2. On one point only do we obtain any further informa- tion concerning the proceedings of Seleucu's in the East, and although the statements preserved to us are even here imperfect and unsatisfactory, they possess nevertheless a great interest. It is certain that Seleucus not only renewed the relations with 1 Appian, Syriaca, 0- 62- with the view indicated in the text. 2 D1‘0V8911 (Hellenismus, V01. i. p. > But the authority of Appian on such a 606) assumes it as certain that such a l point is unfortunately worth very little. measure was adopted by Sclcucus, and ' 5 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIV. India which Alexander had begun, but materially extended them, and acquired important additional information concern- ing parts of that country to which the Greek arms never penetrated. So far as we are able to discern, the Greek satraps, who had been appointed by Alexander to the command of the pro- vinces adjacent to India, maintained the Macedonian power in the regions west of the Indus; but having taken part with Antigonus in the contest for Upper Asia, they naturally found themselves on hostile terms with Seleucus, who was consequently obliged successively to reduce them under his authority. This he appears to have successfully accomplished; and there is no doubt that he had at one period reunited under his government all the provinces that lay west of the Indus. Meantime an important revolution had taken place in the regions east of that river. A native chief, whom the Greeks name Sandracottus—the original Indian form of the name is Chandra-gupta—had overthrown in succession the petty native dynasties that were reigning in the Punjab and had thus become the founder of a powerful monarchy, which he had subsequently extended by his arms eastward, from the banks of the Sutledge to the J umna and the Ganges, where he overthrew the powerful monarchy of the Prasians, and thus established his authority over the whole of the north of India.3 As soon therefore as Seleucus had recovered possession of the countries west of the Indus, he found himself in the presence of a powerful and warlike monarch, commanding resources of .a very different character from any that had been opposed to Alexander.4 In such a state of things the occurrence of hostilities between the two was almost inevitable, but of the 3 Justin, xv. 4. For the history of pp. 356-361. ' Sandracottus and the revolution effected 4 Of course the statements that San- by him in the kingdom on the banks of dracottus was at the head of an army the Ganges, founded nearly a century of 400,000, or, according to others, before by the Magadha dynasty, see 600,000 men, are to be taken merely as Lassen’s Indische Alterthiimer, vol. ii.; instances of Oriental exaggeration Duncker, Gesch. des Alterthu-ms, vol. ii. (Strabo, xv. p. 709; Plut. Alex. 62). SE01‘. 1. SUCGESSORS OF ALEXANDER. SSS circumstances and character of these we are almost wholly ignorant. It appears certain that Seleucus crossed the Indus and invaded the dominions of the Indian king, but it is probable that he did not advance far ;5 at all events it is clear that the final result of his operations was not successful, as by the treaty which he ultimately concluded with Sandracottus, he not only yielded all claims to the Punjab, but ceded a portion of the districts of the Paropamisus and Arachosia, which had been previously held by the Macedonians, so that both. banks of the Indus were now united under the authority of the Indian king.6 Seleucus however received in exchange not less than five hundred elephants—an addition to his forces which he may well have regarded as an ample equivalent for the precarious possession of a few remote and barbarous dis- tricts. He hastened to return to Babylon to take part in the coalition against Antigonus (B.O. 302). Throughout the remainder of his reign Sandracottus con- tinued to maintain the most friendly relations both with Seleucus and his son Antiochus. He sent repeated embassies to Babylon, while the Syrian king in return sent a Greek of the name of Megasthenes, who appears to have enjoyed a high place in his confidence and favour, as envoy to the court of Sandracottus. Megasthenes repaired to the capital of the Indian king, at Palibothra on the Ganges, where he resided for some time, and brought away an amount of information concerning India in general, which became for a long time 5 The supposition that Seleucus himself had penetrated as far as the Ganges, though adopted by several early writers, some of whom even sup- posed him to have advanced to Pali- bothra, is wholly untenable; and is deservedly rejected by Robertson and others. Pliny’s expression, “Reliqua inde Selenco Nicatori peragrata sunt,” would certainly seem to imply that the king had really advanced as far as the Jumna, but even this is at least very doubtful, and very little reliance can be placed upon the precise phraseology of such a writer as Pliny (compare his expressions concerning Patroclcs). The subject is well discussed by Schwanbeck in his edition of the frag- ments of Megasthenes, pp. 12—18. 6 Strabo, xv. 2, § 9, p. 72%; Appian, Sym'aca, c. 55. The extent of the ter- ritory ceded is not clearly indicated. It probably comprised only the districts between the west bank of the Indus and the mountains. 556 HISTORY or ANCIENT GEOGRAPHY. crap. XIV. the foundation and principal authority for all that the Greeks knew in regard to that country. § 3. The work of MEGASTHENES has unfortunately perished; but large extracts from it have been preserved to us, especially by Arrian and Strabo, both of whom have taken it as the base of their descriptions of India, in regard as well to its natural features, as to its political institutions and to the manners and customs of its inhabitants.7 A considerable amount of informa- tion had indeed been already collected by the companions and generals of Alexander, and given to the world by Nearchus, Onesicritus and others, and it is not always possible to separate distinctly the statements derived from these different sources. In taking a brief review of the real extent and value of the knowledge of India acquired at this time by the Greeks, it is of little consequence to observe this distinction; but in general it may be assumed that the statements of Strabo and Arrian, for which no other authority is expressly cited, are derived from Megasthenes. § 4. Already before Alexander had reached the banks of the Hyphasis he had received information of the existence of a great and powerful empire established on the banks of the Ganges, the resources of which were doubtless greatly exaggerated by the native reports that were transmitted to him. According to these statements, Xandrames, the king of the Prasians, who ruled over the whole of the territories east of the Jumna, commanded an army of not less than 200,000 infantry, 20,000 cavalry, 3000 elephants, and more than 2000 war-chariots.8 But with every allowance for exaggeration, there seems no doubt that even at this. period there really existed a powerful monarchy, com- bining under its sway the whole of the fertile regions from the Jumna to the mouths of the Ganges: while the overthrow of the then existing dynasty by Sandracottus seems to have im- 7 The fragments of Megasthenes have are contained also in the 2nd volume been collected, and published, with an of the Fragmenta Historicoram Graz- excellent introduction, containing a corum, edited by C. Mtiller. full inquiry into his life and writings, 8 Diodor. xvii. 93. Compare Plut. by Schwanbeck, 8vo, Bonn, 1846. They Alex. 62, whose numbers are still higher. SEGT. l. MEGASTHENES. 557 parted fresh life and vigour to the whole empire, at the same time that he so materially extended its confines towards the west. The seat of government had been already established at the city which the Greeks called Palibothra, a form under which we readily recognize the native appellation of Patali- putra; it was situated at the junction of the Ganges with the Sone, a few miles above the modern city of Patna.9 §5. On his journey to the capital Megasthenes followed what was termed “ the royal road,” and it is to him that we are indebted for the only definite geographical data that have been preserved to us by ancient writers concerning this part of India. Unfortunately his statements of distances have been transmitted to us only by Pliny,1 and the numbers as they stand cannot be regarded as trustworthy. But the general line of route may be clearly followed. The principal points or stages are thus given :——“ From the Hyphasis to the Hesydrus (Sutledge), 168 Roman miles; thence to the Iomanes (Jumna), the same distance; thence to the Ganges, 112 miles; to the Ithodapha, 119 miles (though others made this distance 325) ; to the city of Galinipaxa, 167 miles (or according to others 265); thence to the confluence of the Iomanes and Ganges, 620 miles; to the city of Palibothra, 425; to the mouth of the Ganges, 638 miles.”2 Setting aside the distances, which are involved in hopeless confusion,3 we find the most important points on this line of route correctly given ; it proceeded across from the Sutledge to the Jumna, thence to the Ganges, and afterwards descended 9 The true site of Palibothra was first pointed out by Major Itennell (Memoir of a Map of India, 1). 50). His view has been generally adopted by recent writers. 1 Eratosthenes appears to have availed himself of the same authority, when he reckoned 10,000 stadia from the Indus to Palibothra (ap. Strab. xv. 1, §11). The startling discrepancy between this statement, and the num- bers given by Pliny is an additional proof of the utter unworthiness of the latter. 2 Plin. Hist. Nat. vi. 17, § 63. 3 Pliny himself tells us that the numbers were variously given by his authorities, and the instances in which he adds the different statements show how wide was the divergence between them. When we add to this cause the diversity of numbers found in our existing MSS. of Pliny, we may well pronounce any attempt to reconcile or explain them to be wholly futile. 553 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIV. the valley of that river as far as the junction with the J umna at Allahabad. The site of Calinipaxa, which, from the way in which it is mentioned, would seem to have been a place of im- portance, is wholly unknown: nor can we identify the river Rhodapha among the numerous tributaries of the Ganges. Megasthenes, indeed appears himself to have enumerated nine- teen tributaries of the great stream, of which the names have been preserved to us,4 but in no correct geographical sequence, and without any statement of distances: hence their identi- fication is very uncertain.5 Among the names that can be most clearly recognized is that of the Sonus, mentioned both by Arrian and Pliny, and still called the Sone, which falls into the Indus at Palibothra. §6. Megasthenes was probably the first Greek who ever reached the banks of the Ganges—certainly the first writer who transmitted any account of it from personal observation. That his account should be tinged with exaggeration is not to be wondered at. We have seen how greatly they magnified in their reports the real dimensions of the Indus and its tributary waters: and the sacred stream of the Ganges was still more calculated to impress the imaginations of those who gazed upon it for the first time. We are told that it was 100 stadia in width at its narrowest point, and that where its waters spread out freely, one bank could not be discerned from the other.6 Of its sources he had of course no definite knowledge, but was 4 They are given by Arrian (Indica, c. 4), but in a very careless and per- functory manner. Pliny also states the whole number at nineteen, but gives the names of only four, “ besides those ah'cally mentioned,” i. e., the Iomanes and Rhodapha, and apparently regards these as the only ones that were navigable (H. N. vi. 18, § 65), though this is directly at variance with the statement of Arrian. The emission in the former list (that of Arrian) of the Iomanes or Jumna, the most important of all tributaries of the Ganges, and which was unqucs“ tionahly known to Megasthenes, as it is twice mentioned in the description of the “royal road,” is unaccountable. But it shows how little reliance can be placed on such enumerations, when reported at second hand. 5 It has been attempted to determine- them by means of the Sanscrit names, but it must be confessed with very im- perfect success. 6 Arrian, Indica, 4, § 7. The state- ments quoted by Pliny are somewhat more moderate, making it only eight Roman miles at the narrowest, and 100 stadia in average width (Plin. H. N. vi. 18, § 65). SECT. I. MEG ASTHENES. 559 correctly informed that it took its rise in the Indian Caucasus (the Himalaya), and after flowing at first to the south, then turned eastwards, and pursued an easterly course as far as Palibothra.7 Of the great bend that it makes towards the south, after passing that city, he had no knowledge, but described it as holding on the same course to its mouth in the Eastern Ocean. It is remarkable that he particularly stated that it had but one month, as distinguished from the Indus, which formed a Delta.8 So imperfect was his information concerning the course of the river below Palibothra, that he seems to have had no idea of the complicated network of rivers really formed by the Ganges and its numerous arms, before they enter the Gulf of Bengal. Palibothra itself was described as a very large city, situated at the confluence of the Ganges with another river (the Sonus or Erannoboas),9 and built in the form of a parallelogram, eighty stadia in length, and fifteen in breadth: it was sur- rounded only with a wooden wall or stockade, but pierced with loopholes, and flanked by numerous towers, besides an outer ditch of vast dimensions, which was filled with water from the neighbouring rivers.1 But" no details are preserved to us of architectural splendour, or of that magnificence in the deco- rations of the court and palace which we are accustomed to associate with the capitals of Oriental potentates. §7. The other cities of India, we are assured, were so numerous that they could not be counted; but Megasthenes 7 Strabo, xv. 1, § 13, p. 690. 8 Strabo, l. e. 9 Some confusion has arisen in con- sequence of the statement that the river at the confluence of which with the Ganges Palibothra was situated was the Erannoboas (Arrian, In-dz'ca, c. 4. The name is accidentally omitted by Strabo, having evidently fallen out of our text. See Kramer’s note); while the ruins generally regarded as those of Pataliputra are placed near the junction of the Sone, a large river, the name of which, as already observed, is readily recognised in the Sonus of Arrian and Pliny. But although these authors certainly regarded the two rivers as distinct, there can be little doubt that they were in fact only two names for the same stream, the name Erannoboas being only a Greek form of the Sanscrit Hyranyavahas (the golden armed), which is the poetical desig- nation of the Sone (see Bitter, Erd- hunde, vol. iv. pt. 1, p. 508). 1 Strabo, xv. 1, § 36 ; Arrian, Indica, c. 10. According to Diodorus (ii. 39), Palibothra was founded and fortified by Hercules. (l) 56o HISTORY or ANCIENT GEOGRAPHY. CHAP. XIV. estimated—of course from mere hearsay authority —-'the, num- ber of nations or tribes that inhabited it at not less than a hundred and eighteen.2 Of these, the Prasians, in whose country Palibothra ‘was situated, were accounted the chief. Their relation with the Gangaridae,3 whose name is also fre- quently mentioned by Greek authors, is not clearly indicated ; but the latter appellation seems to have been applied in a general way to all the inhabitants of the valley of the Ganges, while that of Prasians is evidently only the Greek form of the Sanscrit Pratschia—the people of the East—and was thus not a true ethnic appellation. It had doubtless been used to desig- nate the kingdom founded on the banks of the Ganges, in contradistinction to those of the Punjab. All the information collected by Megasthenes points to the conclusion that at this period the Indian monarchy was a powerful and well-organized government: the king was sur- rounded by a number of counsellors, who assisted him in maintaining a strict superintendence over all branches of the administration: justice was well administered: care bestowed upon the maintenance of the highways: and the revenues diligently collected. Sandracottus is said to have kept up a standing army of 400,000 men, with 30,000 cavalry and 9000 elephants.‘ These numbers are doubtless exaggerated, but there seems no reason to doubt the fact that his forces were really very numerous and well organized. The king himself, except when he went out to war or the chase, or for the performance of some special sacrifices, was hardly ever seen. Within his palace he was attended only by women, some of whom even accompanied him on his hunting expedi- tions and in war, and were especially charged to watch over his personal safety. 2 Arrian (Indica, c. 7) justly remarks ‘ on the inadequate means of information possessed by Megasthenes for such a statement. Herodotus, who dwelt west of the Indus 4 Strabo, xv. 1, gi 53, p. 709. Pliny, as usual, adopts the still more exag- 3 The name is written by Plutarch (Alex. 02) Gandaridze; but they had nothing to do with the Gandarii of geratcd statement of 600,000 as the number of the infantry (H. N. vi. 19, §os). Sizer. 1. MEGAS'l‘I-IENES. 56! § 8. Of the peculiar institutions of the I'Iindoos, their divi- sion into castes could not fail to attract the attention of the Greeks, and the statements of Megasthenes upon this subject, though not free from error, are in themselves very curious and interesting. According to his account the Indians were in his time divided into seven classes, every one of which kept itself quite distinct from the others, and could neither intermarry nor pass from one class to another. The first of these was formed by the wise men or philosophers, by whose advice and direction everything was managed: the second consisted of the agriculturists or tillers of the soil: the third, of shepherds and hunters: the fourth, of artisans and traders: the fifth, of soldiers: the sixth, of inspectors, or official superintendents appointed to examine every detail of the lives of others: the seventh, of other officials and councillors to whom the ad- ministration of public affairs was entrusted.4 Megasthen-es was undoubtedly led into error by supposing these two last classes, which really formed very distinct bodies of men, to constitute separate castes. He was also misled in describing the shep- herds and hunters as similarly separated from all others: but these errors are easily accounted for, from the great tendency which has in all ages prevailed in India to render all such occupations hereditary, and thus maintain subordinate lines of distinction within the broader limits of the castes properly so called. These, as is now well known, were really only four in number, namely; the Brahmins, answering to the philosophers of the Greeks, but comprising also the oflicial inspectors and ministers, whom they regarded as constituting the sixth and seventh classes; the Kshatriyas or warriors: the Vaisyas, including herdsmen and hunters as well as agriculturists; and the Sudras, or menial class, comprising also artisans and handi- craftsmen of all descriptions. Besides these however there exist what are, called the mixed castes, several of which are, 4 Arrian, Indica, c. 11, 12; Strabo, are taken from Megasthenes, and are xv. 1, §§ 39-41, 46-49; Diodor. ii. 40, almost exactly the same. 41. The accounts of all three authors VOL. I. [O Q 562 HISTORY or ANGIENT GEOGRAPHY. CHAP. XIV. according to law at least, confined to the practice of certain trades or occupations.5 One circumstance which naturally attracted the particular attention of the Greeks was the absence of slaves: slavery as an institution, in the form in which it was universal in the West, being unknown among the Indians: and though the condition of the Sudras was in reality little removed from that of a servile class, this appears to have been so far ameliorated in practice as to escape the observation of foreign visitors. Nor do we find any allusion to the still more degraded class, called Pariahs, which new forms an important element in the social constitution of India. In other respects too the Greeks were led to form too favourable an estimate of the state of society among the Indians, as well'as of their moral character. But this tendency to find a kind of Utopian perfection in_ any form of society widely different from that with which the observer is familiar, is an error of frequent occurrence in all ages.6 Thus Megasthenes represented the warrior caste as leading a life of perfect ease and enjoyment, when not called upon to go out to war, while the agricultural peasantry pur- sued their occupations in undisturbed tranquillity, being never interfered with, even when their district became the theatre of hostilities. §9. Much attention was bestowed, both by Megasthenes and other Greek writers, upon the Brahmins, or philosophers as they termed them, whose doctrines and tenets were regarded by them as having much resemblance with those of Pytha- goras. For us on the contrary it is interesting to compare the notices that have been thus preserved to us, with the full knowledge that We now possess of the philosophical and reli- gious systems of the Hindoos; and it must be admitted that considering the difficulties under which the Greeks must have 5 See Colebrooke’s Enumeration of Voltaire (Essai sur les Zlfaeurs et l’Esprit Indian Classes in his Essays, pp. 270- ! des Nations) is an instance that will 279. , readily present itself to the mind of the 6 The flattering pictiu'e of China by 1 modern reader. SECT. l. MEGASTHENES. s63 laboured in obtaining such knowledge, their information is singularly correct.7 They mention also that besides the Brach- manze or Brahmins, there existed another sect of philosophers whom they term Garmanae or Sarmanze, who led a life of hard- ship and privation in the woods, but enjoyed a great reputation for sanctity. It is probable that we have here a notice of the Buddhists, who were at this time rising to eminence, though still inferior in importance and consideration to the Brahmins, as they are justly described by the Greek writers. Sarmanzc is probably a corruption of Sramana, the native appellation of the Buddhist priesthood.8 Others again prac- tised austerities similar to those of the modern Fakeers, remaining immoveable in the same position, standing or lying naked in the full heat of the sun, &c. The practice of self- immolation by voluntary burning was also regarded by the Greeks as one characteristic of the Brahmins; a belief na- turally confirmed by the conspicuous instance of Calanus, who having accompanied Alexander on his return from India, put an end to his life in this manner.9 On the other hand the custom of widows burning themselves on the funeral pile of their husbands—so general among the Hindoos in' modern times—is noticed by the historians of Alexander only as one of the peculiar customs of a particular tribe—the Cathzeans— in the northern part of the Punjab.1 §10. Of the natural productions of India the accounts given by Megasthenes were in general very accurate. Fore- The name _ 7 It is mentioned in one passage that the communications of the Brah- mins with the messengers sent to them by Alexander had to pass through three successive interpreters, all of them illiterate and ignorant men (Strabo, xv. 1, § 64, p. 716). 8 The name is written I‘apuaves and Fepudz/es in our MSS. and editions of Strabo; but Clement of Alexandria, whose account is obviously derived from the same source, writes it Zap/ritual, which is doubtless the more correct form, for the reason given in the text (Clem. Alexandr. Stromata, i. 15, §71). 9 He is expressly described as having done this in accordance with the cus- tom of the country (Ital d1r06aue'iy 'TQ 1ra'rpicp vii/1.9:, 'refle'z/Ta e’1rl 1ruprca'idv. Strab. xv. 1, §64). Mega-sthenes, how- ever, denied that self-immolation was inculcated by the precepts of the philo- sophers. Ibid. §68. 1 Aristobulus, ap. Strab. xv. 1, § 62, p. 714; Diodor. xvii. 91. 202 564 HISTORY or ANCIENT GEOGRAPHY. CHAP. XIV. most among these may undoubtedly be reckoned the ele- phants, with which the Greeks had for the first time become acquainted during the expedition of Alexander. But not only has be preserved to us many curious particulars concerning their natural history, and some interesting anecdotes of their sagacity, but has given a full account of the mode of catching and taming them, which agrees in almost every particular with those recorded by modern travellers.2 The size and ferocity of the tigers in the land of the Prasians is also men- tioned with wonder; and the gigantic serpents—the Pythons of modern naturalists—were perhaps still more calculated to excite astonishment. Nor can we wonder if the accounts of these monstrous reptiles were considerably tinged with ex- aggeration.3 Even the parrots and monkeys were strange to the eyes of the Greeks, and bore a prominent part in their narratives. The abundance of peacocks too had been early remarked by the companions of Alexander.4 They could not fail also to be struck with the vast forests of trees wholly different from any they knew elsewhere; and the peculiar character and mode of growth of the Banian or Indian fig- tree is described with great accuracy. In this instance indeed even their dimensions are not exaggerated. The vast size of the reeds (bamboos) also attracted particular attention, and the circumstance of trees growing actually in the sea (evidently mangroves) is noticed with wonder.5 Among the quadrupeds characteristic of India we find mention also of the rhinoceros, the name of which first became K statements for which Onesicritus is justly called by Strabo r6311 1rapa56§wv tipXucvBepz/rj'rnv. Aristobulus and N6- archus, as usual, were much more moderate. The latter mentioned a 2 Megasthenes, Fr. 36, 37, ed. Schwanbeck. Arrian,,Indica, c. 13, 14; Strabo, xv. 1, § 42. 3 It was Onesicritus alone, who was notorious for the gross exaggerations in which he was wont to indulge, that spoke of two serpents kept in confine- ment by Abisares, king of a mountain tribe above Taxila, of which one was 80 cubits, the other 140 cubits long! (Onesicrit. ap. Strab. xv. 1, § 28, p. 698.) This is one of the marvellous serpent of 16 cubits in length as the largest that was actually caught by the Macedonians, but added that the In- dians asserted they were found of much larger size (Arrian, Indica, c. 15, § 10). 4 Q. Curtius, ix. 2, § 13. 5 Megasth. Fr. 19. SECT. 1. MEGASTHENES. 565 known to the Greeks at this period.6 \Ve have already seen how much they were struck with finding crocodiles in the Indus, though subsequent observation showed them that they were neither so numerous nor so formidable as those of the Nile. The resemblance of the two rivers in this respect seems to have led them to expect to find hippopotami also in the Indus; as they thought it worth while to notice especially that these were not found.7 The great size and power of the Indian dogs—some of which were said to be a match for a lion, and to despise any meaner foe—are especially mentioned as having attracted the admira- tion of Alexander.8 The most remarkable specimens were those exhibited by Sopeithes,'the king of a tract at the foot of the Himalaya, near the banks of the Hyphasis, and the dogs of these mountain regions are to this day described as the largest and most powerful in India. Of the more useful productions of the country the sugar- cane is noticed,9 though in a manner that would hardly lead us to suppose it was in very extensive use; rice is well described, as well as the mode of its cultivation; and cotton, or tree-wool, as it was called, the general use of which by the natives in the place of linen and woollen textures is attested by all Greek writers. They had even noticed the fact of its stones or hard seeds, which had to be extracted before the cotton could be carded.1 Silk also is mentioned, though obscurely, and was supposed to be procured from the bark of a tree.2 Cinnamon and other spices were said to be produced in the south of India, from whence doubtless they were transmitted in the way of trade to the regions on the banks of the Ganges.3 6 Q. Curt. ix. 1, §5. 7 Onesicritus filuDO, with his usual inaccuracy, maintained that there were hippopotami in the Indus (Strabo, xv. 1, p. 690; Arrian, Indica, c. 6). _ 8 Strabo, xv. 1, §§ 31, 37; Diodor. xvii. 92; ZElian, Hist. Nat. viii. 1. The fame of these Indian dogs had indeed previously reached the ears of the Greeks, as they are mentioned both by Xenophon (Cynegetica, c. 10) and by Aristotle (Hist. An. viii. 28). They had probably been already imported into Persia. ‘J Strabo, xv. 1, § 20. 1 Aristobulus, ap. Strab. xv. 1, § 21, p. 694. 2 Strab. l. c. 3 Id. § 22, p. 695. 566 HISTORY or ANcIENT GEOGRAPHY. carp. XIV. § 11. It is strange that a writer who has preserved to us so many valuable and trustworthy notices should have allowed himself to be led astray so as to relate the most absurd and extravagant fables. But not only did Megasthenes repeat the story of the gold-digging ants, as large as foxes—a fable for which there was certainly some foundation, however strangely perverted,4 and assert the existence of unicorns, which he describes as horses, with heads like deer, and a single horn --a belief that still prevails in some parts of India; but he described winged serpents, winged scorpions, and other creatures equally anomalous. Still more absurd were the fables that he repeated concerning races of men of pigmy stature—some only five spans in height, some only three- of others without nostrils, others without mouths, supporting themselves only by smells; of wild men who had their feet turned backwards, so that the heels were in front; of men with ears like dogs, and only a single eye in the middle of their forehead; and others again with ears reaching to their feet, on which they used to lie down to sleep.5 Some of these marvellous tales were certainly Indian fables, which had been really told to the travellers,6 but others are unquestionably fictions already familiar to the Greeks, which were transplanted at pleasure from one unknown region to another. § 12. The direct information obtained by Megasthenes was confined to the two great valleys, or plains, of the Indus and the Ganges; of the mountain ranges and table-lands of the interior he seems to have had no knowledge. Probably the 4 See this subject discussed in Chapter VII. p. 229, Note D. Megas- thenes added the curious statement that these gold-digging ants were found among the Derdee, a nation Daradae or Daradrae of Ptolemy, whom that author places near the sources of the Indus (Ptol. vii. 1, § 42), and it is still retained by the Derds or Dards, who dwell in the mountains on the dwelling in the mountains in the east of India, where there was a great table- land (6p01ré5wv) in which the gold was found (Fr. 39, ap. Strab. xv. i. p. ‘706). This points clearly to the gold being derived from the valleys of the Hima- laya. The name of the Derdee is obviously the same with that of the frontiers of Thibet, near the upper valley of the Indus. 5 Megasthen. Fr. 29, ed. Schwanbeck. Strabo. xv. 1, § 57, p.711. Compare Pliny, Hist. Nat. vii. 2, §§ 25, 26. 6 See on this subject the remarks of Schwanbcck in his introduction to the ‘ fragments of Megasthenes, pp. 61-71. SECT. 1. MEGASTHENES. 567 dominions of Sandracottus were limited towards the south by the Vindhya mountains. But of the general configuration and extent of India, as a whole, he obtained a more accurate idea than under the circumstances could well have been expected. Thus while Ctesias had asserted that India was equal in extent to the whole of the rest of Asia, and even Onesicritus had maintained that it was a third part of the inhabited world, Megasthenes greatly reduced its dimensions, stating that its extent from the mountains on the north to the Ocean on the south was about 20,000 stadia (2000 G. miles). He added the important observation that its greatest length was from north to south; its breadth from east to west—~which almost all other writers had regarded as the direction of its greatest extent—being really only about 16,000 stadia.7 Both statements are indeed considerably in excess of the truth—the real length of Hindustan being in round numbers about 1800 English miles, and its greatest breadth about 1500 miles ; but considering the nature of the information to which alone Megasthenes could have access, and the vague ideas previously entertained by the Greeks of the geography of these countries, it must be admitted that his numbers present a marked approximation to the real distances. §I13. Another point upon which Megasthenes appears to have been the first to obtain any definite information was concerning the large and important island of Taprobane, or Ceylon. Its name indeed had already reached the ears of the companions of Alexander; and was mentioned by Onesicritus, who stated that it contained large numbers of elephants, of larger size than those of India, and added vaguely that it was 5000 stadia in extent, without specifying whether he meant length or breadth 8——and was distant twenty days’ voyage from the main land. Megasthenes reduced this last statement to 7 Strabo, ii. 1, §§ 4, 7,12; xv. i. §12; ment of singular preciseness, consider- Arrian, Indica, 3, §§ 7, 8. According ing the nature of his information. to the latter writer, Megasthenes gave 8 Strabo, xv. 1, § 15, p. (391. the length at 22,300 stadia; a state- 568 HISTORY OF- ANCIENT GEOGRAPHY. can». XIV_ Seven days—still a great exaggeration—and added that the island produced abundance of gold and pearls.9 His informa- tion was doubtless derived from merchants who traded thither for this last article. But his geographical idea of the position of the island was evidently very vague. § 14. Of the subsequent intercourse of the Greek monarchs ‘with India we have very little information. We learn indeed that after the death of Sandracottus, a Greek of the name of DAiMAcHUs was sent by Seleucus as ambassador to his suc- cessor Allitrochades (Amitraghades) and that he wrote some account of his journey, but his authority is rarely cited, and he is treated by Strabo as worthy of very little credit.1 Much more trustworthy, according to that author, was Patrocles, an officer who held important commands both under Seleucus N icator and his son Antiochus, and was entrusted for some time with the chief government of the provinces on the frontiers of India. But it is not clear that he himself visited that country, or at least that he ever crossed the Indus. Nor have we any account of the title or character of the work in which he had collected the results of his researches; but his geo- graphical data were considered by Eratosthenes as more trust- worthy than those of Megasthenes, and were made by that great geographer the foundation of his own description of India.2 Still less do we know concerning a writer of the name of Dionysius, mentioned only by Pliny, according to whom he was sent ambassador to one of the Indian kings by Ptolemy Philadelphusf” But such a mission wasjnatural enough, with a view to counteract the influence of the Syrian kings, at the time that the Egyptian monarchs were striving to open a direct trade with India. 9 Megasthen. Fr. 18; Plin. H. N. vi. 2 Strabo, ib. pp. 68, 69, 70. 22, § 81. 3 Plin. Hist. Nat. vi. 17, § 58. 1 Strabo, ii. 1, §9, p. 70. SEOT. 2. SUCCESSORS OF ALEXANDER. 569 SECTION 2.—Bactrian Prouincea—Patrocles. § 1. From India it seems natural to turn to another quarter, where we might have reasonably hoped to find that the long continued Macedonian dominion would have led to increased geographical knowledge. But in this case we are destined to complete disappointment. We have seen how Alexander had carried his arms as far as the Iaxartes, and had not only subdued the remote provinces of Bactria and Sogdiana, but had founded several cities there, and established colonies of Macedonian soldiers to form a nucleus of Hellenic civilization. Nor did these settlements prove abortive. So far as our information enables us to judge, the cities thus founded main- tained themselves for a considerable period, while the pro- vinces in question, after continuing for more than sixty years under the government of Greek satraps, appointed by the Syrian kings, threw off the yoke and declared themselves independent of the Seleucidan monarchy.‘ Their rulers how- ever were still of Greek origin, and there can be no doubt that they continued to retain a strong tincture of Greek culture, notwithstanding their remote and secluded position. But the disadvantages resulting from this cause were greatly increased, when the Parthians—a purely oriental tribe, probably be- longing to the same race with the Turks—who had revolted from the Syrian kings much about the same time as the Bac- trians, extended their dominion over the whole table-land of Iran, and ultimately established their authority even in the valley of the Tigris. 4 The date of the first declaration of independence by the Bactrian satrap, Theodotus or Diodotus, cannot be vol. iii. p. 18.) Hence the revolt of Baetriana has been fixed by Bayer in 255 B.0., and by Visconti and others. fixed with certainty, but it may be determined within approximate limits. We learn positively from Strabo (xi. 9, §2) that it preceded the revolt of the Parthians under Arsaces, and this is assigned by the best authorities to the year 250 B.0. (Clinton, Fast. Helten. including Professor Wilson, in 256 (Wilson’s Ariana, p. 216). General Cunningham would assign it to the year 216 B.0., but the arguments in favour of the earlier date appear to me to preponderate. 570 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XIV. . § 2. The effect of this was completely to isolate the Greek kings of Bactria, and the regions over which they had esta- blished their rule—including Ariana and the Paropamisus, as well as the provinces of Bactria and Sogdiana—from the rest of the Greek world : and though it is certain that the germs of Greek civilization, once planted in these remote regions, main- tained their ground with singular pertinacity for a long period, we are almost wholly without information concerning their progress and diffusion. The Bactrian provinces, north of the Paropamisus, were the first to succumb under an invasion of Scythian tribes from the north-east;5 while the Greek rulers who had established themselves at the foot of the Paropa- misus, and on the borders of Ariana and India, maintained their independent position down to a later period, and were able at times to extend their arms over the Punjab, and even the lower part of the valley of the Indus. But almost all definite historical information is wanting in regard to these remote and petty kingdoms: we know nothing of their poli- tical institutions or relations with the native powers, while in a geographical point of view the result is absolutely null. \Ve find indeed a vague statement that these monarchs had carried their arms farther into India than Alexander had done, 5 The period of this, like almost all the dates and facts connected with these Bactrian kingdoms, is uncertain: but the final extinction of the Greek dynasties in this part of Asia is assigned to the year 126 B.0.: and it appears probable that the provinces north of the Hindoo Koosh fell into the hands of the barbarians some time before these on the western border of India. Recent researches in connection with the coins which have been discovered of late years in large numbers in this part of Asia, have established the ex- istence of not less than twenty—nine kings or princes bearing pure Greek names, and consequently in all proba- bility of purely Greek extraction. So great a number of rulers during a period of only 120 to 130 years appears clearly to show that some of them were con- temporaneous, and reigned separately in Bactria proper, Ariana, or the lndian provinces. But all attempts to arrange them in any definite chronological order, or determine their geographical relations, are based almost entirely upon mere conjecture. See Wilson’s Ariana, pp. 215-30!) ; and an elaborate series of papers by Major-General Cunningham in the Numismatic Chro- nicle, vols. 8, 9, 10. The whole subject has been still more recently submitted to a careful re-examination by Dr. von Sallet in the Zeitschrift fiir Nu'mismatih, Berlin, 1878; but his researches have only led him to acknowledge, even more strongly than his predecessors, the hopelessness of the task (see p. 207). SE01‘. 2. SUCCESSORS OF ALEXANDER. 57 I and opened out a more extensive knowledge of that country: and two of the kings in question, Menander and Demetrius, are especially mentioned as having passed the Hyphasis, which had been the limit of Alexander’s conquests to the east, and advanced as far as the Jumna: they also recovered possession of the delta of the Indus, or Pattalene, and extended their dominion over some adjoining districts of the sea-coast.6 But it is probable that these conquests were short-lived. Strabo distinctly observes that they had added nothing to the in- formation previously possessed: and it is certain that we do not find in our extant authorities the slightest trace of any increased acquaintance with India, derived from the establish- ment of this Greek kingdom on its immediate frontiers. One curious circumstance is however recorded, which proves the extent of the commercial relations maintained by these kings with the adjacent parts of India; that when the Greek and Roman merchants visited Barygaza and the ports of Guzerat, more than a century after the Christian era, they still found the silver coins of Apollodotus and Menander commonly cur- rent among the traders of that country.7 § 3. Equally little do we learn during this period con- cerning the provinces beyond the line of the Indian Caucasus. There is reason to believe that the Greek rulers of these countries founded there several flourishing cities, in addition to those already established by Alexander, and that the country rose under their government to a very prosperous condition.8 The manner in which the same result was pro- 6 Apollodorus ap. Strab. xi. 11, § 1, ‘ placed earlier than 120 13.0. The frag- p. 516; xv. 1, § 3, p. 686. This Apol- ‘ ments are published by C. MllllOl‘ in lodorus, who was a native of Artemita his Fragmenta Historicoram Grwl-ornm, in Babylonia, wrote a history of the vol. iv. p. 308. Parthians, which is repeatedly quoted 7 Periplus ZlIa-rz's Err/threes’, c. 47. by Strabo, and appears to have con- 8 Eight new cities were founded by tained some interesting notices eoncern- Alexander in Bactriana and Sogdiana ing the Bactrian kings. His date is (Strab.xi.11, §4). We have no details unknown, but he seems to have written concerning those founded by his suc- after the time of Eucratides. and appa- cessors, but the name of Eucratidia rently after that of Menander also. points distinctly to its having been His work can therefore be hardly founded or rebuilt by Eucratides. [Antiochia 572 HISTORY OF ANCIENT GEOGRAPHY. CIIAP. XIV. duced, at a much later period, under Timour and his suc- cessors, is suficient to show how easily such a state of things may have arisen; and the scanty and passing notices found in the Greek writers of the wealth and power of such monarchs as Euthydemus and Eucratides have been abundantly con- firmed in modern times by the discovery of their coins in such numbers as to bear the strongest testimony to the truth of these statements. But of any extension of their commercial relations beyond the frontiers of the empire of Alexander, or any increased geographical knowledge of Central Asia we hear nothing.9 It is much to be regretted that the history given by, Polybius of the campaigns of Antiochus the Great in Upper Asia, against the lately revolted Parthian and Bactrian kings, has not been preserved to us, but as that monarch did not penetrate beyond the frontiers of these provinces, it could scarcely have added anything very material to our geogra- phical knowledge. § 4. There was indeed one author of earlier date, from whom Eratosthenes appears to have derived much valuable informa- tion. This was PATEOGLES, whose name has been already mentioned in connection with India, and who held for a con- siderable time under Antiochus I., the chief command of all the provinces from the frontiers of India to the Caspian Sea. In this position he appears to have taken great pains to collect statistical and geographical information concerning the coun- tries under his government, and there can be little doubt that we should have derived many valuable details from his work had it been preserved to us. He is praised by Strabo for the accuracy of his statements and the soundness of his udgment, Antiochia in Margiana, and Sotcira in Aria are also ascribed to Antiochus Soter. J ustin’s phrase, where he calls Theo- dotus prtet'ectus ” (xli. 4), is a mere rhetorical exaggeration. 9 We know not what value to attach “mille urbium Bactrianarum. to the vague expression quoted from Apollodorus by Strabo (xi. 11, §1) that the Bactrian kings “extended their dominion as far as the Seres and the Phryncs,” but the passage is interest- ing as being the first geographical notice of a people of the name of Sercs, so celebrated in after times. SECT. 2. SUCCESSORS OF ALEXANDER. as well as for his abstinence from the fables with which so many writers had disfigured their narratives. \Ve are told moreover that he enjoyed the great advantage of a statistical account of the provinces under his rule, which had been drawn up by order of Alexander himself, and was afterwards given to Patrocles by one Xenocles, who had charge of the trea- sury.1 Though it is probable that such a document was drawn up with statistical and financial rather than geographical objects, it must have been in any case a valuable contribution to the knowledge of the countries in question. But it is un- fortunate that the most important geographical statement that is transmitted to us on the authority of Patrocles is one that we now know to be entirely erroneous. In the speech which Alexander himself is represented as addressing to his army on the banks of the Hyphasis, he pointed out that it was a comparatively short distance from thence to the Ganges and the Eastern Sea, and added that this Eastern or Indian Sea was continuous with the Hyrcanian (Caspian) on the one side and with the Persian Gulf on the other, all of them, as well as that which bounded Libya, being in fact portions of the one great circumfluent ocean.2 The voyage of Nearchus conclusively established the fact that this was true as regards the Persian Gulf. This discovery appears to have tended strongly, by a false but natural analogy, to strengthen the belief that a similar communication existed with the Hyrcanian Sea also. Alexander had, as we have seen, already taken measures for an exploring voyage on the Caspian, when all his schemes of this description were interrupted by his untimely death. Nor does it appear that any of his suc- cessors found leisure to resume the project. But from the 1 Strabo, ii. 1, § 6, p. 69. Strabo i ander we have no means of judging; calls it an aua'ypaqbh, or “ register,” an but it is certainly not the composition expression which clearly points to an of Arrian, and on such a point as this official document, not a mere literary would, in all probability, correctly give description- us the views of the king himself, which 2 Arrian, Anab. v. 26. How far the were doubtless shared by the leading speech, as preserved to us, may repre- officers of his army. sent that actually delivered by Alex- 574 HISTORY or ANcIENT GEOGRAPHY. can». xIv. position of Patrocles he would be naturally led to make inquiries into the point, and it is strange that the result of these inquiries was to confirm him in the received view, so that he even maintained expressly that it was possible to sail round from India to the Hyrcanian Sea.3 His authority on this point was unfortunately universally received, and was adopted as conclusive both by Eratosthenes and Strabo. Thus, as Humboldt has observed, the conquests of Alexander became in ,7 this instance the means of retarding, instead of promoting, the progress of geographical knowledge.4 His information concerning the dimensions of the Caspian was more correct, as he justly reported it to be about the same size as the Euxine,5 while in regard to the Oxus and the Iaxartes he stated that they both flowed into the Caspian on its eastern shore, the mouths of the two rivers being separated by an interval of eighty parasangs.6 It is clear therefore that he, in common with the other geographers of his day, had no knowledge of the Sea of Aral, as distinct from the Caspian. But we are certainly not entitled to assume in consequence, as some modern writers have done, that the former was not then in existence, and that the Iaxartes really pursued an unbroken course to the Caspian. It is much more probable to suppose that the Greek settlers on the banks of the Upper Iaxartes, when they were told by the native tribes that the river flowed into a great salt lake or sea, took for granted that this could be no other than the Caspian, with the northern end of which they 3 Strabo distinctly says that it was ' Caspium Seleuco et Antiocho praefec- not an ascertained fact that any one toque classis eorum Patrocle,” Plin. had actually sailed round from India 11. N. vi. 17 , § 58). into the Hyrcanian Sea; but that 4 Aristotle, writing just before the Patrocles affirmed that it was possible expedition of Alexander, appears to to do so (xi. 11, § 6. oz’m 61.4.0A07050'L have held the same view as Herodotus, ‘6%, 8n nepie’rritevadu rules are. riis that the Caspian was a wholly separate ’1V5uci75 é-zrl'rnu‘TpKaz/iav'ftri'oé 'o‘vva'rév, basin, not communicating with any Ila'rpoltkiis e’ipmce). This statement is ‘ other (Meteorologica, i. 13, § 29; ii. 1, strangely misinterpreted by Pliny, who § 10). But his information was evi- says that Patrocles himself had actually dently very imperfect. See Chap. XI. performed the voyage, and in company p. 401. with Seleucus and Antiochus! (“cir- t 5 Strab. xi. 7, § 1. cumvectis etiam in Hyrcanium mare et 6 Strab. xi. 11, § 5. SE01‘. 2. SUCCESSORS OF ALEXANDER. were wholly unacquainted; otherwise they could never have supposed it to communicate with the Northern Ocean. Nor could they have been ignorant of the great river Volga, which flows into it at its northern extremity; but of which no mention is found in any Greek author before the time of Ptolemy.7 1 It has been already pointed out (Chapter VI. p. 213) that the identi- fication of the Oarus of Herodotus with the Volga rests upon very slender foun- dation; but, even if it be admitted, that author had no knowledge of its flowing into the Caspian: on the con- trary he supposed it to fall into the Palus Maaotis, of the extent of which he had a very exaggerated idea. ( 576 ) CHAPTER XV. THE PTOLEMIES IN EGYPT. SECTION l.-—The Ptolemies in Egypt—The Red Sea—The Nile. § 1. MEANWHILE the geographical knowledge of the Greeks was receiving important accessions in another quarter. Among the different kingdoms constituted out of the empire of Alex- ander there was none which inherited a larger portion of Greek civilization than that of Egypt. Founded in the first instance by a statesman of great wisdom and enlarged intelligence, it continued under his son and grandson Ptolemy Phila- delphus and Ptolemy Euergetes, to be not only the centre of literary cultivation and learning for the Hellenic world, but became at the same time the most important emporium of trade and centre of commercial enterprise. The position of Alexandria was admirably selected with a view to its being the commercial capital of the Eastern Mediterranean; its port was spacious and secure, and it was free from that tendency to fill up with sand, which is the inevitable disadvantage of all the ports along the coast of the Delta. But it was not to the Mediterranean alone that the Ptolemies directed their atten- tion. Their new capital was situated in a convenient position for communicating with the Red Sea, and by that means with the spice and incense bearing tracts of Arabia and the opposite coasts of Africa, as well as for the ‘more distant, but still more important, trade with India. § 2. It must be remembered that, throughout the reigns of these first Egyptian monarchs, the cities of Phoenicia, which had always continued to carry on a flourishing commerce, and had hitherto possessed the exclusive monopoly of the trade of the Red Sea,—the Egyptians themselves having never shown SEC'I‘. 1. THE PTOLEMIES IN EGYPT. 577 any aptitude for maritime commerce—were subject to the dominion of Egypt; and the Ptolemies could thus avail them- selves both of the skill of'the Phoenician navigators and of the relations which their merchants had already established in these quarters. It was only necessary to divert the line of traffic,—-which had been previously carried across the isthmus of Suez to the Mediterranean, and thence direct to Tyre,—to the Egyptian capital. With this view Ptolemy Philadelphus began with founding at the extreme head of the Red Sea, nearly in the position of the modern town of Suez, a city to which he gave the name of Arsinoe; and he next proceeded to open a direct communication by canal between that city and Alexandria.1 But notwithstanding the facilities thus obtained, it was found that the dangerous character of the navigation of the inner bight or gulf of the Red Sea—the Hcroopolites Sinus or Gulf of Suez—was such as to outweigh the advantage of the proximity to Alexandria; and a port was in consequence established nearly five hundred miles lower down, on the Egyptian shore, which gradually became the chief emporium for all the trade of the Red Sea. To this city Philadelphus gave the name of his mother Berenice, as he had called the other after his wife Arsinoé. From thence the merchandise had to be carried over land, a distance of more than 200 miles, to Coptos 011 the Nile, from whence it descended that river to Alexandria.2 Two other settlements were founded during the same reign on the coast between Arsinoe and Berenice, Philo- tera and Myos Hormus, the latter of which, at a later period, 1 This canal was in fact only the completion or restoration of one, which had been begun long before by the Egyptian king Necho, and afterwards continued by Darius Hystaspes; but had apparently been neglected and fallen into disuse (Herodot. ii. 158; and see the note of Sir G. Wilkinson on the passage, in Rawlinson’s Herodotus, vol. ii. p. 243). It began from the Pelusiac branch of the Nile, above the city of Bubastis, and was carried from VOL. I. thence to the Bitter Lakes, where it turned to the south, and thence pro- ceeded direct to Arsinoé (Herodot. l. c. : l’lin. II. N. vi. 29, § 165). In this latter part of its course, therefore, it must have followed nearly the same line with the modern canal across the isthmus of Suez. 2 Strabo, xvii. p. 815. The ruins of Berenice were first visited by Belzoni, and have been more fully described by Sir G. \Vilkinson and by Lieut. Wellsted in the Journal of Geogr. Soc. vol. v1. p. 96. 2 P 578 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XV. obtained the preference over Berenice, and became the chief trading-place with India.3 Its greater proximity to Coptos was probably the chief cause that led to this change.‘ § 3. But these establishments were by no means adequate to fulfil the objects of Ptolemy, who aspired to nothing less than securing the complete command of the Red Sea. With this view he founded towns, or at least established permanent stations, from distance to distance, all along the Egyptian coast of that sea from Berenice to the straits of Bab-el- Mandeb. One of the most important of these—situated in latitude 18° 40', about 50 miles below the modern port of Suakim —-was named Ptolemais Epitheras, and destined principally, as its name indicates, as a station from which to carry on expeditions into the interior, with a view to the capture of elephants. This had indeed become a great object with the Egyptian monarchs. Ever since the conquests of Alexander in India had brought the Greeks into familiarity with the use of elephants in war, they had been eagerly sought after by the rival monarchs, and when the present of five hundred of them to Seleucus by the Indian king Sandro- cottus5 had for a time given to the Syrian kings a pre- ponderance in this respect with which no others could compete, the Ptolemies turned their attention to the possi- bility of training African elephants in the same manner as the Indian ones, and thus deriving an inexhaustible supply from the regions within their own command.6 The position of Ptolemais Epitheras7 gave them direct and ready access to the tracts on the banks of the Atbara and its tributaries, the very 3 Strabo, ii. p. 118; xvii. p. 815. Concerning the position of Myos Hor- mus, see Note A, p. 607. 4 Thus we find the Roman General, ZElius Gallus (in the reign of Augustus), on his return from his expedition to Arabia, landing at Myos Hormus and marching with his army from thence to Coptos (Strabo, xvi. 4, p. 782). 5 See Chapter XIV. p. 555. 6 See Note B, p. 607. 7 The site of Ptolema'i's Epitheras may be fixed (approximately at least) in the neighbourhood of Cape Mugda or Mikdam, in latitude 18° 40' (see C. Miiller’s note to Agatharchides, p. 172). It was correctly placed within a short distance of this position by D’Anville, but Dr. Vincent transferred it much farther south, being misled by the error of Ptolemy, who followed Eratosthenes in placing it in the same latitude with Meroe. SECT. 1. THE PTOLEMIES IN EGYPT. same districts which have recently formed the well-known hunting-grounds of Sir Samuel Baker; and which no doubt then, as now, abounded in elephants and all other kinds of “large game.” 8 §4. Beyond this again towards the south, the name of another city of Berenice (called for distinction’s sake Berenice- ad-Sabas), within less than a degree of the straits of Bab-el- Mandeb, another Arsinoe immediately close to the actual straits, and a third Berenice just without them, adjoining the headland of Deiré (Itas Bir), sufficiently prove the zeal and energy with which the Ptolemies carried out their policy of occupying and securing the whole of these coasts. The barren and inhospitable nature of the adjoining regions would exclude the possibility of colonization in the higher sense of the word, and all these stations in the neighbourhood of the Straits appear to have been employed principally with a view to catching elephants in the interior. But immediately beyond Deiré began a long line of coast stretching out as far as Cape Guardafui—the Noti Keras or Southern Horn of Strabo --which was an object of considerable interest to the Greeks from its producing, not only myrrh and frankincense, like the opposite shores of Arabia, but cinnamon also in such quantities that the whole tract came to be known both to Greek and Latin geographers, as “ the Land of Cinnamon.” 9 No trace is found of any settlement in these parts being established by the Ptolemies, though successive Greek navigators set up stelw as landmarks along the coast,——probably indicating in each case the farthest point attained,—-which bore the names of those who had erected them. It was'thus that the names of Pytholaus, Lichas, Pythangelus, Leon and Charimortus were perpetuated.1 8 It was probably from the same I (Athenazus, v. 32, p. 201). quarter that they imported the various 9 f7 mvuanwuécpopos. This designation strange animals that figured in the was clearly well established in the festive processions at Alexandria, time of Eratosthenes, as we find it con- among which we find mention as early tinually cited from his work by Strabo. as the reign of Philadelphus of a ! Concerning the production of cinnamon camelopard and an Ethiopian rhino- ; in these countries, see Note C, p. 608. ceros, besides oryxes, bubali, &c. i 1 Strabo, xvi. 4, §15, p. 774. 2P2 580 HISTORY or ANCIENT GEOGRAPHY. CHAP. xv, ' With this exception very little notice has been preserved to us of the enterprising commanders by whom the exploring expeditions of Ptolemy Philadelphus were conducted, or of the successive steps by which the chain of Egyptian outposts already described was established along a line of coast fifteen hundred miles in extent.2 There is however no doubt that the stations near the Straits, which bore the names of Berenice and Arsinoé were erected either during the reign of the second Ptolemy, or at latest in that of his successor, Ptolemy Euergetes. This period, indeed, appears to have been that when the Egyptian commerce was at its height, as it was that in which the kingdom in general enjoyed the greatest prosperity, and it may be reasonably presumed that their trading voyages had already been extended to their utmost limits. § 5. Much the most interesting and important question in regard to the maritime commerce of the Ptolemies is that which relates to their trade with India. That the Egyptian Greeks under their rule carried on an extensive trade in Indian commodities, and that Alexandria became at an early period the chief emporium of this lucrative commerce is a point upon which all writers are agreed. But of the nature and character of this trade we have very imperfect information. It has been assumed by many modern writers that they traded direct with the Indian peninsula, although from their ignorance or imperfect knowledge of the monsoons, and probably also of the general configuration of the countries bordering on the Erythraean Sea, their ships were compelled to follow the circuitous course from the Straits of Bab-el—Mandeb along the coasts of Arabia, Carmania and Gedrosia to the mouths of the Indus. But it will be found on examination that there is great 2 The names of Satyrus and Eumedes §§ 5, 7). We learn also from Diodorus have been, however, recorded among that one of the captains sent out by those who were thus employed by Ptolemy Euergetes, to explore the Ptolemy Philadelphus, the former as Tloglodytic coast, was named Simmias. the founder of Philotera, the latter of His account of those regions was cited Ptolema'is Epitheras (Strabo, xvi. 4, by Agatharchides (Diodor. iii. 18). SECT. 1. THE PTOLEMIES IN EGYPT. , 581 reason to doubt whether any direct trade with India ever existed under the Ptolemies. While we have abundant notices of their commerce with both shores of the Red Sea, and the countries on each side after passing out of the Straits of Bab- el-Mandeb, for a certain distance; we find all information suddenly come to an end at Cape Guardafui on the one side, and at the boundary of the land of the Sabaeans on the other. Eratosthenes indeed was acquainted with the name of Hadra- maut (Chatramotitae), as one of the provinces or portions into which Arabia was divided; but the later writers, Agatharchides and Artemidorus, who describe the coasts in detail, give no names or stations along this part of Arabia or the adjacent district of Oman. The name of the headland of Syagrus- (Cape Fartak) which assumes so important a part in the works of later geographers, is not found in Strabo, and was therefore probably unknown to Eratosthenes, as well as to his immediate successors. The same thing may be remarked of the island of Socotora, on the opposite shore of the narrow sea, which under the name of the island of Dioscorides became familiar to the Greeks in later times, when they carried on direct trade with India. On the other hand it is remarkable that they not only speak of the land of the Sabaaans (the district now known as Yemen) as one of the richest and most flourishing countries in the world—a character which it certainly does not deserve on its own account, though a tract of considerable fertility as compared with the rest of Arabia—- but they expressly state that this prosperity was owing in part to their extensive commerce, the Sabteans themselves having many ships, some of them of large size, while their ports were frequented by trading vessels from all the neighbouring nations, especially from the mouth of the Indus, where Alex- ander had established his naval emporium.3 It thus becomes extremely probable (to say the least) that the Alexandrian merchants may have derived their Indian wares from the ports 3 Agatharchides, § 102; ap. Diodor. iii. 47 ; Artemidorus ap. Strab. xvi. 4, §1-9. 582 HISTORY or ANCIENT GEOGRAPHY. CHAP. xv. of Sabaea, whither they would be brought by native traders, whether Arabian or Indian; and they would thus avoid the necessity of the long and perilous voyage to the coasts of Hindustan.. § 6. The account given by Agatharchides of the commercial character of the Sabaeans is confirmed at a later period by the author of the Periplus of the Erythraean Sea, who speaks of the merchants of Muza (a port just within the straits) as carrying on an extensive commerce and sending their own ships to the east coast of Africa and to the well-known Indian port of Barygaza.4 The same writer expressly tells us that the port of Arabia Felix, by which he unquestionably means Aden,’ was in former days, .when navigators did not yet venture to proceed from Egypt direct to India, or from India to Egypt, the emporium at which they mutually exchanged their commodities.5 Though he does not in terms apply this to the trade of Egypt under the Ptolemies, the comparison which he implies with the state of things in his own day (when the course of trade had totally altered) leaves little doubt that this was the period to which he alludes.6 Had the Greek navigators under the Ptolemies been in the habit of trading directly with India, it is inconceivable but that they should have brought home some notices of the country, its sea-coasts and the ports they visited, but, as we shall hereafter see, neither Eratosthenes nor any other of the geographers consulted by Strabo—though writing for the most part at Alexandria—contain the slightest additional informa- tion from this source, their knowledge of the continent of India being derived exclusively from the writings of Megas- thenes and the other contemporaries of Alexander. Even the name of Barygaza, so important an emporium of Indian com- 4 Peripl. Maris Erythr. § 21. found in any of these earlier writers 5 Ibid. §26. He expressly compares I (Agatharchides, Artemidorus, or Era- it in this respect with Alexandria, as a tosthenes) of the port and promontory place that exported not only its own of Aden, which is not only one of the commodities, but those which had been most marked features on the line of brought thither from a distance. coast, but must always have been the 6 It is, however, very remarkable, chief natural emporium for all this part upon this supposition, that no notice is of Arabia. SECT. 1. THE PTOLEMIES IN EGYPT. 58 3 merce, and situated at so short a distance from the mouths of the Indus, is not mentioned by Strabo and was apparently unknown to the Alexandrian geographers.7 \Ve can hardly therefore be mistaken in concluding that, at all events under the reigns of the earlier Ptolemies, the voyages of the Alexandrian traders did not extend farther than Cape Guardafui on the one side, and the neighbourhood of Aden on the other; and that the lucrative trade with India, of which they possessed to a great extent the monopoly, was of a kind similar to that of the Venetians in later times, who supplied the rest of Europe with Indian commodities, which they them- selves purchased at Alexandria. § 7. We have much less information concerning the progress made by the Ptolemies in the exploration of the interior of Africa; but we know that on this side also they made great additions to the geographical knowledge previously possessed of the countries vaguely known by the designation of Ethiopia. It was indeed impossible that a civilized and powerful state should be established in Egypt, without attempting to push exploring expeditions up the valley of the Nile, both in the hope of solving the long controverted question of the origin of that river, and with the more practical object of making acquaintance with the nations and countries on its banks, which were known to produce ivory, ebony, and other articles eagerly sought for by the Greeks, and which were supposed, though with little foundation, to be‘ equally abundant in gold. The Ethiopians of the neighbourhood of Meroé had at this period assumed something of the character of a civilized community, and under the government of a native ~ruler, called 7 Strabo in describing the extensive trade carried on in his day from Myos Hormus with India direct, expressly contrasts it with the state of things when very few ventured to sail thither for commercial objects (1rpd'repov e’1r2 'rc'iw I'I'roixeptai'xibv Bao'me'wv 6Ai'ywz/ ‘trav- 'ro'nrarn tiappouv'rwu 'n'Aeiv rcal 'rbv ’Iv5ucbv é/uropei'ieoflar ¢6p'r01/, ii. 5, § 12, p. 118). The voyage of Eudoxus of Cyzicus (about 120 12.0.), who certainly did proceed to India, is distinctly spoken of as something altogether new and exceptional (Strabo, ii. 3, §4, p. 98). See Chapter XVIII. sect. 5. 584 HISTORY or ANCIENT GEOGRAPHY. 01m. xv. by the Greeks Ergamenes, appear to have maintained friendly relations with Ptolemy Philadelphus, which would materially assist in the prosecution of researches farther south.8 At a somewhat later period Meroé itself appears to have become a dependency of, if not actually subject to the Egyptian monarchy. This circumstance readily accounts for the in- creased knowledge obtained by the Greeks at this period of the course of the Nile above the second cataract and of the several confluents or tributaries of that great stream. Herodotus, as we have seen, knew nothing of any rivers flowing 'into the Nile, and justly regarded it as one of the great peculiarities of that river, that during so very long a course it had no tributaries. Moreover, while he was familiar with the name of Meroé, as the capital city of the Ethiopians, he has no mention of the “island,” as the surrounding territory was called by later geographers, in consequence of its being nearly encircled by the Nile and one of its tributaries. Eratosthenes, on the contrary, correctly described the “island ” of Meroé as formed by the junction of the Nile with the Astaboras,—the stream which is now called, in the upper part of its course, the Tacazze, but near its junction with the Nile still preserves the name of Atbara. Above this was another similar island, formed by two rivers called the Astapus and Astasobas, concerning which there seems to have been some confusion, some writers describing them as two separate affluents of the Nile, others more correctly asserting that the Astapus was in reality the main stream of the Nile itself.9 It was said to flow from some lakes farther to the south; but no one pretended to have any accurate knowledge of its source. On the other hand, the correct view of the cause of its periodical inundations, that they were produced by the summer rains in these southern regions, appears to have been generally adopted, though 8 See the article ERGAMENES in Dr. 9 Eratosthen. ap. Strab. xvii. l, p. Smith’s Diet. of Ancient Biography, 786. vol. ii. SE01‘. 1. THE PTOLEMIES IN EGYPT. > 585 Various other fanciful theories were from time to time promul- gated by philosophers.1 §8. There is therefore no doubt, notwithstanding the con- fusion in regard to the names, that the Greeks were at this time well acquainted with the course of the Nile as far as Khartoum, at the junction of the two streams now known as the Blue and \Vhite N ile—the Bahr el Azrek and Bahr e1 Abiad; and for a certain distance up those rivers. It was above their junction, in the so-called island formed by the two, that dwelt a people termed by Eratosthenes and later writers the Sembritae, who, according to the current tradition, were the descendants of a body of Egyptian exiles who had quitted their country in the reign of Psammetichus.2 7We here recognize at once the Automoli of Herodotus, though the reports which had reached that historian had greatly exaggerated the remoteness of their situation.3 They evidently occupied in reality the region now known as Sennaarf" It was the natural consequence of this intercourse with Meroe that the Greeks of Alexandria should acquire accurate information concerning the course of the Nile between that city and Syene, and from this period accordingly we find them well acquainted with the great bend formed by the river in this part of its course, as well as with the second, or as it was termed, “ the great” cataract, concerning which such absurd - fables were propagated in later times? But the difficulties presented by this, as well as by the other obstacles to the navigation of the river in this part of its course, would naturally I Strabo, xvii. 1, p. 789. i above Khartoum. 2 Eratosthen. ap. Strab. xvii. 1, p. 5 No distinct mention of the Second 786. Cataract is found in Herodotus, though 3 Herodot. ii. 30. See Chapter VIII. l he was aware that the navigation of 4 The name of Sennaar is given at the Nile in this part presented great the present day to the district between difliculties (see Chapter VIII). The the Bahr el Azrek or Blue Nile and first mention of “the Great Cataract” the Bahr el Abiad or White Nile, con- is found in Eratosthenes ; but he does stituting a quasi. island precisely not appear to have countenanced the analogous to the so-called “ islanc ” of absurd exaggerations concerning it, Meroii. The town of the name is situ- which are found in Pliny and later ated on the Blue Nile, about 150 miles writers. 586 HISTORY or ANCIENT GEOGRAPHY. CHAP. XV. O prevent any considerable trafiic from being carried up or down the Nile; and there is reason to infer that the intercourse with Meroe was carried on principally by caravans across the desert, from the ports on the Red Sea. The direct distance by this line of communication is stated at only ten or twelve days’ journeyf‘ N o allusion is found in any ancient writer to the route which has been generally followed by modern travellers, which quits the Nile at Korosko and strikes across the desert to Abu Hamed, thus cutting off the whole of the great bend of the river, as well as avoiding the cataracts. § 9. We are told by Diodorus that Ptolemy Philadelphus himself carried his arms into Ethiopia, and thus led to a more extensive knowledge of the regions of the interior :7 but the expressions of the historian are vague, and it is very doubtful whether that monarch himself conducted any expedition of importance. Mero'é also appears to have continued virtually independent, and we have no proof of any attempts being made to establish the authority of the Egyptian monarchs farther up in the interior. The supposition that Ptolemy Euergetes had made extensive conquests in Ethiopia, at the head of an army which he conducted in person, rests solely on a misconception of the celebrated monument of Adulis, which records the exploits of an Ethiopian king of much later date. That monument was undoubtedly erected in the first instance during the reign of Ptolemy Euergetes, but there is no reason to suppose that it was set up by that monarch in person or that he had himself visited this remote corner of his dominions.8 A considerable amount of information seems to have been collected by the explorers and traders of the Egyptian mo- narchs concerning the various wild tribes in the interior, with 6 Eratosthenes ap. Strab. xvii. 1, p. 786. In another passage, where he is following Artemidorus, Strabo states the distance from Meroé to the Arabian Gulf at fifteen days’ journey for an active traveller (engrave was. xvi. 4, p. 771). At the present day, since the foun- dation of Khartoum, and the establish- ment of regular communications with the upper regions of the Nile, most travellers proceed by sea to Suakim, and thence across the desert to Khar- toum. ’ Diodor. i. 37. 8 Note D, p. 609. SE01‘. 1. THE PTOLEMIES IN EGYPT. 587 whom they came in contact in their expeditions in pursuit of elephants and other game: but no notice is found of the rugged and lofty table-land of Abyssinia, which’ forms so im- portant a feature in the physical geography of these regions: nor even of the great mountain barrier that bounds it on the east, and presents so conspicuous an object, when viewed from the Red Sea. Nor do they appear to have obtained any knowledge of the countries lying to the west of the Upper Nile—now known as Kordofan and Dar Four; while below these the great Nubian desert opposed an impenetrable barrier to their explorations in this direction. The Egyptian kings had indeed early extended their dominion over the whole of the Cyrenaica, as far as the confines of the Carthaginians, and had founded a city of the name of Berenice on the very border of the Great Syrtis ; but we hear nothing of their having instituted any exploring expeditions into the interior of Libya, or established any com- mercial relations with the native tribes of Central Africa. It is remarkable that the Greek monarchs of Egypt, like the native Egyptian kings, do not appear to have ever thought of introducing the use of camels into Africa: without which useful auxiliary it was almost impossible to carry on any extensive journeys or explorations into the interior of the continent. § 10. To the period of the earlier Ptolemies belongs a writer, who is frequently cited by extant geographical authors, and whose work appears to have exercised considerable influence upon the progress of geography in the times immediately succeeding him. TIMosTHENEs, a native of Rhodes, who held the important position of admiral of the Egyptian fleet under Ptolemy Philadelphus,9 drew up a treatise “ concerning Ports,” 9 Strabo (ix. 3, p. 421) terms him His work, which was in ten books 6 vai'iapxos 1'03 Bev'répov H'rohenaiov, but (Strabo, l. 0.), would appear to have Marcian of Heraclea calls him apxua- .been still extant in the days of Marcian. Bepm'y'r'ns, a term which would imply It is also referred to by Agathemerus more scientific acquaintance with prac- (Geog'raph. i. § 7), but it is always tical navigation (Epit. Artemidor. §2). ! difficult to judge whether these cita- 588 HISTORY OF ANCIENT GEOGRAPHY. GHAP. xv. which is repeatedly cited by Strabo, and appears to have been one of the chief authorities followed by Eratosthenes. It was evidently not confined to the mere description of the cities or harbours on the Mediterranean, but gave the distances from one to the other, with indications of their relative position and bearings.1 Such a work was apparently designed as a prac- tical guide to the navigator, but must have contributed also materially to the more definite geographical knowledge of the seas and coasts which it comprised: and it is not surprising to find that it became one of the principal authorities upon this particular branch of the subject. Eratosthenes indeed, whose work so long retained its position as the standard treatise on geography, is said to have made such extensive use of the work of Timosthenes, as to amount to direct plagiarism.2 But from the total loss of his writings we are unable to judge of the real extent of the obligations of the great geographer of Cyrene to his predecessor. It appears certain however (so far as we can judge from the extant citations) that the treatise of Timosthenes was confined to the “Inner Sea ” or the Medi- terranean and its dependencies: it could therefore hardly have added much to the geographical knowledge of the Greeks in point of extent, however much it may have contributed to accuracy of details. Strabo indeed speaks of Timosthenes, as well as his successor Eratosthenes, as showing great ignorance of Gaul, Spain, and the western regions of Europe in general: a criticism that is probably well-founded, though Strabo him- self (as we shall hereafter see) was far from possessing as correct ideas as he himself imagined, concerning the western half of the Mediterranean.3 tions by late authors are really taken from the original works, or copied at second hand. ‘ Thus he stated that Metagonium on the coast of Africa was opposite to Massilia, an error for which he is justly censured by Strabo (xvii. 3, p. 827). 2 Marcian of Heraclea goes so far as I to assert that Eratosthenes transcribed the whole treatise into his own work, making only a few additions (Epit. Artem. § 3). But this statement seems wholly at variance with the much more trustworthy authority of Strabo, that while Eratosthenes praised the work of his predecessor above all others, he differed from him upon many points (251/ e’vrawei‘ ,uév e’rceTz/os fwbua'ra 76W Zim- Awv, filacpwvév 5’e’Aé1/Xeq'at 1rpbs afrrbu 'n'AeTo'ra. Strabo, l, §40, p. 92.) 3 Marcian also, in whose days of course all parts of the Mediterranean Seen 2. PROGRESS IN WESTERN EUROPE. 589 In one respect, Timosthenes (we are told) made a step in advance of all his predecessors—that he was the first to point out the arrangement of different countries according to the winds that blew from the different quarters, or as we should say, the different points of the compass, and in so doing intro- duced a greater number of these divisions than had been before in use.‘i Thus he placed the Bactrians in the extreme east (Apeliotes), the Indians in the south-east (Eurus), the Ethiopians and the Red Sea in the south-south-east (Phoenix), the Ethiopians above Egypt in the south (N otus), the Gara- mantes above the Syrtis in the south-south-west (Leuconotus), the western Ethiopians above the Mauri in the south-west (Libs); the Columns of Hercules, and the confines of Africa and Europe in the west (Zephyrus), Iberia (Spain) in the north-west (Argestes), the Kelts in the north-north-west (Thras- cias), the Scythians above Thrace in the north (Aparctias), the Pontus, Maaotis and Sarmatians towards the north-north-east (Boreas), and the Caspian Sea and the Sacae towards the north-east (Caecias).5 SECTION 2.-—Progress of Discovery in Western Europe.— Pytheas—Tz'mazus. § 1. \Vhile the conquests of Alexander the Great and the kingdoms established by his successors were extending the geographical knowledge of the Greeks towards the east, other causes were producing the same effect in regard to the west and north-west of Europe. We have seen how imperfect was the information possessed by the Greek geographers as late as the time of Ephorus and the coast-describer Scylax, concerning everything beyond the Pillars of Hercules; and although we were well known, censures Timosthenes i expresses his surprise that Eratosthenes for his ignorance of the countries around should have followed him so implicitly the Tyrrhenian Sea, and from thence I‘ (Epit. Artem. § 3). to the Straits, as well as of the southern ' 4 Note E, p. 610. coast from Carthage to the Straits, and l 5 Agathemerus, e. 2, § 6. 59O HISTORY OF ANCIENT GEOGRAPHY. CHAP. XV. know that the name of the Cassiterides—as islands in the western sea, from whence the Phoenicians brought large supplies of tin—was well known before the time of Herodotus, that historian had been unable to obtain any definite informa- tion concerning them,6 and the Greek writers who followed him appear to have been equally ignorant in this respect. But about the period at which we are now arrived—the generation following the death of Alexander—we find alto- gether new names and new geographical notions concerning the western regions of Europe, which, although still in a very vague and fluctuating form, had acquired a hold upon the popular belief, and undoubtedly indicate an increased ac- quaintance with this portion of the world. For these notions, and for the first introduction into the domain of geographical science of some leading facts, though confused and distorted by many errors and fables, the Greeks were indebted to a writer named PYTHEAS, a native of Massilia. Though the discoveries, or alleged discoveries, of Pytheas, were a subject of much controversy among ancient writers, and were fully discussed by some authors still extant, no definite statement has been preserved to us as to the period at which he lived, or the date of the publication of his work. Even as to its form and character we are left almost wholly in the dark,7 and with this additional disadvantage, that what little informa- 6 Herodot. iii. 115. See Chapter VI. a general expression. So far as we can I) 164 gather, it seems probable that Pytheas 7 The title of his work is uncertain. The ancient authors cite it under various names. Marcian of Heraclea classes him amongst the writers who had left Hepurao'i, but it is clear that he uses the word in a general sense, as he includes Eratosthenes in the same category. The Scholiast on Apollo- nius Rhodius (iv. 761) cites it as l‘fis wepiooos, but it is very doubtful whether he had himself seen the work. An astronomical writer of the first century B.0. refers to his authority, év 'ro'ls 1rep2 'rofi mum/oi); but this is probably only wrote a geographical work, in the course of which he gave an account of his own voyages and explorations, but described other countries also, from the information that he had collected. There is no reason to suppose that he left more than one treatise of the kind. The original work may probably have been lost at an early period. Strabo evidently knew it only through Eratos- thenes and Polybius ; and there can be little doubt that the citations of Pliny, and later Greek writers, are in like manner all made at second hand. SE01‘. 2. PYTHEAS. 591 tion we possess concerning it is derived principally from the controversial notices of hostile critics, who were disposed to decry the whole production as a tissue of fables. \Vith regard to his date, it is certain that he wrote not only before Eratos- thenes, who relied much upon his authority, but before Dicaearchus, who was a pupil of Aristotle, and died about B.0. 285.8 He may therefore probably be regarded as about contemporary with Alexander the Great. § 2. According to Polybius, Pytheas had given an account of a voyage undertaken by himself, in which he had not only visited the island of Britain, but had travelled over a con- siderable part of it, and stated it to be more than 40,000 stadia (4000 G. miles) in circumference.9 Beyond Britain to the north lay another island, to which he gave the name of Thule, and he stated that the sea in these parts assumed a thick and sluggish character, like neither land nor sea, but resembling in consistence the molluscous animal, or jelly-fish, called the pnlmo mar/inns, or sea-lung. This substance he had himself seen, but the other phenomena he described only from hearsay. Returning from this expedition to Britain, he visited the whole of the coasts of Europe bordering on the ocean, from Gadeira (Gades) as far as the Tanais.1 The whole of this account is rejected by Polybius in a summary manner, and he elsewhere asserts broadly that the whole of the north of Europe, from N arbo in Gaul to the Tanai's was still wholly unknown, and that everything which was related concerning it was mere fiction.2 Eratosthenes was more cautious, and while he appears to have doubted some of the statements of Pytheas, he gave credence to those which related to Britain as well as ‘to the coast of Iberia (Spain) and Gadeira.3 He also received as a 8 According to Strabo, Polybius 1 Polyb. ap. Strab. ii. p. 104. censured Eratosthenes for believing 2 Polyb. iii. 38. If these expressions Pytheas, when Dicaearchus had not are to be taken literally, he must have done so (Strab. ii. p. 104). It appears treated the existence of Britain as also from Pliny (xxxvii. 2, §36) that equally fabulous with that'of Thule. Timaeus, who wrote about 264 13.0., had But this is scarcely probable. made use of the work of Pytheas. 3 Eratosth. ap. Strab. l. c. 9 Note F, p. 612. . 592 HISTORY OF ANCIENT GEOGRAPHY. 01m. xv. fact the existenceof Thule, as the most northern land known, a view which, though rejected both by Polybius and Strabo, came to be part of the generally received system of ancient geographers. §3. In attempting to determine the degree of credit that can be attached to Pytheas, we labour under the very great disadvantage that the original work is lost to us, while the few notices that have been preserved relate almost exclusively to the most disputed points, and to those statements that were con- troverted by later authors. As we learn from Polybius that Eratosthenes relied mainly upon Pytheas for the account of Iberia, we must presume that the latter had given some account of his voyage along the Atlantic coast of Spain, and doubtless that of Gaul likewise. Had we possessed this account we should have been far better able to judge of the general trustworthiness of his information, and could hardly have been left in doubt as to whether he had really made the voyage or not. It is true that Strabo speaks in disparaging terms of the ignorance of Eratosthenes concerning Iberia and the west of Europe in general :4 but his censures are not always well- founded; and it is remarkable that the only statement which has been preserved to us from the work of Pytheas concerning the western coasts of Gaul, is one in which he shows himself better informed than Strabo. He represented the land of the Ostimii—the Osismii of later authors5—who inhabited a part of Bretagne, as forming a great promontory, extending far out to sea.6 Strabo, on the contrary, attached no importance to the projection of this part of the coast, and thus gave an entirely wrong configuration to the Gaulish coast adjoining the ocean, by neglecting one of its most essential features. 4 Strabo, ii. 1, p. 93; 4, p. 104. parison of the three, which all unques- ‘’ The name is indeed found in our tionably refer to the same people, existing text of Strabo only in a corrupt leaves little doubt that the true reading form; the MSS. giving in one passage is, as restored by Kramer, ’nf Hippar- 1 made observation with the gnomon), chus that Byzantium and l‘assilia lay ! was almost exactly correct (Strab. ii. 5, in the same parallel of latittle. But I p. 115). it is remarkable that here '. was the 7 Note H, p. 613. 600 HISTORY or ANCIENT GEOGRAPHY. Grow. xv. had visited Iceland, where he could himself have really wit- nessed the phenomenon. But he was undoubtedly correct in asserting that Thule, which he placed six days’ voyage north of Britain, lay far to the northward of the lands on the Borys- thenes, in the interior of Scythia, which were regarded by many of the Greeks as the most northern region of Europe. In this conclusion he was followed both by Eratosthenes and Hipparchus, though Strabo rejected it with undeserved con- tempt. On another point also Pytheas was the first to com- municate to the Greeks more definite and correct notions. This was with regard to the tides of the Ocean, concerning which he reported that they increased as the moon became full, and diminished as she waned.8 Though of course this statement is not accurate, it not only shows a clear perception of the main fact that the tides were produced and regulated by the moon, but an acquaintance with their periodical fluc- tuations, in accordance with the phases of that luminary.9 § 8. It is remarkable that no mention is foundin connection with Pytheas and his voyage of the Cassiterides J1‘ Tin Islands, from which the Phoenicians continued to drawtheir principal supplies of that valuable metal—the discoverf or exploration of which would seem to have naturally forned one of the principal objects of his voyage. This is tle more remark- able as we find in the time of Posidonius tiat Massilia had opened a regular trade with these islands, 1nd their tin was brought overland to that city from the weste'n ports of Gaul.1 There can be little doubt that at this periid Massilia, which had long been a flourishing commercial cty, was extending its trade in all directions through Gaul, aid to the shores of the Ocean, both towards the west and the rorth. The prospect of competing with the Phoenicians in th lucrative trade in tin would naturally offer special attractins; as did that for and to havr confounded these fluctua- tions withahe daily variations of the 9 Plutarch, however, appears to have tides. misunderstood what he was reporting, l 1 Posida. ap. Strab. iii. 2, p. 147. 8 Plutarch, dc Placit. Phz'losoph. iii. 17. SECT. 2. TIMZEUS. 601 amber with the North Sea. In both cases it is probable that the Greeks of Massilia had very imperfect ideas of the position of the lands from which these valuable commodities were brought: but they had doubtless obtained vague information concerning them from the native traders, and these would stimulate them to further inquiries, and explorations on their own account. This would account for the extension of their geographical knowledge in this direction, and may have been the cause that induced Pytheas to undertake a voyage from Gades to Britain and the shores of the North Sea. § 9. Another writer, who appears to have contributed in a con- siderable degree to the extension of the geographical knowledge of the Greeks in regard to the west of Europe, was TIMZEUS, a native of Tauromenium in Sicily,2 who Wrote an elaborate historical work, in which he treated very fully of the founda- tion and history of the Greek colonies in Italy and Sicily; a subject that naturally led him to treat of the western parts of the Mediterranean, and the countries bordering upon them. Thus we find from existing fragments that he gave some account of the Tyrrhenians or Etruscans, the Romans,3 and the Carthaginians, as well as of the principal islands in this part of the Mediterranean, Sardinia, Corsica, and the Gymnesian Islands, known to the Romans as the Balearic. He is severely censured by Polybius for the inaccuracy of his geographical statements, and especially for his ignorance of the natural productions of the different countries that he described ;4 but he himself boasted of the great pains and labour he had in- curred in collecting information concerning the manners and 2 The exact age of Timaeus cannot be determined, neither the date of his birth nor that of his death being accu- rately known ; but his great historical work (of which that of Polybius was in some sense a continuation) ended with the year B.0. 265 (Polyb. i. 5). As he lived to the age of 96 he probably sur- vived its completion by many years; ‘ but the date of its publication is the most material. 3 Timaeus was the earliest author who assigned a date for the foundation of Rome, which he regarded as contem- porary with that of Carthage, and placed it 38 years before the First Olympiad, or B.0. 810 (Dionys. Halic. Ant. Rum. i. 74). 4 Polyb. xii. 3, 4. 602 HISTORY OF ANCIENT GEOGRAPHY. OHA'P. XV. customs of the Ligurians, Celts, and Iberians ;5 and there is little doubt that had this part of his work been preserved,‘ we should have found in it many curious and valuable notices. Of geographical observations, in the strict sense of the word, we find very few cited, and it is evident that Timeeus paid much less attention to this branch of his subject than his predecessor Ephorus had done.6 The only example that has been preserved to us of his attempts at the explanation of physical phenomena is singularly unfortunate. We are told that he accounted for the flux and reflux of the waters of the Ocean, by supposing the water to be driven back by the flood- ing of the great rivers that flowed from the mountains of Gaul, and to return as these subsided.7 Such an explanation shows an entire want of comprehension of the nature of the phenomenon itself. It is curious only as showing that the Greeks were already familiar with the fact of great rivers (the Garonne and the Loire) flowing through Gaul into the Atlantic Ocean. _ Timaeus appears to have derived his information concerning the lands on the shores of the Ocean, to the west and north of Europe, chiefly from Pytheas: he followed that writer in regard to the land from whence amber was brought, though he called the island Basilia, which was named by Pytheas Abalus.8 But‘ the most curious notice that is preserved to 5 Id. xii. 28 a. If, however, these | dinia is in fact, according to the com- researches were in any degree the result : putation of Captain Smyth, slightly of personal observation, they must have larger than Sicily, while Majorca is not been made at an early period of his only entitled to rank next after Lesbos, life, as he himself stated in his history but is considerably larger than that that he had resided at Athens nearly island, though not usually comprised fifty years without once quitting it by ancient writers among the seven (Ibid. 25 d). chief islands of the Mediterranean 6 His views on two points, however, (see note on Scylax). In this respect, which are regarded as erroneous by therefore, Timaeus was certainly better Strabo (xiv. 2, p. 654)—that Sardinia informed than most of the Greeks. was larger than Sicily, and that the 7 Timaeus, Fr. 36. Plutarch (1e chief of the two Gymnesian or Balearic Placz't. Philosoph. p. 901. Islands (Majorca) was the largest 8 Plin. H. N. xxxvii. 11, § 36. In island in the Mediterranean after another passage, however (iv. 27, § 94), Lesbos—were far from meriting the Pliny cites Timaeus as calling the reprobation of the geographer. Sar- island Rauronia, and placing it in the SECT. 2. THEOPHRASTUS. 603 us from this part of his work is the statement that tin was brought from an island named Mictis, which was distant six days’ voyage from Britain, to which the Britons navigated in vessels of wicker-work covered with leather.9 There can be little doubt that we have here the first mention of the Cornish tin-trade, in connection with Britain, and that the island of Mictis is the same as that called by Diodorus Ictis, which he describes as the chief emporium to which the British tin was brought for exportation.1 § 10. Independently of any professed geographical treatises, it is evident that at this period the Greeks had acquired a general familiarity with the countries of Western Europe, Gaul, Spain, and Liguria, as well as with Italy and the Italian islands, wholly different from that which they possessed a century before. This is sufficiently attested by the incidental notices of the productions of those countries and of natural phenomena observed there, which we find in THEOPHRASTUS, who wrote about the commencement of the third century 13.0., and still more in the treatise “ On YVonders,” ascribed to Aristotle, but really belonging to the middle of the same century. Thus we find in Theophrastus mention of several plants and trees as flourishing in Tyrrhenia, Latium, &c.: a special notice of the extensive forests of Corsica, the pine- trees in which were said to exceed all others in size : 2 a curious and accurate description of the promontory of Circaeum or Monte Circello;3 an account of certain kinds of sea-weed Ocean to the north of Scythia, at a circumsutis navigare” (Plin. H. N. iv. day’s voyage from the mainland. There 16, § 104). Here it is impossible to can belittle doubt that both statements say what sense we are to attach to the refer to the same story, but whether the word “ introrsus,” upon which the in.. error rests with Timzeus or with Pliny terpretation of the whole passage, in a we are unable to decide. geographical sense, depends. But there 9 Unfortunately this notice is trans- seems little doubt that we have here mitted to us by Pliny in such a form the first obscure intimation of the story as to be almost unintelligible. His which we find developed in a more words are: “Timaeus historicus a complete form in Diodorus. Britannia introrsns sex dierum navi- 1 Diodor. v. 22. gationc abesse dicit insulam Mictim, in 2 Hist. Plant. v. 8, §§ 1, 2. qua candidum plumbum proveniat. Ad 3 Ibid. § 3. cam Britannos vitilibus navigiis corio ‘. 604 HISTORY OF ANCIENT GEOGRAPHY. OHAP. XV. which grew in the Ocean beyond the Pillars of Hercules :4 and the fact that native cinnabar was found in the interior of Spain.5 The notices preserved by the unknown author of the treatise “ De Mirabilibus ” are unfortunately mixed up, as might natu- rally be expected from the character of the work, with much of fable; and its date is in some degree a matter of uncertainty, but it contains some notices of interest. Among these may be mentioned the statement that the Ister rose in the Her- cynian forest—a name which here appears for the first time.6 At the same time the author repeats the popular notion that one arm of that river flowed into the Adriatic and the other into the Euxine. He placed also the 'Electrides or Amber Islands, in the inmost bight of the Adriatic, but connected them with the Eridanus, which he undoubtedly supposed to be the Po." He is also the first extant author who mentions the iron mines of ZEthalia (Elba), though these had doubtless been worked by the Tyrrhenians long before ; 8 and has a very curious notice of certain vaulted buildings or Tholi, in the island of Sardinia, which probably refers to the singular edifices called N uraghe still found in that island.9 It is remarkable also that the name of the Rhenus or Rhine had reached his ears, and was mentioned by him in conjunc- tion with the Ister, as one of the two great rivers of the north. Both of them, he adds, were navigable in summer, but in winter were frozen hard, so that you might ride across them. Equally interesting is it that he describes the Rhine as flowing by the land of the Germans (I‘ep/tavobs‘), as the Danube did by that of the Paaonians.1 This is the first mention that is found in any ancient author of the name of Germans? 4 Ibid. iv. 6, § 4; 7, § 1. the Pannonians of later writers. The 5 De Lapid. c. 8, § 58. 6 Mirab. Auscult. § 105. Aristotle himself (Meteorologica, i. 13, § 19) describes the Ister as rising in the Pyrenees. 7 Ibid. § 81. 8 Ibid. § 93. 9 Ibid. § 100. 1 Ibid. §168. The Paaonians here meant are clearly the same people with confusion between the two names is found in Greek authors of much later date. 2 It is remarkable that this passage should have been overlooked by Ukert in his elaborate investigation of the origin and introduction of the name (Geographic, vol. iii. pt. 1, p. 71). SECT. 2. TREATISE “ ON woNnERs.” 605 § 11. We are indebted to this compiler also for two inter- esting notices concerning the little known regions beyond the Pillars of Hercules. The one relates that the Phoenicians who dwelt at Gadeira, having sailed for four days with an east wind, discovered some banks, dry at low water and covered with quantities of sea-weed, where they found vast quantities of tunny fish of such superior quality that, when salted or pickled, they were carried to Carthage and retained by the Carthaginians for their own exclusive use.3 The other refers to the discovery in the sea beyond the Pillars, at a distance of several days’ voyage, of an island of considerable extent, uninhabited, but abounding in timber of all kinds, possessing navigable rivers, and admirably fertile in all kinds of fruits. It was repeatedly visited by the Carthaginians, and partially colonized, but subsequently abandoned by order of the govern- ment, and all intercourse with it prohibited, for fear that it should attract too great a number of colonists to the detriment of the mother country.4 This account agrees so closely with that of Diodorus5 as to leave no doubt that they are both taken from Timeeus. It is worthy of notice that in this, the earliest notice that is preserved to us of the Fortunate Islands of the West—which we shall find reappear in so many forms—— one island only is spoken of, and that is described in a manner that leads us at once to identify it with Madeira.6 § 12. Various causes must have no doubt contributed at this period to awaken increased interest and curiosity con- cerning the nations of Western Europe. The sudden inroads of the Gauls, who had not only overrun the northern provinces 3 Ibid. § 136. 4 Ibid. § 84. 5 Lib. v. c. 19, 20. Diodorus, how- ever, represents the island as inhabited and even abounding in splendid build- ings! His whole account is indeed much more highly coloured and exag- gerated than that of our author. 6 The abundance of wood, presenting so striking a contrast to eyes accus- tomed to the barren shores of Spain and Africa, is characteristic of Madeira, the name of which is derived from this circumstance, madera being Portuguese for timber. The existence of navigable rivers is of course an exaggeration, but with this exception there is nothing in the account given by our author that is not probable enough. 606 HISTORY or ANCIENT GEOGRAPHY. can». xv. bordering upon Thrace and Macedonia, but had for a time overthrown the Macedonian monarchy itself, and then crossing over into Asia rendered themselves equally formidable to the Greek rulers of Asia Minor, would naturally direct the attention of the Greek writers to the original abodes, as well as to the national characteristics, of these formidable invaders. At the same time the extension of the Carthaginian power in Spain could not fail to lead to a more complete and accurate knowledge of that country, at all events of its eastern portions, and though this may have been but imperfectly transmitted to the Greeks, there are not wanting indications of increased freedom of intercourse among all the commercial nations of the Mediterranean. In addition to Alexandria, Rhodes had assumed a prominent place among the trading communities of Greece, and continued for a considerable period to enjoy the ‘highest wealth and prosperity.’ Byzantium also, from its position, commanded all the commerce of the Euxine. Syra- cuse was at the height of its prosperity, and under the mild and beneficent rule of Hieron enjoyed a long period of unwonted tranquillity, during which it rose to the greatest opulence. Carthage, though declining in power after the First Punic War, still retained a very extensive commerce, and appears to have held more communication with the Greeks than at an earlier period, though the latter were still jealously excluded from the trade beyond the Pillars of Hercules, which continued to be centred exclusively at Gades. 7 Diodor. xx. 81 ; Polyb. iv. 47. In by the complaints of the other trading B.0. 219 the Rhodians were compelled cities, 51d. '1'?) EOKGTV airroils 7rpoeo’roiuat to engage in war with the Byzantines, 'ré‘w Ka'ro‘r Octavia-Tau. Polyb. Z. c. NOTES A, B. THE PTOLEMIES IN EGYPT. 607 NOTE A, p. 578. MYOS HORMUS. THE position of Myos Hormus has been the subject of considerable doubt. It was placed by l/Vilkinson and YVellsted at Abu Schar, in latitude 27° 24’, nearly opposite the angle of the Sinai'tic penin- sula, where there are considerable remains of an ancient town. But Dr. C. Miiller has shown that it was in all probability situated at Ras Abu Somer, about half a degree farther to the south, where there is a good port with three small islands, and in the immediate neighbourhood a remarkable mountain, conspicuous by its red colour, thus answering precisely to the Scarlet Mountain (é’pos ,ailvri’bdes) placed by Agatharchides in the neighbourhood of Myos Hormus. (Agatharchid. § 81, ed. Muller.) In this case the ruins visible at Abu Schar probably mark the site of Philotera, which is placed by Artemidorus to the north of Myos Hormus, though Ptolemy places it farther south, having apparently transposed the two positions. (See C. Muller’s elaborate note on Agatharchides, la.) The same view is adopted by M. de St. Martin (Le Nord de l’Afrigae, pp. 255-258). ' NOTE B, p. 578. AFRICAN ELEPHANTS. It is well known that at the present day the African elephant is generally reputed to be untameable, and nothing seems to have more excited the wonder of the Abyssinians, during the late war, than to see the manner in which the Indian elephants that accom- panied the English army were trained to perfect obedience. Yet it is certain that not only were the Ptolemies able to train the elephants of Ethiopia for purposes both of war and parade, but their example was quickly followed by the Carthaginians, who employed elephants in Sicily as early as the First Punic War, before the death of Ptolemy Philadelphus. All the elephants used 608 HISTORY or ANCIENT GEOGRAPHY. ‘can. xv. by them, including those carried by Hannibal into Italy, were un- doubtedly of African origin. The elephants also which are repre- sented on Roman coins and monuments belong in all cases to the African and not the Indian variety: the very large size of the ears constituting a criterion by which they are easily recognized. The inscription on the monument of Adulis, erected in the reign of Ptolemy Euergetes, distinctly ascribes to his father Philadelphus the merit of being the first to procure Ethiopian elephants and train them to service in war. A large number of these accom- panied the army of Euergetes when he carried his arms into Asia and overran the dominions of Seleucus II. (Mon. Adulit. ap. Clinton, F. H. vol. iii. p. 382.) The statement of Agatharchides that Ptolemy Philadelphus was the first to train elephants for war, which justly excited the surprise of Photius, was doubtless intended to apply only to African elephants, and as thus understood, was strictly correct. (Agatharchides ap. Phot. p. 7 17 NOTE O, p. 579. LAND OF GINNAMON. There is no doubt not only that the Regio Oinnamomifera (i7 Kwvanwnéqtopos) of the ancients was the tract of Northern Africa extending to Cape Guardafui, but that it was from thence that both the Greeks and Romans actually derived their principal, if not their sole, supply of that valuable spice. The geographical term is first found in Eratosthenes (ap. Strab. ii. pp. 63, 72, &0.) who appears to have applied it to the whole coast extending from near the Straits of Bab-el-Mandeb to Cape Guardafui, and at a later period we find the Periplus of the Erythraean Sea enume- rating cassia—a term which was usually applied to cinnamon—— among the productions exported from all the ports along this line of coast, from Malao (Berbera) to Cape Guardafui, or the Promon- tory of Spices (Aromatum Promontorium), as it was called by all later writers, evidently from this very circumstance. (Periplus Mar. Erythr. §§ 8-12.) But while no difficulty arises on the geographical point, there has been much question raised as to the ancients having really derived their cinnamon from thence. In modern times cinnamon is almost exclusively procured from Ceylon, or from regions still NOTE D. PROGRESS IN WESTERN EUROPE. 609 farther to the east and still more unknown to the ancient traders, China and Java. Nor is it now known to exist in the part of Africa from whence the Greeks and Romans procured it: though that region still abounds in myrrh and frankincense. Hence it has been supposed by some modern writers that it was only brought by sea to the ports in the neighbourhood of Cape Guardafui, and thence imported by the Arabian and Greek merchants, in the same manner as the Indian merchandise was from Aden and other ports on the coast of Sabaea. The testimony of the ancient writers however is too distinct and precise to be thus set aside: and Mr. Cooley has moreover shown that its cultivation in Ceylon dates from a comparatively recent period. The same view is adopted and confirmed by Sir E. Tennent, who has investigated the subject with much care. It is certain that no ancient writer alludes to cinnamon being brought from thence or from the adjoining coasts of India, even after the direct trade was opened with those countries. It seems therefore impossible to doubt that the cinnamon used by the Greeks and Romans—which was probably of an inferior quality to that of Ceylon—was really brought from the north-east corner of Africa, the land of the Somali, a tract still very imperfectly known, and where it is not improbable that the cinnamon may still be found Wild. (See Mr. Cooley’s paper on the Regz'o Cimzamomifera of the Ancients, in the Journal of Geogr. Soc. vol. xix. and Sir E. Tennent’s Ceylon, vol. i. p. 599—604.*) NOTE D, p. 586. MONUMENT OF ADULIS. It is now well known that the celebrated Monument of Adulis, for our knowledge of which we are indebted to a monk of the sixth century, Cosmas Indicopleustes, consisted in fact of two separate portions, the inscriptions on which were copied by the traveller, as if they had formed part of the same record. The former part belonged to a stela or pyramid set up in the reign of Ptolemy Euergetes, and recounting the exploits of that monarch, * The same view had been already urged by Dr. Vincent, vol. ii. p. 512, VOL. I. 2 R 610 HIs'roRY oE ANCIENT GEOGRAPHY. CIIAP. xv. especially his invasion of the Syrian monarchy, in which he pene- trated beyond the Euphrates, and conquered or at least reduced to submission the provinces of Babylonia, Susiana, Persis, Media, and the rest of Asia as far as Bactriana ; but containing no allusion to any conquests in Ethiopia, beyond the incidental notice already alluded to, of his having made use of Ethiopian elephants. The second part, which is of much later date, recounts the exploits of an Ethiopian king of Axum in Abyssinia, who appears to have conquered all the neighbouring tribes in that part of Africa, and even carried his arms across the Red Sea into Arabia. The con- fusion resulting from the union of the two, which had misled even Dr. Vincent and Ukert, was first cleared up by the English tra- veller Mr. Salt. (A full account of the monument and its inscrip- tions will be found in Boeckh’s Corpus Inscript. Groecarum, tom. iii. p. 508 and foll.) See also St. Martin, Le Nord de l’Afr'igue, p. 224. Clinton H. vol. iii. p. 382 note) has given the first part of the inscription relating to Ptolemy Euergetes, but has erroneously added as referring to the same monarch the conclusion which belongs to the Axumite king. NOTE E, p. 589. THE WINDS AS KNOWN TO THE GREEKS. The number of divisions thus established by Timosthenes was twelve in all : and this appears to' have been the number generally recognized among the later Greek geographers. No trace is found of a subdivision into sixteen parts, according to the custom of modern ‘navigators and geographers. But the statement of Aga- themerus that Timosthenes distinguished twelve winds, by inserting four additional ones between those previously known and admitted, is certainly not correct. Aristotle, in his Meteorologica (ii. 6), distinctly enumerates twelve winds, and the quarters from which they blow; and though his list differs slightly from that ascribed to Timosthenes, this arises only from the variations of names, many of which were of local attribution. His enumeration is as follows. The west wind, Zephyrus, blew from the equinoetial setting: and opposite to this was the east wind, Apeliotes, from the equinoetial rising. The north Wind, called Boreas and Aparctias, blew from the north, the region of the Great Bear (Arctos). Opposite to this NOTE E. PROGRESS IN WESTERN EUROPE. 611 was the south wind, Notus. The Caecias blew from the summer rising (north-east), and opposite to this was the Lips, from the winter setting. The Eurus blew from the winter rising (south- east), and opposite to this was the Argestes (north-west) known also as Olympias and Sciron. These therefore formed four pairs respectively opposed to each other; but besides these there were others which were not so precisely opposite: these were the Thrascias, intermediate between the Argestes and the Boreas : the Meses, intermediate between the Boreas and Caecias; the Phoenix, between the Eurus and Notus; the Libonotus in the corresponding position, between the Lips and Notus, is not mentioned, the name being apparently not familiar to the Greeks in his day. But though this amount of subdivision was recognized by the more scientific writers, there can be no doubt that eight winds only were popularly known. This is the number found on the monument at Athens, commonly called the Tower of the Winds, where their names are thus given: Boreas, Caecias, Apeliotes, Eurus, Notus, Lips, Zephyrus, and Sciron.8 It is evident however that the ancients, even the geographers, made no attempt to divide the circle of the heavens into regular portions corresponding with our quarters (N.E., S.W., &c.), inde- pendent of the winds. The only mode in use to designate these points of the compass (as we call them) was by reference to the summer and winter changes in the place of the rising and setting of the sun: as is done by Aristotle in the passage just cited, and by Strabo throughout his work. Of course such a mode of expres- sion was inaccurate, and had the further disadvantage that it ought in strictness to vary with the latitude of each place : as for instance between Athens and Alexandria. But no such accuracy was ob- served in practice: and the expressions of “the winter sunrise ” or “ the winter sunset” would be used generally as equivalent to south-east and south-west respectively. At the same time most scientific observers would be aware that there was a greater in- terval between these points and the four cardinal points, on the one side than the other; and hence arose the intercalation both by Timosthenes and Aristotle of four such points, without the four corresponding ones introduced in our modern division. 8 This monument, more correctly termed the Horologium of Andronicns Cyrrhcstes, belongs probably to the second century, no. 2 R 2 612 HISTORY or ANCIENT GEOGRAPHY. CHAP. xv. It will be‘ seen that where the whole circle is thus divided into twelve regions instead of sixteen, none of the points, except the four cardinal ones, can exactly correspond with those of our modern divisions. Hence the use of the terms North-north-East, South- south-West, &c., in the translation of the ‘passage from Timosthenes is necessarily inaccurate, and has been adopted merely for the sake of convenience. NOTE F, p. 591. DIMENSIONS OF BRITAIN. It is worthy of remark that Diodorus, Writing after Caesar had given so much more information concerning Britain, and an ap- proximate estimate of its true dimensions, should return nearly to the statement of Pytheas, though with an affectation of accuracy, which was evidently founded on no real authority. He makes the whole circumference 42,500 stadia, the three sides being respec- tively 7500, 15,000 and 20,000 in extent. (Diodor. v. 21.) But he correctly calls the side opposite to Gaul, the shortest side of the triangle: while Strabo, who had a wholly erroneous idea of the position and extent of Britain, makes its side opposite to Gaul, which does not exceed (he says) 4400 stadia, the greatest length of the island. (Strabo, iv. 5, § 1, p. 199.) It is not improbable that in this part of his work Diodorus followed Timaeus. NOTE G, p. 598. SIR G. LEWIS ON OREDIBILITY OF PYTHEAS. This has been especially the case with Sir G. Lewis, who in his dissertation on the Navigation of the Phoenicians (inserted in his Historical Survey of Ancient Astronomy, chap. viii.) has treated Pytheas with the same contempt as Polybius did, without adverting to the points in which our present full knowledge of the northern regions of Europe has shown that Pytheas was right, and Polybius and Strabo were wrong. That Pytheas did not really visit Thule may be readily admitted, and it is improbable that he really advanced along the northern shores of Germany farther than the Elbe. Even NOTE H. rnoennss IN WESTERN EUROPE. 613 at a much later period we find the Roman fleet that first penetrated to the mouth of that river claiming to have reached the confines of the known world (Mon. Ancyran. p. 34.) But this is no reason for disbelieving altogether the fact of his having personally explored a considerable portion of the countries to the north of Europe that were previously unknown to the Greeks. Besides the chapter of Sir G. Lewis just referred to, the credi- bility of the voyage of Pytheas and of his geographical information is fully discussed by Gossellin in his Recherches sar la Geographic des Anciens (vol. iv. p. 168—179), by Ukert (Geographic der Griechen and Homer, vol. i. pt. 2. p. 298-309; vol. iii. pt. 1. p. 5, § 6): and in the article PYTHEAs in Dr. Smith’s Dictionary. References are given by Ukert to the earlier authorities on the subject. See also Redslob’s Thule, cited in note to p. 597. NOTE H, p. 599. ASTRONOMICAL PHENOMENA AT THULE. This is the distinct statement of Pliny in one passage (iv. 16, § 104), “ Ultima omnium quee memorantur [insularum] Thule, in qua solstz'tz'o nallas esse aoctes indicavimus, cancri signum sole transeunte, nullosque contra per brumam dies.” But he adds: “ Hoe quidam senis mensibus continuis fieri arbitrantur” : and in the previous passage to which he here refers (ii. 75, § 187) he him- self cites Pytheas as having stated that this—the continuous day ' for six months, and continuous night for the other six—was what actually occurred at Thule. It is much more probable that Pytheas had made the more correct and rational statement, and had been misunderstood by those who did not see the absurdity of the other supposition. In another passage Strabo refers to him as having placed Thule, which he calls “the most northern of the British Islands,” in a latitude where the arctic circle coincided with that of the summer tropic—a distinct astronomical statement which would be equiva- lent, in the language of modern geographers, to saying that it was situated under the Arctic Circle. (Strabo, ii. 5, § 8, p. 114.) This is of course incorrect, if we suppose Thule to have any connection with the British Islands, so as to represent the Orkneys or Shet- lands (even the latter group extending only to 60° 50’ N. lat.), but 614 HISTORY OF ANCIENT GEoGRAPHY. oHAP. xv.- by no means so gross an exaggeration as it was supposed to be by Strabo and others, who were ignorant of the true position of these northern countries. It is one also to which Pytheas would have been easily led, if we suppose him to have picked up his infor- 'mation concerning Thule at second hand : the absence of any appa- rent night in these high northern latitudes being readily transformed into" the assertion that the actual, or astronomical, day was of twenty-four hours’ duration. Pytheas was doubtless astronomer enough to discern that this must be the case somewhere, as one advanced towards the north, and he might therefore readily accept the hearsay statement that it was the fact in the island to which he gave the name of Thule. _ It may be observed that at a much later period we find Caesar himself repeating the tale that there was continuous night for thirty days at the winter solstice, but which had been transferred to the islands between Britain and Ireland. (Caes. B. G. v. 13.) (615) CHAPTER XVI. ERATOSTHENES. § 1. WE are now arrived at the period when geography first began to assume something offla, regular and systematic cha- E953; and “to, be based, however imperfectly,_up_onn fixed scientific principles. It is to the Alexandrian school that we are indebted for the first steps in this direction: and ERATos- THENES, who presided over that school during the space of more than forty years,1 may be regarded as the parent of scien- tific geography, as he was also in great measure of systematic chronology. The way had been undoubtedly prepared for him by the astronomical researches of his immediate prede- cessors, and he was himself fully acquainted with all the astronomical science of his time: while the recent extension of geographical knowledge among the Greeks from the various causes which we have been just passing in review, had accumu- lated a mass of information greatly exceeding that at the disposal of earlier writers. The position of Eratosthenes himself was peculiarly favour- able. He was born at Cyrene in 13.0. 276, and having early devoted himself to the study of philosophy and learning, passed a considerable time at Athens, from whence he was invited to Alexandria by Ptolemy Euergetes, and placed at the head of the Library, a position of the highest literary distinction, which gave him the command of the accumulated stores of learning that had been brought together by the enlightened monarchs of Egypt during three generations. Eratosthenes continued to 1 According to Clinton (F. H. vol. iii. continued to hold this situation till his p. 37), he succeeded Zenodotus as libra— death in 13.0. 196. rian at Alexandria, about 15.0. 240, and 616 HISTORY or ANGIENT GEOGRAPHY. CHAP. xvI. hold this important post till his death, about B.0. 196/; and left behind him a number of works, some scientific, others purely literary. / A_l_l__t_l_1e~s~e phaye unfortunately perished ,;_ and though enough has been preservedloflhflis" geographical treatise to enable us in great measure to judge of its character, and to form a tolerably clear idea of his system, yet we shall find, as we pursue our examination, abundant reason to regret the imperfect character of our materials? In another respect also we are unable to estimate justly the amount of merit due to Eratosthenes, from our very deficient knowledge of the actual state of geographical science previous to the commence- ment of his labours. § 2. It is certain indeed that Eratosthenes was not the first , among the writers of this period who attempted to bring toge- the; in one general view the results of recent discoveries and observations, and survey the progress that had been madein the knowledge of the different countries of the world. Die/rs; ARCHUS, a pupil of Aristotle and a friend of Theophrastus, who flourished about a hundred years before Eratosthenes, B.0. 326—296,3 had left several geographical works, one of which, termed I‘iiq wept’odoq, was apparently, from its title, a general geographical treatise: and would seem to have com- prehended a summary view of the position and dimensions of the habitable world, as then known, not altogether dissimilar from that given by Eratosthenes. But the few citations that are preserved to us are not sufficient to enable us to judge fairly of the relation which it bore to the work of the later author. We learn however from Strabo,4 that Polybius, while he passed by the earlier writers on geography as unworthy of serious consideration, entered into an elaborate criticism of - Dicaearchus and Eratosthenes as representing an improved ‘'_~\ 2 The fragments of the geographical works, by Bernhardy (Eratosthenica, work of Eratosthenes were first 001- l 8vo, Berolin. 1822). lected and edited by Seidel (Eratos- 3 Clinton, F. H. vol. iii. p. 474; G. thenis Geographicorum Fragmenta, 8vo, Miiller, Fragm. Hist. Gr. vol. pp. Goettingae, 1789); and more fully, 225,226. together with the remains of his other 4 ii. p. 104. crap. xVI. ERATOSTHENES. 617 state of geographical knowledge. We may therefore infer that the former also had treated the subject in something like a scientific manner. Dicmarchus had also written (besides many valuable works of a historical character) a complete geographical, or rather topographical, description of Greece—a treatise which would have been of the highest interest to us had it been preserved ; and we learn incidentally that this was accompanied by maps, which were still extant in the days of Cicero, and were regarded by him as of high authority.5 Three fragments of considerable length belonging to a work of this description have been pre- served to us, and are published in all the editions of the minor Greek geographers, to which the name of Dicaearchus has been usually applied, but there is no authority for this attribution, though they probably belong to about the period of that author.6 § 3. On another account also Diceearchus deserves a place in reviewingthg progress of__geographicalknowledge, that he was thezfirst, so far as we know, to attempta scientific measurement of mountains ; a task which he undertook, we are Itdldl,lthe request of certain kings,7 probably those of Mace- donia. It was not likely that the results of such a first attempt should be satisfactory: but the few and scanty notices of them that have been preserved to us are so uncertain and incon- sistent with one another, that we cannot safely judge of their character. Thus we are told on the one hand that he deter- mined the height of Cyllene in Arcadia to be rather less than 15 stadia, or 9000 Greek feet, and that of Atabyrius in the 5 “ Peloponnesias civitates omnes Miiller in his edition of the Fragmenta maritimas esse hominis, non nequam, Historicorum Grazoorum, vol. ii. pp, 229 sed etiam tuo judicio probati Dicaearchi -232. The fragments themselves are tabulis Credidi” Cic. Epist. ad Att. published by the same author in his vi. 2. Osann considers these tabulav to Geographz' Greece‘ Minores, tom. i. pp. have been attached to the I‘fis wepiooos, 97-110. They are contained also in while C. Miiller doubts their having the second volume of Hudson’s edition any reference to maps at all. The pas- ' of the same writers. sage is certainly not conclusive. 1, Begum “aura inontes pormensus, 6 See this point fully discussed by C. ‘ Plin. H. N. 11. ()5, § 162. 6I8 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XVI. island of Rhodes (a mountain of very inferior altitude) to be 14 stadia.8 On the other hand Pliny asserts that he ascer- tained Pelion to be the highest of the Greek mountains, but that it did not exceed 1250 paces (6250 feet) in perpendicular height.9 If he really came to the conclusion that Pelion ex- ceeded in elevation the neighbouring mountains, Ossa and Olympus, his method of observation must have been singularly inaccurate; the former being in fact at least 1000 feet higher than Pelion, while Olympus exceeds it by 4700 feet, little less than half its total altitude]l Olympus is indeed by far the highest mountain in Greece, while Pelion is inferior to Par- nassus, Cyllene, Taygetus, and several others. But Dicaearchus at least deserves credit for having attempted to determine the perpendicular altitude by a mathematical process. § 4. There can be no doubt that before the time of Eratos- thenes the ideas of the learned world upon the subject of geo- graphy had assumed a more regular and systematic form. And it is certain also that these had been embodied in the form of maps, which, however imperfect, were unquestionably very superior to anything that had preceded them. We have seen that the first use of maps had been introduced at a very early period by Anaximander. and that maps of the world were not uncommon in the time of Herodotus, though based on the crude ideas of the period, and on hasty assumptions that excited the ridicule of the historian. Nor can it be doubted that the discoveries resulting from the conquests of Alexander, and the extension of geographical knowledge under his suc- 8 Geminus, Elem. Astronom. c. 14. The actual height of Cyllene is 7788 feet, that of Atabyrius only 4560. The former was generally regarded as the highest mountain in the Peloponnesus : according to Strabo it was reckoned by some to be 20 stadia in perpendicular height; by others (probably referring to Dicaearchus) only fifteen. But another statement, cited from Apollo- dorus, made its height 80 feet less than nine stadia, or only 5320 Greek feet (Steph. Byz. s. v. KvAMiW); Eustath. ad. Hem. Odyss. p. 1951). 9 Plin. H. N. ii. 0. 65,§ 162.. “ Dicaa- archus, vir in primis eruditus, regum cura permensus montes, ex quibus altissimum prodidit Pelion MooL passuum ratione perpendiculari.” ‘ Admiral Smyth gives the height of Pelion at 5200 feet, Ossa at 6100, and Olympus at 9850 feet above the sea. can». xvi. ERATOSTHENES. 6 19 cessors, would have gradually found their way into such maps; but we know from frequent experience, even in modern times, how slowly established errors are discarded, and how long they maintain their ground, even in the face of more accurate information. The same thing was still more the case in ancient times, and it is highly probable that if we could now recover the map of the world as it was generally received in the time of the first Ptolemies, we should find it still retaining many of the erroneous views of Herodotus and Hecataeus. It appears indeed from repeated statements of Strabo that Eratosthenes made it the object of his special attention to “reform the map of the world,” as it had existed down to hisv time,2 and to reconstruct it upon more scientific principles. It is this enlarged and philosophical view of the subject which constitutes his especial merit, and entitles him to be justly called the father of systematic geography. The materials at his command were still very imperfect, and the means of scien- 1 tific observation were wanting to a degree which we can, at the present day, scarcely figure to ourselves ; but the methods which he pursued were of a strictly scientific character, and his judgment was so sound that he proved in many instances to be better informed and more judicious in his inferences than geographers of two centuries later. § 5. In regard to the fundamental idea of all geography— the position and. figu1;e/of___the_earth—Eratosthenes adopted the views that were current among the astronomers of his day, which had been received almost without exception from the times of Aristotle and Euclid.3 He regarded the earth as a sphere,4 placed in the centre of the universe, around which the views will be found in Sir G. Lewis’s Historical Survey of Ancient Astronomy (pp. 187, 188), extracted from the Pheenomena of Euclid. The great 2 Aroptiéio'ar 'rbv dpxa'iov 'yew'ypacpucbv Will/(Zita was the problem that he had set before himself. Strab. ii. c. 1, § 2. These words point clearly to the pre- vious existence of a generally recog- geometer had preceded Eratosthenes nizcd map of the inhabited world I by nearly a century. ('rfis oZnovnéz/ns), probably that of l ‘1 Strabo repeatedly censures Eratos- Dieecarchus. l thencs (i. pp. 62, 65) for dwelling at 3 A convenient summary of these : unnecessary length upon the proofs of 620 HISTORY OF ANCIENT GEOGRAPHY. GHAP. XVI. celestial sphere revolved every twenty-four hours : besides which, the sun and moon had independent motions of their own. The obliquity of the sun’s course to that of the celestial sphere, was of course well known: and hence the great circles of the equinoctial, and the ecliptic, or zodiacal circle, as well as the lesser circles, called the tropics, parallel with the equi- noctial, were already familiar to the astronomers of Alexandria. Moreover it appears that these conceptions, originally applied ‘to the celestial sphere, had been already transferred in theory to the terrestrial globe. Thus the idea of the globe of the lhearth, as it would present itself to the mind of Eratosthenes, ]or any of his more instructed contemporaries, did not differ ,materially from that of the modern geographer. For all geo- graphical purposes, at least as the term was understood in his day, the difference between the geocentric and the heliocentric (‘theories of the universe would be unimportant. § 6. But Eratosthenes had the merit of making one valuable addition to the previously existing ideas upon this subject, by a more careful and successful measurement than had ever been previously attempted, of the magnitude of the earth, or circumference of the terrestrial globe. He was not indeed the first who had attempted the solution of this problem, which would naturally engage the attention of astronomers and geometers, as soon as it was agreed that the earth was of a spherical form. Aristotle refers to the calculation of “mathematicians” who had investigated the subject (without naming them) that the earth was 400,000 stadia in circum- ference.5 At a later period Arpliimedes speaks of 300,000 the spherical character of the earth; a fact which he appears to have regarded as too well known to require demon- stration. But though it was undoubt- edly a received tenet among philosophers in the days of Eratosthenes, it had pro— bably not yet acquired the same general acceptance as in the time of Strabo, two centuries afterwards. 5‘ Dc Coclo, ii. 14, § 16. It is a singular instance of that blind rever- ence for antiquity which has misled so many modern writers, that this passing notice of Aristotle, on which he himself evidently laid no stress, and had taken up (as he himself tells us) on the authority of others, should have been received as unquestionably correct. Hence D’Anville and Gossellin, in order to explain its apparent inaccuracy, 01m. xVI. ERATOSTHENES. 62 I stadia as the measurement usually received, a statement apparently founded on the calculations of Aristarchus of Samos, one of the earlier astronomers of the Alexandrian school.6 But we have no information as to the data on which these first -crude attempts were based, or the mode by which the authors @Ilived at their results. The _method pursued by Eratosthenes was theoretically sound, and was in fact identical in principle with that which has been adopted by astronomers in modern days. Assuming (in accordance with the general belief) that Syene in Upper Egypt was situated exactly under the Tropic of Cancer: assuming also that Alexandria and Syene were on the same meridian, and at the distance of 5000 stadia from one another, he measured the shadow of the gnomon at Alexandria in order to determine its latitude, and thus ascertained that the arc of the meridian intercepted between the two, was equal to one-fiftieth part of a great circle of the sphere. Hence he at once deduced the conclusion that such a great circle, or the circumference of the globe, would amount to 250,000 stadia.7 § 7. The only theoretical error in this mode of calculation was in the assumption—which was inevitable in the days of Eratosthenes—that the earth was exactly spherical, instead of being as it really is, a slightly oblate spheroid, and that therefore a meridian great circle was equal to that of the equator. And the error proceeding from this cause, which would not exceed --1-flth part of the whole, is wholly unimportant 300 have assumed that Aristotle must have ] accurate investigations of the same been employing a smaller stade than problem. that generally known to the Greeks, l 6 Archimed. Arenarius, p. 320, ed. such as would really give 400,000 stades Torelli. 7 The method piu'sued by Eratos- thenes is fully stated and explained by the astronomer Cleomedes, in his work on the Circular Motion of the Heavenly Bodies (KvKAuci') Oewpi'a Me'reu'apwu, i. c. 10, ed. Bake), and will be found in Bernhardy’s Eratosthenica, Fr. 42. The date of Cleomedes is uncertain, but he may probably be assigned to the first century before Christ. for the circumference of the earth, and I have then proceeded to make use of I this smaller or Aristotelian stade for ‘ the measurement of distances in the ! marches of Alexander and the voyage ' of Nearchus (see Chapter XIII. p. 545). i Yet it is evident that no value was at- , tached to this statement by any ancient ‘ writer from the time when Eratosthenes , and Hipparchus had instituted more ,\ 622 HISTORY OF ANCIENT GEOGRAPHY. GHAP. XVI. as compared with the practical errors arising from the defective means of observation. In the first place it was assumed that Syene lay directly under the tropic, it being a well-known fact that at the summer solstice the sun could be seen from the bottom of a deep well, and that at the same time the gnomon cast no perceptible "shadows But though these facts were perfectly correct as matters of rough observation, such as could be made by general travellers, they were far from having the precise accuracy requisite as the basis of scientific calculations. Syene is in (fact situated in latitude 24° 5' 30",9 or nearly 37 G. miles to the north of the tropic. In the next place Alexandria, instead of being exactly on the same meridian with Syene, lay in fact not less than three degrees of longitude to the west of it: an _error of no trifling moment when the distance between the two was assumed as the basis of calculation. But a much graver error than either of these two was that caused by the erroneous estimate of the actual distance between the two cities. What mode of measurement had been resorted to, or how Eratos- thenes arrived at his conclusion upon this point, we are wholly without information: but it may well be doubted whether he had recourse to anything like actual mensurati'on.1 Indeed the difficulty which modern experience has shown to attend this apparently simple operation, where scientific accuracy is required, renders it highly improbable that it was even attempted; and the round number of 5000 stades at once points to its being no more than a rough approximation. But even considered as such, it exceeds the truth to a degree that 3 Strabo, xvii. p. 317. The same 1 We are told indeed by Martianus thing is told by Seneca, Pliny, and Capella (cited by Bernhardy) that Lucan. It is remarkable that no men- this distance was derived from actual tion of it occurs in Herodotus, though measurement, but the authority of so the fact must have been well known to late a compiler is altogether worthless; the Egyptian priests, and was one and there is no doubt that the Greeks likely to attract the attention of all in the time of Eratosthenes had no travellers. \\ power of making a trigonometrical sur- 9 Wilkinson’s Egypt and Thebes, vey, without which any such measure- p. 415. ment is impossible. can». XVI. ERATOSTHENES. 623 one could hardly have expected, in a country so well known as Egypt, and in an age so civilized as that of the Ptolemies. Alexandria is in fact situated at a distance of about_5m30 geo- graphical miles (5300 stadia) from Syene, as measured on the map along the nearest road: but the direct distance between the two, or the arc of the great circle intercepted between the two points, which is what Eratosthenes intended to measure, amounts to only___453 G. miles or 4530 stadia.2 “Eratosthenes “Lil-174976- ill fixingii the leestllretfihisere... at 5.00.0 317191.177’ “as 470 DGYWQEWQlWJQlIQUIJI‘III’II; But this was not all. The difference in latitude between Alexandria and Syene really amounts to 'only 7° 5’, so that the direct distance between the two cities, supposing them to have been really situated on the same meridian as Eratosthenes assumed them to be) would not have exceeded 425 G. miles, or 4250 stadia, instead of 5000. His arc was therefore in reality 750 stadia too long. i It is remarkable that while the terrestrial measurement was thus grossly inaccurate, the observation of latitude as deduced§ frpm the gnomon at Alexandria was a very fair approximation% to the truth: a fiftieth part of a great circle being equivalent; r6311 arc of 7° 12', thus exceeding by about 7' only the true‘; interval between Alexandria and Syene,3 while falling short of that between Alexandria and the real tropic by about 30' or half a degree. § 8. It appears indeed almost certain that Eratosthenes himself was aware of the imperfection of his data, and regarded the result of his calculation only as an approximation to the truth. Hence he felt himself at liberty to add 2000 stadia tol the 250,000 obtained by his process, in order to have a number i that would be readily divisible into paiaworinto degrees) of 360 to a greatacircglef The result would of course‘ be that 2 Leake On some disputed Questions 4 It is by no means clear whether of Ancient Geography, p. 101. the division of the great circle into 3 The latitude of Alexandria is 31° degrees was known to Eratosthenes, or 10' 45"; that of Syene as already stated at least was adopted by him. It would 24° 5' 30", the exact difference between rather appear from a passage in Strabo the two is therefore 7° 5’ 15". 1 (ii. p. 113) that he divided the equator 624 HisronY or ANGIENT GEOGRAPHY. GHAP. XVI. each degree would be equivalent (according to his calculation) to 700 stadia; though it in reality contained no more than J 600 of the ordinary Greek stadia of 600 Greek feet in length. “a, Hence it has been supposed by many modern writers that Eratosthenes really employed stadia of this length, or in other words that where he estimates distances in stadia, he is not using the term in the usual sense—the only one that would be under- stood by his contemporaries, or by subsequent Greek writers —but means everywhere stades of 700 to a degree, or rather more than 514 Greek feet each.5 These writers in fact assume that Eratosthenes must have known the true length of a degree, and the real circumference of the globe, and that therefore his. stade must have been a different measure from that in ordinary‘ use. But the account of his operation, which is given us with great clearness, proves—as might indeed almost have been assumed without it—that he in reality followed the converse method. He started from what he believed to be a well-ascer- tained terrestrial measurement, and deduced from thence the circumference of the globe and the length of its aliquot parts. His conclusion was erroneous, because his data were inaccurate, and his observations defective. But none of the writers who have transmitted to ES the details of his calculation, have given us the slightest hint, or evidently had any idea, that he was not employing the customary Greek stade, the length of which was familiar to them all. After all it must be admitted that the calculation of Eratosthenes, considering the disadvantages under which he laboured, came surprisingly near the truth. His measure- ment of 250,000 stadia (the immediate result of his calcula- ition) would be equivalent to 25,000 geographical miles, while degree, rather than 4200 to a siatieth, as involving no substantial error, though <- 3 Strabo, i. p. 65. The passage is signification than that given in the text. unfortunately mutilated, and cannot 4 Strabo, i. p. 64. é’m're ei ,uh To be restored With certainty, but the ,ué'yeeos 1'09 ’A'r2\a1/'rucofi wekoi'yovs words To Aourbv pe’pos 1rap6t 'rb Aexflév élcu’mvs, KEV WMTV 'r‘mds e’lc 'rfis ’IB')7pias Szda'r'npa i'nrc‘sp 'rb 'rpi'rov pe’pos 511 To?) eis 'riyv ’lv5uc'))1/ Std 705 airroi) 1rapaA— ZAou mixAov can clearly have no other , AfiAou. 2s2 628 HIsroRY or ANGIENT GEOGRAPHY. carp. XVI. /‘ continued under the name of Caucasus along the northern frontier of India, until it ended in the Eastern or Indian ‘Ocean? A similar proceeding had been already resorted to by Dicaearchus, who had in like manner divided the known world by a longitudinal line, traversing its whole extent from the Columns of Hercules to the range of Imaiis, which he took as the line of separation of the northern and southern portions, describing the different regions and countries with reference to this imaginary line, and not according to the customary division into three continents.6 It does not appear that Era- tosthenes attached any such special importance to this par- ticular line, but it was evidently selected by him as traversing a number of points the position of Whic_hwas__ known, or sup- posed to be known, and as ‘being the only linewhich through its whole length passed through regions with which the Greeks were in some degree acquainted’. It was also supposed to be the line which traversed the inhabited world in its greatest ‘length, from the farthest extremity towards the west to the supposed extreme point to the east; and hence the distances jmeasured along this line would give as their result the total J length of the earth’s surface, as known to geographers. It is evident that the determination of any such line as this, with any approach to accuracy, required the possession of a number of correct observations of latitude for different points along its whole extent; but such observations were almost _wholly_wan_ting. It was not merely that in the age of Era- tosthenes the only methods available for this purpose were of a rude and imperfect description, and could not be relied upon for accuracy, except within very wide limits, but it is certain that no such observations were in existence, unless in 5 Strabo, ii. 1, § 1. 'rfis OZKOUILLEIV‘IIS‘; and refer to the above 6 Agathemerus, Geograph. 1. c. 1, § 5. l passage of A gathemerus as their autho- It is remarkable that many modern ' rity. But no such statement is found 11‘I\;+fi‘l‘0 flfihjl‘jl‘l‘l'lfi" +lwn nnrnvnnl-n nnln“ n1 Q.“ A call- ._..._ ,jf--. . _ ‘l g T J- o,‘ “Alums”, ‘Mud-“\ALLJ uuv wuvuA-wuu vulvrxv; j 1.1.1 AlbchbllUl-Llbl UH, .LlUl uttvb .l. Illbb WlBll Leake, should state that this llIlG was i the expression in any ancient author. termed by Dicsearchus the 8wt4>pa'y,ua \ .- ._/' can». XVI. ERATOSTHENES. 629 a very few and isolated cases. Hence we cannot wonder that the attempt thus made came out in some points very wide of the truth; the remarkable thing is rather that—for the , western portion of its course especially—the line should have made as near an approximation to correctness, as proves to be the case. §12. This fundamental parallel of latitude (as it may be called for want of a better distinctive term) was supposed to begin at the Sacred Promontory (Cape St. Vincent) which was generally (though erroneously) regarded as the westernmost point of Europe, whence it passed through the Strait of the Columns, or Straits of Gibraltar, the Sicilian Strait, and the southern extremities of the Peloponnese and Attica, to Rhodes and the Gulf of Issus. Of these points, if we take the parallel of 36°, which really passes through the Straits of Gibraltar, as representing that intended by Eratosthenes, we shall find . that the Sacred Promontory is placed just about a degree too far south, while the Sicilian Strait, or Strait of Messina, is brought down more than two degrees to the south, into the position of the island of Gozo near Malta; on the other hand the same parallel passes within less than half a degree of Cape Malea (generally regarded by the ancients as the southernmost point of the Peloponnese); while it actually traverses the southern portion of the island of Rhodes, about 25 miles from the city of that name, and approaches within the same distance of the entrance to the Gulf of Issus or Iskenderun. It is certain indeed that Eratosthenes himself was to a great extent aware of the imperfection of the means at his command, and did not regard his proposed line as a matter of scientific accuracy, but only as a rough approximation. This is evident from the manner in which he speaks of the Gulf of Issus without indicating any particular point of it, and of Rhodes, without stating whether the island or the city was meant; but still more from his repeatedly referring to the parallel in question as passing through Athens and 630 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XVI. Rhodes,7 though he elsewhere indicated distinctly that the parallels of the two cities were separated by an interval of 400 stadia (40 G. miles).8 Such a difference he regarded as immaterial in considering such distances as the diameter of the known world. After making due allowance for this admitted vagueness and laxity, it will be seen that the only instance in which any very serious error is involved in the line of the assumed parallel in regard to the position of the Sicilian Strait, which is brought down nearly into the latitude of Malta.// But so grave an error in regard to a place which must have been so familiar to the Greeks, sufficiently shows how little attempt could as yet be made to base these geographical conclusions upon trustworthy data. In the present instance the mistake ,Cmade by Eratosthenes was continued by almost all subsequent geographers till the time of Ptolemy.9 I Closely connected with this error was the false conception formed by all the earlier Greek geographers of the north coast of Africa, which they supposed to have a nearly uniform 7 Hence we find him (as reported by Strabo) continually referring to this line, sometimes as the parallel through Rhodes, at others as that through Athens, without naming any other points. In one instance only does he designate it by the fuller description as i; as 2m- MTW Kai ’A9'm/63V Ical ‘P6501; 'ypapjugi (ii. 1, § 24, p. 79). The strange false reading that had found its way into all our MSS. of Strabo, in several of the other pas- sages, where 6 8;’ 1617111311 mircAos had been corrupted into 6 616i ewe” mines or wapotAA-nAos, and had remained uncor- rected by all editors down to Kramer, had the effect not only of introducing into the geography of Eratosthenes a name first known to the writers of the Roman Empire, and of which no men- tion is found in the description of India and the far East, where it would have found its appropriate place, but of re- ferring to this remote and unknown 0 ll -- u y l locality as a laminar designation for this important parallel. The correction of these passages by Kramer (followed by the most recent editors) so as to accord with the correct reading found in others precisely similar, has had the effect of removing one of the greatest stumbling-blocks to the student of ancient geography. (See Kramer’s note on Strabo, ii. p. 65.) 8 Strabo, ii. 1, p. 87. The real dif- ference is much greater; Athens being situated in 37° 58' N. latitude, while the city of Rhodes was in 36° 26'; but Eratosthenes, like all the other Greek geographers, brought down the pro- montory of Sunium, and with it the whole of Attica, and Athens itself, much too far to the south. 9 Hipparchus was indeed an ex- ception, who brought down the parallel passing through Rhodes (correspond- ing with that in question) considerably to the south of Syracuse (Strabo, ii. p. 134). Strabo, however, recurred to the erroneous view of Eratosthenes. CHAP. XVI. ERATOSTHENES. 63 1 direction from east to west, with the exception of the two bays called the Syrtes, which they regarded as mere indentations, thus ignoring altogether the manner in which that coast projects to the northward in the neighbourhood of Carthage. Carthage itself was thus placed far to the south of its true position,1 while the island of Sicily was brought down to meet it; it being well known that the interval between the two was not very considerable. § 13. Having thus drawn one main line through the whole length of the Mediterranean from the Strait of the Columns to the Gulf of Issus, Eratosthenes next proceeded to draw a \ ,— "m. __l._ .__ _____..___ ..._- meridian line at right angles to ital-passing through— Alexandria; ~and Rhodes, which heco-nsidered to be on ‘the same meridian. This assumption involved in the first instance an error of con- siderable amount, the city of Rhodes being really situated more than a degree and a half of longitude to the West of Alexandria, while Syene, which (as we have already seen) he also regarded as on the same meridian, lay in reality three degrees to the east of it. Producing this line in both direc- tions, he conceived it as passing through Meroé to the south (which is really situated very nearly in the same longitude with Syene), and thence up the course of the Nile to the land of the Sembritae, the remotestrp‘eople in this direction of whom: hewhad knowledge. N orthwards" again from Rhode‘srit was supposed to »pass ‘through Byzantium (which lies in fact between the meridians of Alexandria and Rhodes) to the mouth of the Borysthenes, a very vague designation, as ‘that river forms an extensive estuary, the whole of which however lies two or three degrees farther to the east than Byzantium.2 1 Strabo supposed Carthage to be only 900 stadia or 90 G. miles to the north of the parallel of Alexandria, though there is in reality a difference of more than five degrees and a half of latitude between the two cities (Strabo, ii. p. 133)! It is remarkable that this erroneous position of a city so well known to navigators appears to have been founded on supposed astronomical observations with the gnomon (Strabo, l. 0.). See the passage cited in Note A, p. 661. 2 If the Greek city of Olbia or Olbio- polis, situated near the mouth of the Borysthenes, was the point referred to 632 HISTORY or ANGIENT GEoGRAPHY. GHAP. xvI. § 14. At the same time this erroneous conception of the relative position of these points in longitude was combined with equally mistaken ideas of their distance in latitude, which had the effect of placing the mouth of the Borysthenes 9350 stadia to the north of Rhodes; while the real interval in latitude between the two, is only about 10% degrees, or 630 G. miles. It is evident how defective a map must neces- sarily prove, which was based in the first instance on such data as these; but it cannot be denied that in his whole course of proceeding Eratosthenes showed a clear comprehension of the problem which presents itself to thenscigntific geographer, and the method he pursued was undoubtedly the._ best that was possible for him under the disadvantages in which he “found himself. It was the want of trustworthy observations for latitude and longitude, and the extremely defective cha- racter of all calculations of distances, that opposed an in- superable barrier to the attainment of anything like scientific accuracy. The Greeks were undoubtedly familiar at this period with the use of the gnomon in determining latitudes, and as we have seen in the case of Alexandria, such observations, when carefully made, were susceptible of a considerable degree of accuracy: but this was rarely the case, and in the majority of those instances—very few in number as they are—in which we know that such observations were actually made, the result is far from satisfactory.3 More generally the latitude seems to have been inferred from the determination of the length of the day at the summer or winter solstice, a method which could give at best but a rough approximation, and which was greatly vitiated by the very imperfect means at the command of ancient astronomers for the measurement of time. But defec- tive as these modes of observation were, they would still have offered a comparatively trustworthy basis, had there been a -—which is probable, as it was the only I would be just about three degrees. spot where any observations could have 3 Note A, p. 661. been made—the error in longitude l CHAP. XVI. ERATOSTHEN ES. 633 sufficient number even of such rough observations available: but this was certainly not the case, and the whole course of the subsequent discussion of the positions assumed by Eratosthenes sufficiently proves how little reliance was placed on authorities of this kind by succeeding geographers. § 15. But if the means at the command of Eratosthenes for the determination of latitudes were thus imperfect, far more was this the case with regard to longitudes. Here indeed the want of any precise mode of observing diurnal time, or of com- paring such observations with one another, was absolutely fatal. Hipparchus indeed had the sagacity to point out that the observation of eclipses might be applied to this object; but even if the idea had occurred to earlier astronomers, it is certain that no observations had been made with such a view; and the few general notices of such phenomena were wholly destitute of the accuracy requisite for scientific objects. Even in the time of Ptolemy, more than three centuries later, we shall find that scarcely any observations of this kind were available.‘ The entire want of any accurate knowledge of longitudes, even in the case of well-known localities, is sufficiently proved by the circumstance that Eratosthenes placed Carthage and the Sicilian Strait on the same meridian with Rome, though the one lies more than two degrees to the west, the other more than three degrees to the east of that city I5 The effect of this error, combined with the one already noticed in regard to the latitude of the Sicilian Strait, was of course totally to distort the map of this part of the Mediterranean. 4 See Ptol. Grogr. i, c. 4. The degrees! So that the error amounts most noted example of such observa- tions, to which even Ptolemy refers as a typical instance, was that of the cele- brated eclipse that occurred before the battle of Arbela (B e. 331), which was said to have been observed at Arbela at the fifth hour, and at Carthage at the second hour. This would imply an interval of forty-five degrees of longi- tude between the two cities; the real difference being less than thirty-four to just about one-fourth of the whole distance. The vagueness of the ob- servation is in this case sufficiently attested by the form in which it is reported. 5 Strabo, ii. p. 93. Yet Strabo,while censuring Eratosthenes for his inaccu- racy in this respect, himself falls into the grave error of placing Rome far to the west of Carthage. 634 HISTORY OF ANCIENT GEOGRAPHY. can». xvI. § 16. At the same time that the means of correcting a map by the only secure criterion—the determination of latitudes and longitudes—was thus in great measure wanting ; even the measurement of ordinary distances was, as we have already had repeated occasion to observe, of the rudest description. And-thiswaS especially the case with regard to distances by sea. We have seen that the ancient navigators had no means of determining their progress analogous to the modern log, so that the computation of distances by sea was really nothing more than rendering the number of days or nights’ voyage by a rough estimate into stadia. This was the method avowedly employed by Scylax, and doubtless also by Timosthenes, upon whose statements Eratosthenes appears to have in great measure relied for the geography of the Mediterranean.6 Im- perfect as such a process would be, it would still give something like an approximation to the truth, wherever the voyage was one that was frequently or habitually made : in other cases it would be altogether uncertain. When we compare the prin- cipal distances given by Eratosthenes in the Mediterranean with the results of modern observation, we shall find the fluc- tuations arising from these causes very much what might have been expected. Thus the distance from the Straits of Gibraltar to Carthage is estimated at 8000 stadia, and that from Carthage to the Canopic mouth of the Nile, just beyond Alexandria, at 13,500 stadia. The former of these, which agrees with the compu- tation of Scylax, is a very fair approximation to the truth, while the latter, though considerably in excess, if compared with the direct distance, according to the course which a modern vessel would pursue, is not materially so, if we allow for the principal sinuosities of the coast, of which the ancient navigators were in this instance unaware. The whole length of the Mediterranean was apparently computed by Eratosthenes E/at 26,500 stadia ;7 an estimate naturally in excess of the truth, 3 See Chapter XV. p. 580. given directly; but results from the 7 This number is indeed nowhere combination of the distance from CI-IAP. XVI. ERATOSTHENES. 635 as it was arrived at by adding together distances from point to point, estimated according to the course of navigation, and then computing the whole, as if they formed one continuous and uniform line. Yet, notwithstanding this grave defect in the mode of calculation, the error in excess is by no means so great as might be expected.8 The distance from the Straits to Issus comprises in reality about 411-12- degrees of longitude, which would give in round numbers a little more than 2,000 geographical miles, or 20,000 stadia, for the length of the Mediterranean, as measured along the same parallel of latitude. It would be indeed impossible in reality to sail in a direct course from the one extremity to the other, on account of the projection of the African coast to the north ; but of this Eratosthenes was unaware. Hence one unavoidable source of error in his computation. But erroneous as the result thus attained may appear, when compared with our improved modern methods of calculation, it is remarkable that it is not only much nearer the truth than that adopted by Ptolemy three centuries later, but it is actu- ally a better approximation than was arrived at by modern geographers till about two centuries ago. \Vhile Eratosthenes made the length of the Mediterranean about 6000 stadia, or 600 geographical miles, too great—an error of rather more than one-fifth—it is a fact that the best modern maps in use, as late 5 as the year 1668, assigned it a length of fifteen degrees in excess of the truth, or nearly one-third greater than the _, reality.9 , § 17. A few of the other more important distances given by Eratosthenes will tend to show how far he had arrived at a reasonably correct idea of the map of the Mediterranean. Rhodes (which Eratosthenes looked on as on the same meridian with Alex- andria) to Issus, with those already cited from the Straits to the Canopic mouth. 8 Note B, p. 662. ‘’ Gossellin, Géogmplu'e des by M. Sanson in 1668 placed the Gulf of Issus 60° of longitude from Cape St. Vincent. The real interval is just about 45°, Cape St. Vincent being in 9° W. long, and the Gulf of Issus being intersected by the meridian of Grecs if analyse'e, p. <12. The maps published ‘ 36° E. longitude. 636 HISTORY or ANGIENT GEoGRAPHY. CHAP. XVI. Thus we find that he reckoned 7000 stadia from the Pillars of Hercules to Massilia, and 6000 to the Pyrenees—that is to the point where these mountains abut on the Mediterranean. Both distances are below the truth, but present a very fair approxi- mation, if they are measured on the map without following the minor sinuosities of the coast. On the other hand his esti- mate of the length of the Euxine from the Bosphorus to the mouth of the Phasis was 8000 stadia, though the distance is really less than 600 G. miles (6000 stadia) :1 but to this he added 600 stadia more for the distance from the Phasis to Dios- curias, which he, in common with all the other Greek geo- graphers, regarded as the easternmost extremity of the Euxine? § 18. But while Eratosthenes possessed pretty accurate knowledge of the shores of Spain and Gaul that border the Mediterranean, his information as to the rest of those countries was very imperfect. His knowledge of the western regions of Europe, and the shores of the Atlantic, was indeed derived almost exclusively from Pytheas, to whom, as we have seen, he was far from attaching unlimited credit; but he regarded his account as on the whole the most trustworthy that he pos- sessed, and it is certain that in many respects his information was more correct than that of several later geographers. He even followed Pytheas in regard to Thule, the existence of which he clearly admitted: and adopted the statement of the same author with respect to the length of the solstitial days and nights in that island.3 Hence he was justly led to 1 It is probable that an exaggerated l Dioscurias, which was placed both by idea of the length of the Euxine had ‘ Eratosthenes and Strabo 600 stadia become traditional among the Greeks, (60 G. miles) east of the mouth of the like so many similar errors. We have 2 Phasis, was really situated to the seen how greatly it was over-rated by N.N.W. from that point; and about Herodotus (see Chapter VI. p. 177). 20 G. miles farther west. 2 The permanence of this error, in 3 It must also have been from which even Strabo participated, serves Pytheas that Eratosthenes derived the to show how little care was bestowed statement that to the inhabitants of by ancient navigators upon correct Thule the arctic circle (in the Greek b. arings; and, . consequently, how sense of the term) coincided with the little assistance they could derive from tropic (ap. Strab. ii. 5, § 8, p. 114). these in determining their longiiudes. ' CIIAP. XVI. ERATOSTHENES. ‘ 637 inferthat the parallel of Thule was the most northerly with which he was aequainted,_and placed it 11,500 stadia to the north of that passing through the mouth of the Borysthenes? a Eonclusion for which he is severely taken to task by Strabo,‘ though it in reality comes very near the truth.5 At the same time he placed the northern part of Gaul on the same parallel with the Borysthenes ; which last, as we have already seen, he erroneously placed considerably too far to the north. But the error resulting from this cause was not very material : and though it had the effect of elongating the great island of Britain much beyond the truth, it is certain that the map of these western regions as conceived by Eratosthenes, had con- siderably more resemblance to the reality than that afterwards drawn by Strabo.6 It is remarkable that no mention is found either in the extant notices of Eratosthenes, or of his guide and predecessor Pytheas, of the second of the two great British islands, Ierne.7 But no negative inference can safely be drawn from this, on account of the fragmentary character of these notices. Eratosthenes appears to have had no accurate knowledge of thgregionsgrgrth of the Euxine, so as to have any definite po_itr_1t_s_inthis part of the world to compare with those in the iQst of Europe. Observations would doubtless be here entirely 4 i. 4, p. 63. 7 The names of Albion and Iernc 5 The real difference in latitude between the mouth of the Borysthenes and the Shetland Islands (supposing them to be the Thule of Pytheas) does not exceed 13%0 of latitude, or about 810 G. miles: but if Thule be placed, as Eratosthenes supposed it to be, under the arctic circle, the difference would be about 20° or 12,000 stadia. 6 Yet Strabo confidently asserts that Timosthenes and Eratosthenes “were utterly ignorant” concerning Spain and Gaul, and still more so in regard to Germany and Britain ('reAe’ws i77- vdovu 'roi 76 ’IBnpuca Kal ra Kekruca, uvpiq) 8% ,uaAAou 'ra I‘eppauuca Ical 'ra Bpe'r'ravucoi. Strab. ii. 1, p. 93). are found (as has been already men- tioned, p. 398) in the treatise De Mundo (wept Kdo'aov), ascribed to Aristotle, but that work is unquestionably spu- rious, and belongs to a much later period. The name of _Ierne is first found among extant authors in Strabo, though there can be little doubt it was known long before. Polybius uses the expression “the British Islands” (at Bpe'rawucal V'ho'or, iii. 57), showing clearly that he was aware of the existence of more than one of them. Unfortunately the part of his work containing a fuller account of them is lost (see Chapter XVII). 638 HIsToRY OF ANCIENT GEoGRAPHY. can». XVI. wanting: but it appears singular that the long continued intercourse of the Greek colonies with the Scythian tribes of the interior had not led to any increased geographical know- ledge in this quarter. Eratosthenes must indeed probably have known, as well as Strabo, that the mouth of the Tanai's lay considerably to the north of that of the Borysthenes, but we do not find that any notice was taken of this fact in arranging the distances for his map of the world. He was equally unacquainted with the northern shores of Germany, and though he certainly supposed that there was / continuous sea to the north of Europe, as well as Asia, this Was either a mere hypothetical inference, or ,aconclusion from the erroneous idea that Pytheas had sailed along these northern shores as far as the Tanai's. So far as we are able to judge, his knowledge of the extensive regions to the north of the Danube and the Alps was of the vaguest and most imperfect character. We learn only from an incidental notice that he mentioned the name of the Hercynian forest ;8 by which he probably meant the Black Forest, in which the Danube takes its rise. The same statement, as we have seen, is already found in the treatise De Mirahilibns ascribed to Aristotle. § 19. The parallel of Thule being thus taken for the most northern limit of the habitable world, that which passed through the land of the Sembritae on the Upper Nile was assumed to be the southernmost. This parallel was placed by Eratosthenes 3400 stadia to the south of 'Meroe—which would bring it down to the south of Sennaar—alndhe correctly e§tigaied that . the. same paralletif nmdilnedsastwards, Would-- pass through the ‘Land of . Cinnamondmand, the easternmost- promlontbry of Africa, which was als_o__at that time the__m_qst stfiiiiéiifttiat of “the Hmtinent known to navigators. He fufiheiimassumedééfor here he had unquestionably no real information—that Taprobane, the southernmost land of which he had heard in connection with Asia, lay on the same parallel 8 Caesar, B. G. vi. 24. CHAP. XVI. ERATOSTHENES. 639 with the Region of Cinnamon, so that this line passed through all the most remote regions towards the south, and might thus be fairly taken as the southern limit of the habitable It is unfortunate that we do not know with certainty at what distance from the Equator Eratosthenes conceived this line to be drawn, but it may be inferred from a comparison of other authorities that he fixed it at 8300 stadia from the equinoctial line :1 a position very near the truth, “if we suppose Eratosthenes to have here been calculating by his own estimate of 700 stadia to a degree, as in this case he would naturally do.2 § 20. The other distances along his principal meridian line he estimated as follows: from Meroe to Syene 5000 stadia; from Syene to Alexandria 5000; from Alexandria to Rhodes 37 50 ; from Rhodes to the Hellespont 4350; and from thence to the mouth of the Borysthenes 5000 ; thus giving as the sum total 26,500 stadia from his southernmost parallel to the Borysthenes, or, with the addition of the 11,500 stadia assumed from thence to the parallel of Thule, 38,000 stadia for the total width of the habitable world. Of these distances that from Alexandria to Rhodes was the result of a gnomonic observation made by Eratosthenes himself 3 --the distance having previously been estimated by navigators at 4000, or according to others as much as 5000 stadia—a striking instance of the vagueness of their mode of reckoning? The distance from Syene to Alexandria was supposed to have 9 The information possessed by the Greeks in the time of Eratosthenes concerning the peninsula of India, and still more concerning Taprobane itself, was so utterly vague and erroneous that it is impossible to regard this assumption as more than a lucky guess, founded on the belief that Taprobane was the most southerly portion of Asia, and perhaps confirmed by the notion (utterly unfounded in itself) of that great island extending from east to west towards the Land of Cinnamon. 1 Note C, p. 664. 2 A line drawn through the Land of Cinnamon and the south of Sennaar would about coincide with the parallel of 12° N. latitude. Of course this would give 8400 stadia as the distance from the equator, reckoning 700 stadia to a degree. The result adopted by Eratosthenes therefore, whatever the process by which he arrived at it, dif- fered from the truth only by 100 stadia, or 10 G. miles. 3 Note D, p. 665. 4 Strabo, ii. 5, p. 125. 64o HrsToRY or ANCIENT GEOGRAPHY. CHAP. xvI. been measured, though, as we have already seen, it was really wide of the truth; the position of Meroe also had been deter- mined by astronomical observation; but all the others could only be the result of vague computation from the distances estimated by navigators. ‘ The effect of such a mode of computation would be unques- tionably to exaggerate the distance; but it seems incredible that such a geographer as Eratosthenes should have taken the actual distances traversed’on such a circuitous voyage as that from Rhodes to the mouth of the Borysthenes, and added them together in order to obtain the direct distance, or interval of latitude between the two points. Yet even this hypothesis fails to explain the full amount of the discrepancy. The dis- tance from Rhodes to the Hellespont, as measured on the map by the nearest course. that would be possible for a vessel to pursue, on account of intervening headlands and islands, is (in round numbers) only about 300 G. miles, or 3000 stadia instead of 4350; that from the Hellespont to the Bosphorus, about 150 miles ; and from the Bosphorus to the Borysthenes, where it was easy to pursue a direct course, about 390; thus giving a total result of 8400 stadia, instead of the 9350 assigned by Eratosthenes. At the same time it must be observed that the effect of this excess in the computation of the distances was in great part neutralized, as afleetiug the latitudes, by his erroneous allow- ance of 7 00 stadia to a degree. Hence he placed the mouth of the Borysthenes in 480 N. latitude, which is only about 175° to the north of its true position. § 21. In continuing the line which he had assumed as his main parallel of latitude, from the Gulf of Issus eastward across thqfiqogntinent (of Asia, EratosthenesUengoiinteredhdifim culties of another kind. Here, however, he derived important assistance from the conquests of Alexander, and the itineraries- of those who had accompanied him, as well as from the state- ‘ ments oIfPatrocles, who (as we have seen) had himself held ‘importantIggyernments in Upper Asia. He conceived the CHAP. XVI. ERATOSTHENES. 641 line in question to be carried from Issus to Thapsacus on the Euphrates, thence to the Pylae Caspian, next, to the foot of the Indian Caucasus, and from thence in a straight line to the Indian Ocean. Throughout its whole length he regarded this line as nearly coinciding with the southern foot of Mount Taurus, which he considered as a vast range of mountains, occupying in parts a width of not less than 2000 stadia, and stretching in a direct line, nearly parallel with the equator, from the interior of Lycia and Cilicia, where they were fami- liar to the Greeks, north of the plains of Mesopotamia and Assyria, as well as of the table land of Persia, and the plains \ of India, till they ultimately ended in the unknown ocean that formed the eastern boundary of Asia. Imperfect as was such a view, there can be no doubt that we trace in it some approach to a true conception of the _ leading facts in the geography of Asia, and to that systematic grouping of the- subordinate ranges of mountains, without which it is impossible to form a clear idea of the physical geography of a great continent. It is perfectly true that the range of Mount Taurus, which occupies so prominent a positiori throughout the south of Asia Minor, is connected with the lofty mountain ranges of Armenia and Koordistan, and througli them with the range of Elbourz to the south of the Caspian; which is again connected, though more irregularly, with the.I stupendous range that was known to the Greeks by the namesi of Caucasus and Imaus, and was justly regarded as forming a continuous mountain barrier to the north of India. The course of this vast mountain system is indeed very far removed from the regular and uniform direction which was ascribed to it by Eratosthenes, and it is connected with various other mountain chains, some of them of great height, that branch off to the north and south, of which comparatively little notice was taken by the Alexandrian geographer : but the great leading fact that Asia is traversed in about the latitude sup- posed, by a broad belt of mountains, was rightly conceived, and tended to throw much light upon its physical configuration. VOL. I. 5 2 r . / 642 HIsToRY 0F ANCIENT GEoGRAPHY. carp. XVI. Nor do the leading points through which Eratosthenes drew his assumed line deviate very widely from the required posi- tion. If we take as before the parallel of 36° to represent the line in question, we shall find that it passes a very few miles ~ to the north of Thapsacus on the Euphrates, and after travers- ing the mountain regions of Koordistan and the plains of Media, passes within less than a degree north of the Pylae Caspiae, and about the same distance to the north of the Hindoo Koosh, where it was traversed by Alexander.5 From this point indeed the great chain of the Himalayas trends away rapidly to the south-east—a fact of which Eratosthenes was wholly ignorant. He accordingly regarded the mountain chain as still preserving its direction from west to east, in accordance with the supposed course of the Ganges, which was believed by all the Greek geographers at this period to hold a due easterly course to its outflow in the eastern ocean. § 22. The distances as computed by Eratosthenes along this main parallel were: from Issus to the Euphrates 1300 stadia ;6 from the Euphrates to the Pylae Caspiae 10,000 stadia; thence to the foot of the Caucasus 14,000; and from thence to the mouth of the Ganges 16,000 stadia.7 These distances were almost all based upon the itineraries of Alexander and his generals, or, in the casve- of India, upon those which had been furnished to Sgldeucus. They therefore inevitably laboured under the disadvantage of being itinerary distances, con- verted into rectilinear ones, and laid down on the map as such, without attempting to correct them either by observa- tions of longitude, or by any trigonometrical 'measurements, neither of which were at the command of the Greek geo- 5 The parallel of 36° actually passes about 50 miles south of the Caspian Sea, and the same distance south of Balkh, the ancient Bactra. It isjust ~ about a degree to the north of the city founded by Alexander at the foot of the Caucasus, which may very probably in a direct line, without any allowance have been the point intended by Era- , for itinerary excess. tosthenes under the vague general 3 7 Ap. Strab. i. 4, § 5, p. 64. designation of “ the Indian Caucasus.” ' 6 No indication is here given of the point on the Euphrates intended; but it is probable that Thapsacus is meant as being the customary place of passage. In that case the distance is almost pre- cisely correct, as measured on the map OHAP. xVI. ERATOSTHENES. 543 grapher. The consequence is, as might be expected, that the distances greatly exceed the truth, and we cannot be surprised to find that in this instance the error is far greater than in regard to the length of the Mediterranean.8 But defective as were the materials at the command of Era- tosthenes for all this part of Asia, they were nevertheless considered so valuable that two centuries later Strabo was still content implicitly to follow him, because he could obtain no better information.9 J The whole sum of the distances thus computed by Era- tosthenes for the length of the habitable world, from the Sacred Promontory to the eastern extremity of India gave a result of 70,800 stadia. To this he added 2000 stadia more to the west of the Sacred Promontory—apparently with the view of allowing for the supposed projection of the coast of Gaul beyond that of Spain 1——and again 5000 more to the eastward, where he considered the extreme point of India to project to the south-east, so as to be 3000 stadia east of the extremity of his main parallel. g _ of" 17,800 stadia ; thus making the length slightly to exceed L_ By these additions he attained atotal sum, twice the assumed breadth.2 8 In attempting, however, to com- pare these distances with the reality we labour under the great difiiculty of not knowing the precise line of route along which they were measured ; and any line of march likely to be followed by an army would of necessity be ex- tremely circuitous. The direct distance, as measured on the map from .Thap- sacus to the Pylae Caspiae but little exceeds 620 G. miles or 6200 stadia: while that from the Pylze to Alexandria at the foot of the Caucasus gives only about 970 G. miles or 9700 stadia. The excess of the itinerary distances em- ployed by Eratosthenes amounts there- fore in the one case to more than two seventies, in the other to nearly two fifths of the whole. The whole interval in longitude between the Gulf of Issus and the foot of the Hindoo Koosh, which was what he in reality was attempting to estimate, is in fact only about 33°, which in lat. 36° is equiva- lent to 1600 G. miles, or 16,000 stadia, instead of 25,300 stadia, which was the result of the computation of Eratos- thencs. 9 wept (5V ’Epa'roo'6éz/'r)s oiz'Tws‘ el’p'qxev' 01’) 701p é'Xo/ae'z/ 'rL Aé'yew Bel/‘@1011 wept aim-6:11. Strabo, xv. 2, § 8, p. 723. 1 He conceived, in accordance with the erroneous view universally enter- tained in his time, that the Sacred Promontory (Cape St. Vincent) was the most westerly point of Spain : but he supposed, still more erroneously, that the promontory of Calbium, the western- most point of Gaul—which evidently corresponds with one of the headlands of Finisterre in Brittany—extended still further to the west. This idea was apparently derived from Pytheas. (See Chapter XV. p. 593.) 2 It is curious that this assumption had come to be regarded so completely 2T2 644 HISTORY OF ANGIENT GEOGRAPHY. GHAP. XVI. §23. Eratosthenes undoubtedly conceived, in accordance with the prevalent belief in his day, that the Ocean was found immediately to the east of India, and that the Ganges flowed directly into it. Just to the north of the Ganges the great mountain chain of Imaus, which he regarded as the con- tinuation of the Indian Caucasus and the Taurus, descended (according to his ideas) to the shores of the Eastern Ocean; and he appears to have given the name of Tamarus to the! headland which formed the termination of this great rangef‘i From that point he supposed the coast to trend away towards/l tthe north-west, so as to surround the great unknown tracts Z'of Scythia on the north, but sending in a deep inlet to the (,south which formed the Caspian Sea. I Of the northern shores of Asia or Europe he had really no i more knowledge than Herodotus, but, unlike that historian, he assumed the fact that both continents were bounded by the I Ocean on the north; a fact which is undoubtedly true, but in a sense so widely different from that supposed by Eratosthenes that it can hardly be held as justifying his, theory. In fact the conclusion of Eratosthenes was mainly based upon the erroneous belief that the Caspian communicated with the Ocean to the north in the same manner that the Persian Gulf did to the south; a view which was adopted by all geographers ,for a period of three centuries, on the authority of Patrocles? It was doubtless from the same authority that Eratosthenes derived his statement as to the dimensions of the Caspian Sea, as well as that concerning the outflow into it of the rivers ./Oxus and Iaxartes, which he assertsin a remarkably distinct and positive manner.5 Yet the erroneous idea of its com- as an admitted principle, that Strabo treats it as a matter of course that Eratosthenes, having fallen into error i with regard to the breadth of the known world, must necessarily be wrong in respect to the length also : for that it was a fact agreed upon by all the best authorities that the length was more than double the breadth (i. 4, §5, p. 64). Eratosthenes therefore having, as Strabo considered, exaggerated the breadth by carrying Thule to so high a latitude, was compelled to give an undue extension to the length also, in order to preserve the assumed propor- tion between the two. 3 Note E, p. 666. 4 See Chapter XIV. p. 574. 5 He stated that the part of the circumference of the Caspian “ which CHAP. XVI. ERATOSTHENES. 645 / munication with the Ocean to the north sufficiently shows how, far from trustworthy the information possessed by the Greeks really was.6 § 24. His ideas of the geographical position and configura- tion of India were in great measure erroneous. He conceived it indeed to be of a rhomboidal form, which may be regarded as a rough approximation to the truth, and he even knew that the two sides which enclosed the southern extremity were longer than the other two. But as he supposed the range of Imaus that bounded the country to the north to have its direction from west to east, while the Indus flowed from north to south, he was obliged to shift round the position of his rhomb, so as to bring the other two sides approximately parallel to the two thus assumed. Hence he conceived the projecting angle of India to have a direction towards the south-east, instead of the south, and even (as we have already seen) sup- posed it to advance farther towards the east than the mouth of the Ganges. He appears in fact to have obtained—probably from the information collected by Patrocles—a correct general idea of the great projection of India in a southerly direction towards Cape Comorin, but was unable to reconcile this with his previously conceived notions as to its western and northern boundaries, and was thus constrained altogether to distort its position in order to make it agree with what he regarded as established conclusions. It was doubtless from the same source that he had learnt the name of the Coniaci, as the ;people inhabiting this southernmost point of India;7 a name .5 which henceforward became generally received, with slight modifications, by ancient geographers. He was familiar also with the name of Taprobane, which was well known to the Greeks” (rbv ' i'rn'b r611 'EAMix/wx/ 'yva’iptpiov) was 5400 stadia in extent; that from thence along the coasts of the Anariaci (?), the Mardi, and Hyrcanians as far as the mouth of the Oxus was 4800 stadia; and thence to the Iaxartes 2400 (Strabo, xi. 0, § 1, p. 507). the latter distance at 80 parasangs (Id. xi. 11, §5), which exactly coincides with the distance given by Eratos- thenes. 6 To this may be added their igno- rance of the great river Volga. the I name of which is first found in Ptolemy. Patrocles had stated 3 7 Ap. Strab. xv. 1, p. 689. 646 HISTORY OF ANCIENT GEoGRAPHY. can». xvI. had indeed been known to the Greeks since the days of Alex- ander, but his ideas of its position were extremely erroneous, as he placed it south of the headland of the Coniaci (Cape Comorin), at a distance of seven days’ voyage from the main land, and extending for a space of 8000 stadia in length, from east to west, towards the extreme point of the Cinnamon Region on the coast of Africa.8 At the same time he placed the two (as already noticed) in the same latitude, as the most southerly regions of the inhabited world. § 25. Imperfect as was the geographical knowledge of India ‘possessed by Eratosthenes, it was not surpassed by any later igeographer, until after the time of Strabo. He was also the first to bring together anything like definite information with regard to Arabia, a country which on account of its proximity to Egypt, could not fail to attract in a high degree the atten- tion of the Greeks under the Ptolemies. We have seen that Alexander, at the time of his death, was preparing a voyage of discovery for the circumnavigation of the great vpeninsula of Arabia. The project thus interrupted was not resumed by any of his successors; but although we have no account of any regular expedition by which the coasts of Arabia, from the entrance of the Persian Gulf to the Straits of Bab-el- Mandeb, were examined and surveyed, in the manner that the coast from the mouth of the Indus to that of the Euphrates had been by N earchus, there appears no doubt that the cir- cumnavigation had been accomplished. It is probable indeed (as has been already shown) that the Greek traders from Alexandria did not usually extend their voyages beyond the southern coast of Arabia Felix; but this does not exclude the possibility of a few more adventurous spirits having con- tinued them to the Persian Gulf, or even to India itself.9 But even if this navigation were in practice confined to native 8 Ap. Strab. xv. 1, p. 600. not take place till after the time of 9 The voyage of Eudoxus, Who 1111- Eratosthenes. doubtedly visited India in person, did I CnAP. xvr. ERATOSTHENES. 647 tradersi'romthe Sabaean ports, it would doubtless be easy for the Greeks to obtain from them sufficient information to ascer- tain in a general way the extent and character of the Arabian peninsula. The fact that it presented no very serious diffi- culties to the navigator was many case sufficiently proved by the existence of this trade. Hence Eratosthenes had no diffi- culty in arriving at a general notion of Arabia so far as thel sea-coasts were concerned. The peculiar characteristics of the interior of that country presented the same obstacles to a more extensive acquaintance in ancient times that they have continued to do down to our own days. But the proximity of two such flourishing and civilized states as the Egyptian monarchy on the one hand, and the Syrian monarchy, with its seat at Babylon or Seleucia, on the other, could not fail to exercise considerable influence even on the wild tribes of Arabia. In consequence of this we find that a considerable commerce had developed itself, which was carried on by caravans across the deserts from one side of the peninsula to the other. Thus Gerrha, on the western coast of the Persian Gulf, had already become an important centre of trade; the and other productions of the southern districts of Arabia, "as well as of the opposite coasts of Africa being brought thither by caravans, and afterwards sent on from thence to Babylon and Seleucia.1 These caravans took forty days for the overland journey from Hadramaut.2 Other similar caravans carried on the communication from ZElana, at the head of the eastern branch of the Red Sea (now called the Gulf of Akabah), through the western portions of Arabia, and reached the district of the Minzeans in seventy days. ZElana was the port of Petra, which seems to have been already rising into importance as an emporium of trade: and from thence other caravans boldly struck across the desert direct to Babylon, a distance estimated by Eratosthenes at 5600 stadia.3 § 26. His general description of the physical characters of 1 Eratosthcn. ap. Strab. xvi. 3, p. 766. 2 Ibid. 1, § “1:, p. 768. 3 Ibid. § 2, p. 767. 648 HISTORY or ANCIENT GEOGRAPHY. can». xvi. Arabia is very correct. The northern portions of the peninsula, bordering on Judqpa and Coele Syria, were barren and sandy, pro- ducing only a few palm-trees, thorny acacias and tamarisks, and devoid of running water, but having wells from distance to dis- tance. The inhabitants were wandering tribes of Arabs, dwelling in tents and subsisting on their herds of camels. The more southern districts, on the other hand, which felt the influence of the ocean, and were subject to periodical rains in summer, were fertile and produced abundance both of grain and cattle.4 This portion of Arabia was divided into four districts, inhabited by four principal tribes: the Minaeans adjoining the Red Sea, whose chief city was Carna or. Carnana; the Sabaeans, whose capital was Mariaba; the Catabanians, extending to the Straits of Bab-el-Mandeb ; and the Chatramotitae, the furthest to the east, whose capital was Sabata.5 The name of this last people is evidently still retained in that of Hadramaut; of the others, the Minaeans were apparently the occupants of the Hedjaz: while the Sabaeans unquestionably held the fertile district of Yemen, the chief city of which retained the name of Mareb till a late period. The site of the Cattabanians is clearly indi- cated, but their name is otherwise little known. According to Eratosthenes their territory produced frankincense, while myrrh was brought from that of the Chatramotitae,6 both of which were important articles of expert; but it is probable that the greater part of these aromatic spices were in reality brought from the opposite coast of Africa, as well as the cinnamon for which it was famous.7 §27. No settlements had been formed by the Egyptian 4 There was however one remarkable exception. It is strange to find him stating that these more fertile regions of Arabia possessed abundance of cattle and beasts, except horses, mules, and swine (Strabo, xvi. 4, §2, Boo'lcmud'rwv 're dxpdomd, 1min! ‘L'rrjzrwv Ical iqjuzeuwv and {611). Nor is there any allusion to horses being more frequent among the wandering tribes of the interior. 5 Eratosthen. ap. Strab. xvi. 4, § 2. 6 Id. ibid. 4, § 4. 7 It may be observed that Nearchus, when he came in sight of the headland of Maceta—the easternmost promontory of Arabia—was told that it was from thence (i.e. from Arabia) that cinnamon and other similar productions (a mud- pw/w'. 're Ital dAAa 'rowv'rdrporra) were conveyed to Babylon and Assyria (Arrian, Indica, c. 32, § 7). can». XVI. ERATGSTHENES. 649 monarchs upon the Arabian coast of the Red Sea, but the African side had, as we have already seen, been lined by an almost continuous chain of tradingstations, forming permanent settlements, though prevented by the barrenness of the ad- joining tracts from ever rising into colonies of importance. Hence Eratosthenes was well acquainted with the extent and dimensions of the Red Sea, which he describes as extending 9000 stadia from the head of the gulf adjoining Herotipolis (the Gulf of Suez) to the station of Ptolemais Epitheras, and 4500 from thence to the Straits. Both statements are very fair approximations, the former, however, somewhat exceeding, while the latter rather falls short of, the truth.8 He also estimated the distance from the Straits of Bab-el-Mandeb to the ex- tremity of the Cinnamon Region (Cape Guardafui) at 5000 stadia—a very close approach to the truth. But he appears to ‘I have had no knowledge of the coast of Africa beyond that point, which indeed continued until long after to be the farthest limit of Greek navigation in that direction.9 . § 28. We have already seen how greatly the knowledge 0 the upper course of the Nile and its tributaries had advanced under the Ptolemies. This was in part the consequence of direct exploration from Egypt, but still more from the establishment of a comparatively civilized state at Meroé, which carried on con- siderable intercourse both up and down the river. It is evident also that Meroe maintained direct communication with the Red Sea, from which it was distant only ten or twelve days’ journey :1 and it was this communication that enabled Eratos- 6 Ibid. § 4, 768. He appears, how-q forming a considerable bend or elbow ever, in one respect to have formed an erroneous idea of the figure of the Red Sea, which he conceived to have a direction nearly due N. and S. from Heroiipolis _to Ptolemais Epitheras, and thence to trend away to the SE. The general direction of this vast inlet is in reality remarkably uniform throughout its whole length from N.N.W. to SSE. But the notion entertained by Eratosthenes of its at the point mentioned, was received by all ancient geographers down even to the time of Ptolemy. 9 Thus Strabo, writing on the autho- rity of Artemidorus, a century later than Eratosthenes, speaks of the coast towards the south, after doubling the Southern Horn (Cape Guardafui), as wholly unknown (p. 774). 1 Eratosth. ap. Strab. xvii. 1, p. 786. At the present day there is a regular 65o HISTORY OF ANCIENT GEOGRAPHY. CHAP. XVI. thenes to connect his observations on the Upper Nile with those on the Red Sea, where he placed Ptolemais Epitheras on the same parallel with Meroé. At the same time he correctly placed Meroe itself at just about the same distance from Syene that the latter was from Alexandria.2 With the intermediate part of the course of the Nile he was well acquainted, and described clearly the great bend made by the river between Meroe and the Second Cataract—an important feature in the geography of this part of Africa, which has only become known to modern geographers in quite recent times.3 The knowledge possessed by Eratosthenes of the relations between the Nile and its tributaries was not indeed, as has been already pointed out,‘1 altogether free from confusion, arising principally from the different names given by the native tribes to the different branches of the river, and even to different portions of the main stream. But with due allowance for this source of error it must be admitted that the informa- tion of Eratosthenes was not only superior to that of any ancient writer down to the time of Ptolemy, but was much more correct than that possessed by modern geographers until a very recent period.5 Thus he clearly understood that the so-called “island” of Meroe was formed by the junction of two streams, the easternmost of which was the Astaboras (still called the Atbara), and the westernmost the Astapus, though according to others its propername was the Astasobas, while caravan trade from Suakin to Berber i remarkable bend of the Nile in Nubia on the right bank of the Nile, just here described by Eratosthenes was below its confluence with the Atbara, first verified in modern times by the and about 60 G. miles below Meroe. information procured by Burckhardt ” 2 This was probably derived from (Journal of Geogr. Soc. vol. ii. p. 24). the astronomical determination of its 4 Chapter XV. p. 584. position already referred to (see Note A). 5 M. Gossellin, writing near the end But there must have been at this of the last century, observes: “Quant period a more direct caravan route in use, as is the case at the present day, without following the circuitous course of the Nile, and this would afford the means of direct measurement, or rather computation of the distance. - 3 Colonel Leake observes that“ the au Nil, Eratosthenes décrit la partie supérieure de son cours avec assez d’ex- actitude pour que, depuis son siecle, on n’ait rien eu d’important a y ajouter, ni a y corriger ” (Geographic dcs Grccs ‘ analysée, p. 21). This first volume of his work was published in 1790. can». XVI. ERATOSTHENES. 651 the name of Astapus properly belonged to another branch which flowed from certain lakes to the south, and formed in fact the main and direct stream of the Nile itself.6 We have here a distinct indication of the \Vhite Nile, or Bahr-el-Abiad ; and this is confirmed by the mention of another “island ” formed by the confluence of two streams, higher up than that of Meroe, which was inhabited by the Sembritze, the descend- ants of an Egyptian colony. This can clearly be no other than the district of Sennaar, comprised between the two main branches of the river, the Blue and White N ile.7 Of the highlands of @yssinia on the contrary he appears to have had no knowledge; though he was certainly not ignorant of the existence of mountains in that direction, and correctly ascribed the periodical inundation of the Nile to the ' regular rains that fell in these elevated regions within the tropics.8 This theory indeed appears to have been generally adopted in his time. It is worthy of remark that everything points to the informa- tion of Eratosthenes having been derived from the inhabitants of the valley of the Nile itself, as would naturally be the case where a civilized community was once established so far up the river as Meroé: while Ptolemy on the contrary, at a much 6 Eratosthen. ap. Strab. xvii. i. p. 786. Thu 5% ’Ao"rd1rovz/ dAAoV ell/at, pe’oy'ra v’s'x 'rwwv Aim/£611 dab ,ueo'npfipfas‘, Kai o'Xe'o‘ov Tb Ica'r’ ei’zdei‘av o'cB/ra 'roi} NeiAov 'rofi'rou 1roze'iv. This notice ap- pears to me conclusive against the theory of Mr. Cooley (Claudius Ptolemy and the Nile, 8V0. Lond. 1854), who maintains that the ancients were alto- gether unacquainted with the White Nile, and that all their accounts of its upper course refer exclusively to the Blue Nile or Abyssinian river. Apart from the direct testimony of Eratosthenes, it is most improbable that the Greeks of the second century 13.0., to whom Meroé was as familiar as Khartoum is to us at the present day, should not have known of the union of 100 miles above it. The lakes (Max/at) from which it was supposed to flow, probably referred to the vast marshes which it forms above lat. 10°. It is very unlikely that any rumour of the Victoria and Albert Nyanza had reached his ears. 7 The comparative civilization, and skill in the mechanical arts, possessed by the natives of Sennaar at the present day, as compared with the other native tribes by which they are surrounded, is supposed by some modern writers to be derived from their Egyptian de- scent. See Cooley’s work above cited (pp. 22, 23) and the authorities there referred to. 8 'r7‘71/ 5% wkr'lpwo'ul ai’z'roz'i TObS Bepwoi‘zs IiuBpovs wapao'lceliagew. Strabo, 1.6. PIO- the two great streams, little more than ! clus ad Platom's T imam-m, p. 37. 652' HISTORY or ANCIENT GEoGRAPHY. CHAP. xvI. later period, seems to have derived his accounts as to the upper course of the river from the Red Sea. Hence arises in part the difficulty of reconciling the two authors. But the statements of Eratosthenes, taken by themselves, are perfectly clear and intelligible. He was also the first to mention the name of the N ubians (Neill-lat), whom he describes as occupying the country on the west of the Nile, from the neighbourhood of Meroé to the bend of the river. They were a great nation, not subject to the ZEthiopians of Meroe, but forming numerous petty sovereignties of their own.9 § 29. Of the rest of Africa Eratosthenes probably knew little, if anything, more than was already known to Herodotus. But as no extracts have been preserved to us from this part of his work, we can only infer that he furnished no information beyond what was in the possession of all later geographers. In one respect however he was certainly better informed than many of his successors,-—that he mentioned the island of Cerne, and many other Phoenician settlements beyond the Columns of Hercules, on the west coast of Africa. His knowledge of these was probably derived from Carthaginian sources 1 : but as the settlements themselves had disappeared at a later period, the fact of their existence was disbelieved, and the statements of Eratosthenes were summarily rejected by Artemidorus and Strabo? 9 Ap. Strab. xvii. p.786. Accord- means of judging whether it was ing to Mr. Cooley (Claudius Ptolemy and the Nile, p. 41) the name of Nubians, which is not a native appel- lation, was at first applied only to the natives of Kordofan (immediately west of Sennaar), and it was not till a later period that that people occupied the lower districts between Meroe and the frontiers of Egypt. This would be entirely in accordance with the state- ment of Eratosthenes, as cited by Strabo. 1 The voyage of Hanno had unques- tionably taken place long before the time of Eratosthenes; but we have no known to the Alexandrian geographer. No allusion to it is found in Strabo; unless it be assumed to be the autho- rity referred to in the next note. 2 Strabo, i. p. 47 (compare xvii. p. 829). 'll'E'll'lO'TGUKE 5% real 1repl 761/ {4350: orrnAc'ov 'HpaKAefwV 'noiUwTs ,uvtlu’adeo'i, Kept/1'71’ 're uiio'oz/ Ical itAAovs 'rdrrovs 5V0- ,uotgwu, rails undapoii uvvi Seucvv/le'vovs. The disappearance of such a settlement as Cerne is readily explained by the parallel case of Arguin, a barren island on the same line of coast, occupied by the Portuguese for a period of centuries, and which at one time carried on a GIIAP. XVI. ERA'I‘OSTHEN ES. 653 There can be no doubt that he conceived Africa to be sur- rounded by the sea to the south, in accordance with the notion prevalent in his time ; and which, as we have seen, was already held by Alexander himself. This much might be clearly in- ferred from his frequent use of the termflAtlantic, as applied to the Indian Ocean; but we are distinctly told that he regarded the whole surrounding ocean as continuous, so that the \Vestern Ocean and the Erythreean were parts of the same sea.3 But it does not appear that he attempted to support this theoretical conclusion by a reference to known facts, and we must there- fore infer that he attached no value to the alleged circumnavi-ii gation of Africa by order of N echof” J § 30. In the absence of the original work it is impossible for us to judge how far Eratosthenes entered into a detailed geographical account of the various countries of which he spoke: but as his whole treatise on geography was comprised in only three books, and the third of these alone was devoted to what can be strictly called geographical details, it is evident that any such review must have been very brief and summary. As far as the fragments preserved enable us to form a conjec- ture, it would appear that he certainly gave a brief enumeration of the leading geographical features of the coasts,5 as well as of the tribes of the interior; and he undoubtedly added some notices of the natural productions of the country, or the manners and character of the inhabitants, but it is probable that these were principally confined to the more remote and less known regions of the world. His object certainly appears to have been rather to present a general picture of the world as then known, than an elaborate geographical description of considerable trade, but is now utterly desolate and inhabited only by a few Arab fishermen. 3 Strabo, i, 3, § 13, p. 56. ml ‘yap rca'r’ ab'rbv ’Epa'roo'6é1/n 'rip/ e’rc'rbs dotAar- 'raV drraoav o'zippovxl Eli/at, 66076 real 1'7‘71/ 'Eo'rre’piov Kai. 'riw ’Epv6pax/ Gotham-av ,afav ell/at. 4 See Chapter VIII. sect. 2. 6 It was evidently in this part of his work that he made so much use of that of his predecessor Timosthenes, that he is accused by Marcianus of Heraclea of having plundered it bodily (Marcian. Epit. Artemidor. § 3. ’Eparoo'6éz/ns at 5 Kvpnuaios, of”: olda 'ri wadcbv, 'rb Ti/aoo- de'uovs ite'rel’yparlze BLBMOV, Bpaxe'a 'ru/d. 'irpoo'dels‘). See Chapter XV. p. 588. 654. HISTORY OF ANCIENT GEOGRAPHY. CHAP. XVI. the different countries of which it was composed. All such details could be easily filled in at a subsequent time, if the general outlines were correctly drawn. The'object of Eratos- thenes was to lay a secure foundation upon which succeeding geographers could build: and so successfully did he execute this task, considering the materials at his disposition, that the map of the world, as laid down by him, received scarcely any material improvement until the time of Ptolemy. § 31. Another part of the system of Eratosthenes, concerning which we have very imperfect information, is his division of the ‘iinhabited world into sections, to which he gave the strange idesignation of Sphragides (o'tppwyldes), or “ seals.” The origin of this term is unknown, and it was not adopted by any other geographer. Nor is it easy to discern what was the purpose of its introduction, or the nature of the division intended. These sections indeed had so far a relation to the fundamental parallel of latitude drawn by Eratosthenes in the manner already described, that they were all placed to the north or south of that line, so that it should form (in part at least) one of their boundaries; but in other respects they were very irregular, being neither conterminous with countries, nor with any geo- metrical divisions of the earth’s surface. The first section indeed comprised India only, and would therefore answer the first condition. The second also might be thought to fulfil the same purpose, though much less perfectly : it comprised Ariana, under which name Eratosthenes appears to have included (as was done by Strabo after him),6 Aria, Parthia, Drangiana, Arachosia, Gedrosia and Carmania, or all the provinces from the Indus to the frontiers of Media and Persia Proper. This section was bounded on the west by an imaginary line drawn from the Pylee Caspiae to the frontier of Carmania on the Persian Gulf, which Eratosthenes conceived as situated on the 6 Strabo, indeed, in the description had no better information (of) yap 5X0- of all these countries, their boundaries uéz/ 'rt Aé'yew Béivriou 'n'epi airrciw, xv. 2, and extent, distinctly follows the autho- § 8, p. 723). rity of Eratosthenes, adding that he CI-IAP. XVI. ERA TOSTHENES. 655 same meridian. , Hence the eastern and western boundaries would be approximately parallel,7 and the whole section presents the form, roughly taken, of a parallelogram. But the third section, which was conceived to be still bounded by the chain of the Taurus on the north (continued in a supposed straight line from the Caspian Gates), and by the imaginary line above described on the east, was limited on the west by a line drawn from the Armenian mountains (the prolongation of Mount Taurus) to Thapsacus on the Euphrates and then down that river to the Persian Gulf. Here we find no apparent connection with any national divisions; while the approach to any regularity of figure is of the most imperfect description. But this is_ still more the case with the fourth section, which extended from the Euphrates to the Mediterranean, and to the Isthmus of Suez on the west, while it was bounded on the south by an imaginary line across the deserts of Arabia from the head of the Red Sea to the Euphrates near Babylon. It thus presented something like a triangular form, having its base on the Mediterranean, and its vertex on the Euphrates. Unfortunately, while Strabo goes into a long and tedious discussion of the boundaries and dimen- sions of this section, he has omitted to continue the enumeration of these divisions, so that we are left wholly in the dark as to the nature and extent of the sections into which Eratosthenes subdivided Africa and Europe, as well as Asia north of the Taurus.8 In the absence of such information it is very difficult to see what geographical meaning Eratosthenes attached to the subdivisions in question, or what purpose they were intended to serve. § 32. On one subject, which occupied an important place in 7 Eratosthenes, as we have already seen, supposed the Indus to flow from N. to S., while the Ganges flowed from W. to E. For this he was censured by Hipparchus, who maintained that the Indus flowed towards the south-east (Strabo, ii. 1, § 34, p. 87). Its real course deviates considerably to the westward of south, so that its mouth, taking the centre of the Delta as such, is situated 471,9 of longitude to the west of Attoek. 8 It seems probable that the penin- sula of Arabia constituted the fifth Sphragis, but this is not stated by Strabo. 656 HISTORY OF ANCIENT GEOGRAPHY. GHAP. XVI. C the work of Eratosthenes, as well as in those of many of his successors, he certainly entertained sounder and more judicious views than most of the Greek geographers. The feeling of blind reverence for Homer and his works, which had grown up in Greece itself, but had attained to a still higher develop- ment in the schools of Alexandria, had led to a belief that his statements were to be received with deference as authorities not only in matters of history and mythology, but on questions of geography also. Nor was this confined to Greece itself and the lands immediately adjoining, where the accuracy of the long array of names exhibited in the Catalogue of the Ships, and the appropriateness of the descriptive epithets so often appended to them, were the subject of well-merited eulogy ;9 but the same confidence was reposed in the statements of the poet concerning the more remote localities described in the wanderings of Ulysses and Menelaiis, and even in his casual notices of the Ethiopians and Scythians.1 Eratosthenes on the contrary had the boldness to assert that while Homer was well acquainted with Greece and the regions near at hand, he was ignorant of those more remote :2 and that vhis narrative of the adventures of Ulysses, interwoven as it was with obvious fables, that no one could dream of understanding in their literal signification, was no more amenable to the test of geographical, than of historical truth. He appears to have 2 given especial offence by saying that people would never find out the real localities described in the Odyssey—the islands of olus, Circe, Calypso, &c., until they found out the cobbler ho had sewn up the bag of rZEolus.3 All these localities had, ong before the time of Eratosthenes, been identified with well- nown spots: and the inhabitants of the places thus pitched upon naturally clung with tenacity to the supposed traditions 9 See Chapter III. p. 42. rd o'zlue'y'yvs ,uduov, é’iorrep ’Eparoo"6é'uns 1 This tone, as has been already el'p'mce, Ital Ta e’u 'roi‘s "EAMqmv, ems ml remarked, pervades all the discussions 7631/ mippw woAAa M5761 and 81’ dIcpLBefas of Strabo in connection with this "Opmpos. subject. 3 Strabo, i. 2, § 15. 2 Strabo, i. 2, § 7, p. 18. ’AAA’ 055% \ CHAP. xvI. ERATOSTHENES. 65 7 that connected them with the works of the great poet. The grammarians and critics of Alexandria sought to support these pretensions by the most far-fetched interpretations, and had recourse to the most ingenious devices, rather than own that the object of their worship could have been ignorant of regions which in his day no Greek had ever visited; or could have ' given the reins to his poetic fancy, without troubling himself about geographical accuracy. Unfortunately we only know the views of Eratosthenes from the adverse criticisms of Strabo, who has undoubtedly in this, as in other instances, taken little pains to do justice to the arguments of his adversary : but it is clear that Eratosthenes maintained that the chief purpose of Homer, as of all other poets, was to delight and amuse, rather than to instruct, his hearers,4 and that he purposely transferred the fables of Circe and Calypso, of ZEolus and the gloomy land of the Cimmerians, to the boundaries of the Ocean and the unknown regions of the far west, that he might freely indulge his fancy, without the stern restraint of reality. In these general views Eratosthenes was much more in accordance with the judgement of most modern critics than were his contem- poraries or successors, of whom Strabo in particular attacks him with a vehemence worthy of an orthodox divine assailing a heretical commentator. But such sceptical criticisms evi- dently made little impression on the Greeks in general: even the grave and sober-minded Polybius adhered to the popular application of the fables, and sought to explain the tales of Bolus and Scylla, as if it was certain that Ulysses had really visited the Lipari Islands and the Straits of Messina.5 How far Eratosthenes carried his doubts we are unable to determine, in the absence of the original work: but it would appear that while doing full justice to the accuracy of Homer’s geography, as far as related to Greece and the neighbouring islands, he had discarded altogether the wanderings of Ulysses, as un- worthy of serious consideration in a geographical point of view. 4 Strabo, i. 2, § 3, p. 15. 6 Polyb. xxxiv. 2, 3, ap. Strab. i. pp. 23, 24. VOL. I. 2 U 658 HIs'roRY 0F ANCIENT GEoGRAPHY. can». XVI. He asserted also that the notices of Egypt in the Odyssey showed great ignorance of that country; and denied that Homer had any knowledge of the voyage of Jason and the Argonauts to the river Phasis.6 For both these assertions he is severely taken to task by Strabo, but on both points the verdict of any impartial reader of the Homeric poems will be in favour of Eratosthenes. § 33. Physical geography, in the modern sense of the term, was still quite in its infancy in the days of Eratosthenes, and it cannot be said that he did much to impart to it a scientific character. We have already seen indeed that in treating the mountain chains of Asia as one continuous range, to which he applied the name of Taurus, he may be regarded as \made a first ‘attempt, hos/avenues, ‘at thatwsystematic descrip: namariimaanaimasggsEasier* ‘wetow *greé'ui'étéae tr‘ Qrogra'phy. He had also, as already stated, arrived at a sound J conclusion concerning the causes of the inundation of theNile —a subject that must naturally have engaged the attention of a geographer resident in Egypt. On the other hand he started a strange hypothesis, that the surplus waters of the Euphrates were carried by subterranean channels to Coele Syria, and thence again underground so as to feed the streams which broke out near Rhinocorura and Mount Casius.7 § 34. Eratosthenes also adopted, and apparently developed at considerable length, an idea first suggested by the physical philosopher Strato,8 that the Mediterranean and the Euxine Seas had originally no outlet, and stood in consequence at a much higher level, but that they had burst the barriers that 6 In the last of these views he was followed by Demetrius of Scepsis (Strabo, i. 2, p. 45). The arguments of Strabo on the other side are confined to the repeated assertion that these were facts unicersallly admitted (T661! 8% amp) 7511 ’Ici.0'ova ovufiau'rwv Ital 'r'hu ’Ap'yz‘o Kai 'roils ’Ap'y0va15'ras 'ré‘w éuoho'yov- . ue'vwv wapd. wilo'w, l. 0.). 7 Eratosth. ap. Strab. xvi. 1, § 12, p. 741. Strabo only ventures to express a doubt concerning this extravagant suggestion (oz’m 0780; 6% ei mBaucBs e’z’pmcev). 8 Strato of Lampsacus, an eminent Peripatetic philosopher, who had de- voted his attention so particularly to the physical branches of philosophy as to be currently known by the surname of 6 gbvo'uais. He succeeded Theo- phrastus as the head of the Peripa- tetic school in B.0. 287. OHAP. xvI. ERATOSTHENES. 659 confined them, and thus given rise to the Straits of the Bos- phorus, the Hellespont and that of the Columns. In proof of this theory he alleged the presence of marine shells far inland in Libya, especially near the temple of Jupiter Ammon, and on the road leading to it, as well as the deposits and springs of salt that were also found in the Libyan deserts.9 Such a speculation has been frequently revived in more modern times ; 1 nor can it be said to be altogether without foundation, though it may safely be asserted that no such violent and sudden disruption as that supposed by Strato and Eratosthenes could have actually caused the formation of such straits as the Bosphorus and Dardanelles, or the Straits of Gibraltar. Nor can the occurrence of marine remains, regarded as a general phenomenon, be ascribed to the cause in question; but the most recent geological researches have all tended to establish the fact that the Libyan desert, as well as a great part of the still more extensive desert of the Sahara, was covered by the sea at a comparatively recent period. It is remarkable that one of the arguments brought forward by Strato in support of this hypothesis was the existence of a submarine ridge or bank, extending across the Straits from Europe to Africa, by which, as he contended, the two had been originally united? Such a bank really exists, though at so considerable a depth as to render it remarkable that its existence should have been discovered by ancient navigators.3 At the same time both Strato and Eratosthenes were aware of the great depth of certain portions of the Mediterranean, especially the Sicilian and Sardinian seas—as compared with other parts of the same great basin.‘ Sta'ré'raxev d'rrb 'riis Eilpfié'll'flS €1rl 79W 9 Strabo, i. 3, pp. 49, 50. AzBdnv. Strabo, i. p. 50. 1 See Admiral Smyth’s Mediter- ranean, p. 114-122. The same theory has been fully developed by M. Bureau de la Malle in his work entitled Géo- graphic Physique de la Mer Noire, de l’I/nte'rieur ole l’Afrique, et de la Médi- terrane’e, 8V0, Paris, 1807. 2 31': Ital 1161/ in 'rawia 'ns i5¢a2los 3 Smyth’s Mediterranean, pp.l 59,160. 4 Strabo, i. 3, p. 50. Posidonius reported the Sardinian Sea to be the deepest of all, and to attain to a depth of 1000 fathoms. Id. p. 54. Such a statement must, however, have been merely conjectural. 66o HISTORY OF ANGIENT GEOGRAPHY. CHAP. xvI. § 35. We learn also from Strabo that Eratosthenes, in order to prove that the world, though spherical, was not exactly so, entered into a long enumeration of the changes of its ‘surface produced by the action of water, of fire, of earthquakes, volcanic eruptions, and other similar causes.5 Though Strabo is un- doubtedly right in regarding the effect of all such operations as insignificant in relation to the figure of the whole earth, it would have been very interesting to have known in detail the facts cited by Eratosthenes, and the view which he took of their results. But here, as in so many other cases, we unfor- tunately know his arguments only through the criticisms of his adversaries. We learn however that he was of opinion that the highest mountains did not exceed ten stadia (6000 feet) in perpendicular altitude,“ a conclusion probably based on the measurements of Dicaearchus already noticed. 5 Strab. ib. p. 49. (ad Arist. de Caalo, ii. p. 136) by Bern- 6 See the fragments cited from Theon hardy in his Eratosthenica, fr. 39, p. 56. Alexandrinus (p. 23) and Simplicius ' arena MAP or THE WORLD ACCORDING To ERATosTHENEs Thule ........ -. r r - e l C‘, ‘EU Q ‘t Win" 6 i=1 {Es-s B- Truman. Y)“ S ilv a i Q Hi (3 ‘4 F H. 19 X _ :1- .i’ *_ r M i _ :"v—il? ~~ -~ _ - - at”! ' s'tunu a7 , _ ,_ ,_._.___. 7. V p, Q) i V V‘ _ t . . I _B H 1. ell? ‘ ___;_.'_ _ '“'_' , j 3’ .' I‘ . “T: ___:. , .. g- “4 .i' I__.-__ ' ' 1.1-] q ' l "' ' -- ’/// A; \'-)ll\\ImH“\/|“‘\‘r\ \§' (Q _-_~ 0 ~ J: — - I J -. .. '/u,\._\.'.i _ p w ‘ I‘ §\~ ‘Mi/w . _Z ‘ \‘fml ‘.illlflfz‘, ‘II/7.1 al'h" ' ' A" 'ffglf‘hjjjrj ‘1.5.1.1,’ I. I Nib/‘qui,- \‘.l: /,/ .-\.\\_\_(u _il H’flm ;\}\\ 7" a ,- y.’ __ - ‘H/"zb. ‘NWT/A“ '7’ “WV/fl ‘ 1'//'"'1 .1; _ _ __ _ ;: "7 '—_ _ ‘ W _ . - , “at. llslvlljll‘lllgvi , -, ,, ,...,-,'-,',W.,,/,, ,,7,/,;,,,,7;:;,v>/, , W; V (0/ Mi’iil’lrr‘ll/Ik- . {@171 IVH‘WIIZj§6mlfl€iizlhz%'"agéi "WW?illtftflfljzihuwfiimhj‘(/@\‘ f g -::'_—i__ —T ‘ IJ/V ~ “A (6‘? ,Q/U “.V/I/("M "71; '3 "\5 '- . z ' 7 " Warn/z.‘ ‘.31., " lv‘ljl‘jQM/r: -- '\°1‘-l\l_lq,\\\\\_l ltd-"a WWI/Ssh“ _I/fil/I/ll/élljlHll/jiillflllilmm/(/f,\\\ I’ ’_ ,////Iy/je ‘ —:f~ -- __ _q; in — _» | 1 A "(my/1',’ . \WhbmxQj \gw/flfijmfihiiw / ,1 , - ‘ * y 1' w (I) y g. s y,‘ r '- » \ ~~ .. ////// _ _ Ff 4;: _— rat-tilt 'l" \ "” r “ ‘\ 6))‘ “ " '< . hark; . tartan: .lihiill’ttitt’iht .. ._ H j ,7 W. _, ,<.,,, . .2. v I,’ i . _ ‘a, S I ~ Arachosia -% A Palibofizra° Gan ‘es R. Parallel of . l Carthage * .3 O cxandria My ‘ "\I = Susie , .- l'hebes P81 Sis l‘iil ill‘ Tropic I A N A y , ill . r . ,t, ‘Till?! 1'." ‘111' | l l S (a1. 1 Ichthyophagi MM“. _ . - _“mwnzscd 4 ,a "—T P 61 Of ‘15:’ (p 01306 ________ _. _ . . :1 _. eroc' I: i. ‘l i a T‘ w E p (N l l l | l l l ||ll I ll"l till ill _, 1' v S ~ " S 61 11' it ‘ c l r1 6% A —-.e ofitnglllwdfia _ ‘S i w rill 5357-1 s; I! 2‘ ii at l :1‘ l ‘I! | ~ .0: Q ~. .‘s HIIJIH l l I Ill l 3,. S’ 4& ll l l , it it! 'l is in 0 (‘I é ‘.O 0 1I lit , I It ll 1 ill!‘ " ll‘ will a i it... till“) ’ ' I C . Equafnv . ——--—~ —_-___4 ~_f * .__. .‘___ _ A. i F: _—_ . ___T_‘A__‘—_VA ST A D l A 2.000 3.000 8,000 13,500 Sumf'or-d's Geograplrl'Esrabr Meridian ‘of 6.300 10.000 14.000 16,000 3000 2,000 exanflna. _ London -. John Mun? ay. NOTE A.. ERATOSTHENES. 661 NOTE A, p. 632. OBSERVATIONS OF LATITUDE. AMONG the few points of which we know with certainty that their latitudes had been determined by direct observation were Massilia and Byzantium : and from the time of Hipparchus onwards it became a received fact among geographers that they were in the same parallel of latitude, though there is really a difference of more than two degrees between them. Yet we are distinctly told by Strabo that Hipparchus himself repaired to Byzantium for the pur- pose of observing its latitude, and found his gnomonic observations to coincide with that of Pytheas at Massilia. (Strabo, i. 4, .~ f- was almost precisely correct. Hence Eratosthenes anhis suc- cessors who accepted the conclusion of Pytheas placedgMassilia very nearly in its true position, which rendered their mil‘p of the western Mediterranean much more correct than that of Strabo: while on the other hand Byzantium was pushed up far to the north of its true situation, and hence the map of the adjoining regions became distorted to correspond with it. Again, we know that Eratosthenes himself made corresponding observations of latitude at Rhodes and Alexandria, (Strjbo, ii. p. 126) and the astronomical result of these observations ‘as very nearly correct, though he was led into error in computing the distance between the two by his erroneous estimate of 7 00 stadia to a degree. But at the same time he placed a point so well known as Athens only 4.00 stadia (40 minutes) to the north of Rhodes, while the difference between the two really exceeds a degree and a half. On the other hand Eratosthenes possessed a determination of unusual accuracy for the latitude of Meroe, which he placed within less than half a degree of its true position. For this observation he was indebted to a certain Philon, who had himself made avoyage up the Nile to Ethiopia, where he had observed both the proportion of the gnomon to its shadow, and the number of days (45) before the summer solstice, when the sun became vertical. (Strabo, ii. 1, § 29, p. 77.) 662 HISTORY OF ANCIENT GEOGRAPHY. CHAP. XVI. We have no account of the authority on which the Straits of Gibraltar were placed in the same latitude with Rhodes: but it was a fortunate circumstance that these two extreme points should have been so correctly assigned, while intermediate positions, such as Carthage and the Straits of Messina, were placed so wide of their true position. In the case of Carthage, as already observed, the erroneous latitude assigned to it was based, or supposed to be based, upon direct observation. Thus Strabo assumes, or refers to it as a well- known fact, though without mentioning his authority, that the gnomon there was to its shadow as eleven to seven (ii. 5, § 38, p. 133), a proportion which would correspond with 32° 30’ N. latitude, about 4% degrees south of its true position! NOTE B, p. 635. GOssELLm’s THEORY. According to M. Gossellin indeed the result attained by Eratosthenes was one of surprising accuracy. Assuming that the latter was employing stadia of 7 00 to a degree, and allowing for the proper reduction of degrees of longitude along the parallel of 36° latitude, he arrives at the remarkable result that the length of the Mediterranean as given by him was within 1° 22’, or less than a degree and a half, of the truth. (Geographic des Grecs analysée, p. 40.) So striking is the exactness of this coincidence, that M. Gossellin himself observes it cannot be ascribed to Eratos- thenes himself, as it proves an amount of mathematical skill and knowledge far exceeding that possessed by any Greek in his day. Hence he is driven to the very extraordinary hypothesis that Eratosthenes derived his information from materials which he found in the Alexandrian library, and which proceeded from an ancient people who possessed the skill and knowledge of astronomy and mathematical geography attested by such a result. These materials Eratosthenes made use of without understanding them (l): hence he fell into errors which require to be explained and rectified before we can do justice to the value of his authorities. (Ibid. pp. 43-50.) This far-fetched hypothesis may justly be discarded without NOTE B. ERATOSTHENES. 663 further investigation, as resting upon no substantial basis whatever. It has however been adopted by M. Bureau de la Malle (Géogr. Physique, &c. p. 147 But it must be observed that almost all the remarkable instances of close conformity with the truth produced by M. Gossellin are obtained by more or less altering the data as given by Strabo, or by introducing arbitrary assumptions of his own. Thus when he proceeds to reckon the distance from the Gulf of Issus eastward to the mouth of the Ganges—a distance which we know Eratosthenes to have computed from the itinerary measures of the marches of Alexander, and other similar materials (Strabo), and which comes out in M. Gossellin’s hands almost as exact as that from the Sacred Cape to the Gulf of Issus, this coincidence .results in great part from his reducing the number of itinerary stades by one-tenth, and by his arbitrarily fixing on Chandernagore —-which is more than 50 miles from the sea and not on the Ganges at all—as the point to compare with the supposed mouth of the Ganges, conceived by Eratosthenes as situated on the Eastern Ocean. M. Gossellin himself remarks that all the intermediate distances given by Eratosthenes are inexact, though the great distances (the sums total) are, or ought to be considered as correct (p. 45). The first admission is no doubt true, but the second is only attained by a series of arbitrary changes and alterations of numbers, which are made with the express purpose of bringing about a correct result. We have seen that Eratosthenes in computing 700 stades to a degree of a great circle was not, as supposed by M. Gossellin (as well as D’Anville and many other modern geographers) employing a difl‘erent stade from that familiar to all the Greeks, but was adopt- ing an erroneous calculation of the length of a degree. Hence the assumption that, in computing distances such as those along the line of the Mediterranean, or the prolongation of it across Asia, Eratosthenes was throughout reckoning by stades of 7 00 to a degree, is wholly without foundation. In regard to the first indeed it is most probable that he was following Timosthenes,'from whom we know him to have borrowed largely [see Chapter XV. p. 588] ; while the latter were undoubtedly taken either from Patrocles or from the authors of the itineraries of Alexander. Yet all these authors wrote before any one had heard of stades of 700 to a degree and were merely using the ordinary Greek stade as univer- sally received. It must be added that though Strabo was well 664 HISTORY OF ANCIENT GEOGRAPHY. CHAP. xvI. acquainted with the measurement of the earth’s circumference by Eratosthenes, and with the resulting value of its subdivisions, he has nowhere given the slightest hint, or evidently entertained the least suspicion, that that author in his numerous statements of distances was employing any other stade than that in general use among all the Greeks. NOTE C, p. 639. SOUTHERN LIMIT OF THE HABITABLE WORLD. The calculation by which this result is arrived at is as follows. We know from the astronomical writers, Ptolemy and his com- mentator Theon, that Eratosthenes fixed the interval between the tropics at 11 / 83ds of the whole meridian circle,7 which would place the tropic of Cancer in N. latitude 23° 51’ 20". This would give in round numbers 16,700 stadia between the tropic and the equator (at 700 stadia to the degree), and as Eratosthenes had reckoned 5000 stadia from Syene to Meroe, and 3400 from Meroe to the Land of Cinnamon, there remain 8300 between this limit of the habitable world and the equator. This conclusion, which is adopted both by M. Gossellin and Dr. C. Muller, is confirmed by the fact that the distances given by Eratosthenes, on this calculation, result in placing Thule at a dis- tance of 46,300 stadia from the Equator, or 16,7 00 from the pole: exactly the same distance as the tropic from the Equator: and this precisely accords with the statement of Eratosthenes himself (de- rived from Pytheas) that at Thule “ the summer tropic coincides with the arctic circle ” (napa of; 6 afrrcis éo'n 11:5 dpK'rLKdi 5 depu/Os 'rpo'n'tKOs miKhog. Strabo, ii. 5, § 8, p. 114), or according to the modern use of the expression, that it was situated under the Arctic Circle. The distances given thus make up exactly the 63,000 stadia re- quired for the quadrant of a meridian great circle, in accordance with the 252,000 stadia adopted by Eratosthenes as the circum- ference of the globe. 7 Ptolemaei Illagna Syntaais, i. 10, (Tara, dis drcpzBc'bs eiAnupél/q), Kai 'ydp 6 p.18. The statement is given _some- ’Epa'roo'0é1ms Eiaipnc-as 'rbv 6M1: mixAov what msre cleifilly by ThGZODGIOH) 111? 1r)’ eil'plo'lce 'r'illl 1161-0061‘! T6511 '1'P01rmc'6v r6311 Oommen my CD. 6 passage p. ,Ka : a __ :v ¢ _ '_' I // ofi'ros 15 A6703 5 az’i'rbs o'xedbv ‘rd 'roi; mil-w” ‘in-Kai 6g“, 00° 7° mobs F'giu'B f‘ ’Epa'r00'6év0vs, (25 real 6 “Imrapxos éxp'r'l- W70” ""7 "P53 “1' NOTE D. ERATOSTHENES. 665 There remains indeed one difficulty which we have no means of solving, that Strabo, who in all that relates to mathematical geography generally follows the authority of Eratosthenes, and in this very passage distinctly refers to his calculations, has placed the parallel of the Region of Cinnamon (which he also assumes as the limit of the habitable world) at 8800 stadia from the Equator, without any indication of his difference from Eratosthenes, or of his reasons for the alteration. (Strabo, ii. 2, § 2, p. 95.) But as he at the same time reduces the distance from Meroe' to this extreme limit to 3000 stadia, instead of the 3400 allowed by Eratosthenes, the resulting difference is in great measure neutralised. The grounds of both changes remain equally unexplained. NOTE D, p. 639. DISTANCE FROM ALEXANDRIA TO RHODES. This we are distinctly told by Strabo (ii. p. 126): aim-()9 8% 8a) 7631/ o'KwQnpLKdW 'yverudvwv dvevpefv TpLo'XLM'ovq érr'raKoo't'ovs 7T€VT’Y;KOVTO.. But it must be observed that Eratosthenes must here have been led into error by his own previous calculation that there were 700 stadia to a degree. For in this case he had no means (if measuring the dis- tance—having himself rejected the estimates of navigators—and therefore all he could do was to ascertain by his gnomon the differ- ence of latitude, and convert this into stades at the rate which he had obtained from his supposed measurement of the are between Syene and Alexandria. This is Well explained by Col. Leake (Disputed Questions of Ancient Geography, p. 92), and is important to bear in mind, as we shall find that Posidonius was led to an erroneous computation of the earth’s circumference by supposing this are between Rhodes and Alexandria to have been actually measured by Eratosthenes. The result of this error would naturally be that Eratosthenes (believing his gnomonic observations to be approximately correct, as was really the case) would overrate the distance; and this we find to be the case, for though his computation gave a material reduction upon the estimateof the navigators, it was still‘con- siderably in excess of the truth—the real distance between Rhodes and Alexandria being only about 330 G. miles, or 3300 Greek VOL. I. 2 x 666 HISTORY OF ANGIENT GEOGRAPHY. CHAP. XVI. stadia instead of 37 50 (Leake, l. 0.). Yet we afterwards find Strabo and other later writers reverting to the original rough estimate of 4000 stadia; so little reliance did they place upon the mathematical calculations of scientific geographers like Eratosthenes ! NOTE E, p. 644. PROMONTORY OF TAMARUS. Strabo, xi. 11, § 7, p. 519. This name of Tamarus is found only incidentally in this single passage. It is difficult to conceive whence Eratosthenes could have derived it, as the promontory in question had no real existence, but was a geographical fiction, or rather inference, that the chain of Imaus must end in some such headland. The name is however again found under the form Tamus, in Pomponius Mela, who describes it as the eastern extremity of the chain of Mt. Taurus (iii. 7, § 68). The Tabis of Pliny, which he calls “jugum incubans mari quod vocant Tabin ” (vi. 17, § 53) is evidently only a corruption of the same name : though it has been strangely identified by Mr. Clements Markham with the northern point of Siberia, a' country of which the existence was as utterly unknown to Pliny as that of America. It was here—if anywhere—that we should have expected to find mention of Thinae, a name so strangely introduced into all our editions of Strabo down to a very recent period. (See note to p. 630 of this chapter.) Yet the name is wholly wanting where it would be appropriately found, as marking the termination of the principal parallel of latitude on the Indian Ocean, though it was introduced repeatedly, according to the old reading, to designate the very parallel in question. END OF VOL. I. Losnox: PRINTED BY WILLIAM oLowEs AND sons, STAMFORD STREET AND cnARmG CROSS. ' DD NOT REMOVE OR MUTILATE CARD “'l_l§. . . . _ _ '*"i":‘exqs ,,-_ *fliDwIquQxpm.“ 'W‘wbmuskaawh, ,, ‘ g A! ' 1 ‘I ‘ NU I A. .I‘ Q'- ~ *1 kpj'su A at’ a.‘ ‘ ~ .My'. I ‘ 6"’. '.“‘:_F°" ‘a n "1': ,s n 5b,,‘ .1 v ‘ ‘ . i Q‘ ..,, 3. P';'fl_-.__‘ , ‘xi, I“ I‘... 1J6!‘ w "in ‘' ~ . N’bw‘01‘ ‘7nd in..." n‘ , ..av\...‘ ' "l~|‘-“.‘.‘A‘€"VI .. . 1 q ‘ I ‘Q 0G I \ 7"‘ . I i i " i ., “M m ‘ I“ 1 wflfibpwa Q ' a. J‘ k ‘a h v i " ‘- i In ‘I in)’ 1'1‘- 3 ~\ I‘ ."“‘JQ“‘ V‘ , “ ' v . ’ “Ml--