majority of subsequent geographers, from Ptolemy. Subsequent to this date, there are few works which fall within the range of classical geography, or which have added any information of real value on the subject of India; while most of them have borrowed from Ptolemy, whose comprehensive work was soon a text-book in the hands of learned men. From Agathemerus (at the end of the second century) and Dionysius Periegetes (towards the end of the third century) some few particulars may be gleaned: -as for instance, from the latter, the establishment of the Indo-Scythi along the banks of the Indus, in Scinde and Guzerat; and, from a work known by the name of Periplus Maris Erythraei (the date of which, though late, is not certainly determined), some interesting notices of the shores of the Indian Ocean. Festus Avienus, whose paraphrase of Dionysius Periegetes supplies some lacunae in other parts of his work, adds nothing of interest to his metrical account of Indian Geography.

Such may serve as a concise outline of the progress of knowledge in ancient times relative to India. Before, however, we proceed to describe the country itself under the various heads of mountains, rivers, provinces, and cities, it will be well to say a few words on the origin of the name INDIA, with some notice of the subdivisions which were in use among the earlier geographers, but which we have not thought it convenient in this place to perpetuate.

The names INDUS, INDIA, are no doubt derived from the Sanscrit appellation of the river, Sindhu, which, in the plural form, means also the people who dwelt along its banks. The adjoining countries have adopted this name, with slight modifications: thus, Hendu is the form in the Zend or old Persian, Hoddu in the Hebrew (Esther, i. 1, viii. 9). The Greek language softened down the word by omitting the h, hence "Ivdos, "Ivdia; though in some instances the native name was preserved almost unchanged, as in the xiveos of the Periplus Maris Erythraei. Pliny bears testimony to the native form, when he says, "Indus incolis Sindus appellatus" (vi. 20. s. 23).

The great divisions of India which have been usually adopted are those of Ptolemy (vii. 1. § 1), into,—(1) India intra Gangem, a vast district, which was bounded, according to that geographer, on the W. by the Paropamisadae, Arachosia, and Gedrosia; on the N. by the Imaus, in the direction of the Sogdiani and Sacae; on the E. by the Ganges, and on the S. by a part of the Indian Ocean: and (2) India extra Gangem (Ptol. vii. 2. § 1), which was bounded on the W. by the Ganges; on the N. by Scythia and Serica; on the E. by the Sinae, and by a line extended from their country to the Μεγάλος κόλπος (Gulf of Siam); and on the S. by the Indian Ocean, and a line drawn from the island of Menuthias (Ptol. vii. 2. §1), whence it appears that Ptolemy considered that the Ganges flowed nearly due N. and S. We have considered that this division is too arbitrary to be adopted here; we merely state it as the one proposed by Ptolemy and long current among geographers. The later ecclesiastical writers made use of other terms, as ʼn évôóтepw "Ivdia, in which they included even Arabia (Socrat. H. E. i. 19; Theod. i. 23; Theoph. i. 35), and ʼn éoxárn Lydia (Sozomen, ii. 23).

The principal mountains of India (considered as a whole) were:-t -the eastern portion of the Paropamisus (or Hindú-Kush), the Imaus (Haimava), and the Emodus (now known by the generic name of the Himalaya.) To the extreme E. were the Montes

Semanthini, the boundary of the land of the Sinae, the Montes Damassi, and the Bepyrrhus M. (probably the present Naraka M.). An extension of the M. Damassi is the Maeandrus M. (now Muin-Mura). In India intra Gangem Ptolemy mentions many mountains, the names of which can with difficulty be supplied with their modern representatives: as the Orudii M., in the S. extremity of the land between the Tyndis and the Chaberus; the Uxentus M., to the N. of them; the Adisathrus M.; the Bittigo M. (probably the range now known as the Ghats), and the M. Vindius (unquestionably the present Vindhya), which extend NE. and SW. along the N. bank of the Nerbudda; M. Sardonix (probably the present Sautpura); and M. Apocopa (perhaps the present Aravelli).

The principal promontories in India are:- in the extreme E., Promontorium Magnum, the western side of the Sinus Magnus; Malaei Colon, on the S. coast of the golden peninsula; Promontorium Aureae Chersonesi, the southern termination of the Sinus Sabaracus, on the western side of the Chersonesus; Cory or Calligicum, between the S. Argaricus and the S. Colchicus, near the SW. end of the peninsula of Hindostán; Comaria (now C. Comorin), the most southern point of Hindostán; Calae Carias (or Callicaris), between the towns Anamagara and Muziris; Simylla (or Semylla, the southern end of the S. Barygazenus, perhaps the present C. St. John), and Maleum.

In the same direction from E. to W. are the following gulfs and bays:- the Sinus Magnus (now Gulf of Siam); S. Perimulicus, and Sabaricus, on the E. and W. side of the Chersonesus Aurea; S. Gangeticus (Bay of Bengal), S. Argaricus, opposite the N. end of Taprobane (probably Palks Bay); S. Colchicus (Bay of Manaar); S. Barygazenus (Gulf of Cambay), and S. Canthi (most likely the Gulf of Cutch).

The rivers of India are very numerous, and many of them of great size. The most important (from E. to W.) are the Dorias (Salven ?) and Doanas (the Irrawaddy), the Chrysoana, Besynga, the Tocosanna (probably the present Arrakan), and the Catabeda (now Curmsul); the Ganges, with many tributaries, themselves large rivers. [GANGES.] Along the W. side of the Bay of Bengal are the Adamas (Brahmini), Dosaron (Mahanádi), Maesolus (Godávári), Tyndis (Kistna), and the Chaberis or Chaberus (the Caveri). Along the shores of the Indian Ocean are the Nanaguna (Tarty), the Namadus (Narmadá or Nerbudda), and lastly the Indus, with its several tributaries. [INDUS.]

The towns in India known to the ancients were very numerous; yet it is remarkable that but few details have been given concerning them in the different authors of whose works fragments still remain. Generally, these writers seem to have been content with a simple list of the names, adding, in some instances, that such a place was an important mart for commerce. The probability is, that, even so late as Ptolemy, few cities had reached sufficient importance to command the productions of an extensive surrounding country; and that, in fact, with one or two exceptions, the towns which he and others enumerate were little more than the head places of small districts, and in no sense capitals of great empires, such as Ghazna, Delhi, and Calcutta have become in later periods of Indian history. Beginning from the extreme E., the principal states and towns mentioned in the ancient writers are: Perimula

on the E. coast of the Golden Chersonesus (in the neighbourhood of Malacca); Tacola (perhaps Tavai or Tavoy); Triglyphon, in the district of the Cyrrhadiae, at the mouth of the Brahmaputra (now Tiperah or Tripura); and Cattigara, the exact position of which has been much disputed among geographers, but which Lassen has placed conjecturally in Borneo. Northward of Triglyphon are a number of small districts, about which nothing certain is known, as Chalcitis, Basanarae, Cacobae, and Aminachae, the Indraprathae, and Iberingae; and to the W., along the swamp-land at the foot of the Himalaya chain, are the Tiladae, Passalae, Coraucali, and the Tacaraei. All the above may be considered as belonging to India extra Gangem.

Again, from the line of coast from E. to W., the first people along the western mouths of the Ganges are called the Gangaridae, with their chief town Gange (in the neighbourhood of the modern Calcutta); the Calingae, with their chief towns Parthalis and Dandagula (the latter probably Calinapattana, about halfway between Mahanadi and Godavari); the Maesoli and Maesolia, occupying nearly the same range of coast as that now called the Circars, with the capital Pitynda, and Contacossyla (Masulipattana ?) and Alosygna on the seacoast; W. of the Maesolus (Godávari), the Arvarni, with the chief town Malanga (probably Mandarágja, the present Madras). Then follow the Soringi and Bati, till we come to the land of Pandion (Пavdiovos xúpa), which extends to the southern extremity of the peninsula of Hindustán, and was a district of great wealth and importance at the time of the Periplus. (Peripl. pp. 31, 33.) There can be no doubt that the land of Pandion is the same as the Indian Pandja, and its capital Modura the present Mathura. Within the same district were Argara (whence the S. Argaricus derives its name), the Carci, and the Colchi. At the SW. end of the peninsula were Cottiara (Cochin), and Comaria, whence the promontory Comorin derives its name. Following the western coast, we arrive at Limyrica (Peripl. pp. 30, 36), undoubtedly in the neighbourhood of Mangalore, with its chief towns Carura (most likely Coimbatore, where a great quantity of Roman coins have been dug up during the last fifteen years) and Tyndis (in the neighbourhood of Goa); and then Musopale, Nitrae, and Mandagara; all places on the sea-coast, or at no great distance from it. Somewhat further inland, within the district known generically at the time of the Periplus by the name of Dachinabades (Dakhinabháda, or Deccan), was the district of Ariaca ('Apíana Zadavwv, Ptol. vii. 1. §§ 6, 82; cf. Peripl. p. 30), with its chief town Hippocura (Nandira or Hydrabad, if not, as Ritter has imagined, the sea-port Mangalore); Baetana, Simylla (on the coast near Bassein), Omenagara (undoubtedly the celebrated fortress Ahmed-nagar), and Tagara (Peripl. p. 19), the present Deoghir. Further N., the rich commercial state of Larice appears to have extended from the Namadus (Narmadá or Nerbudda) to Barygaza (Beroach) and the Gulf of Cambay. Its chief town was, in Ptolemy's time, Ozene (Oujein or Ujjayini), a place well known to the antiquaries of India for the vast numbers of the earliest Indian coinage constantly found among its ruins; Minnagara, the position of which is doubtful, and Barygaza, the chief emporium of the commerce of Western India. North of Larice was Syrastrene (Saurashtran), to the west of the Gulf of Cambay; and still further to the westward, at the mouths of

the Indus, Pattalene (Lower Scinde, and the neighbourhood of Kuráchi), with its capital Pattala (Pótala.)

It is much more difficult to determine the exact site of the various tribes and nations mentioned in ancient authors as existing in the interior of the country, than it is to ascertain the corresponding modern localities of those which occupied the seacoast. Some, however, of them can be made out with sufficient certainty, by comparison of their classical names with the Sanscrit records, and in some instances with the modern native appellations. Following, then, the course of the Indus northwards, we find, at least in the times of Ptolemy and- of the Periplus, a wide-spread race of Scythian origin, occupying both banks of the river, in a district called, from them, INDO-SCYTHIA. The exact limits of their country cannot now be traced; but it is probable that they extended from Pattalene on the S. as far as the lower ranges of the Hindu-Kush, - in fact, that their empire swayed over the whole of modern Scinde and the Panjáb; a view which is borne out by the extensive remains of their Topes and coinage, which are found throughout these districts, and especially to the northward, near the head waters of the three western of the Five Rivers. A great change had no doubt taken place by the successful invasion of a great horde of Scythians towards the close of the second century B. C., as they are known to have overthrown the Greek kingdom of Bactriana, at the same time effacing many of the names of the tribes whom Alexander had met with two centuries before, such as the Aspasii, Assaceni, Massiani, Hippasii; with the towns of Acadera, Daedala, Massaga, and Embolima, which are preserved in Arrian, and others of Alexander's historians.

Further N., along the bases of the Paropamisus, Imaus, and Emodus, in the direction from W. to E., we find mention of the Sampatae, the district Suastene (now Sewud), and Goryaea, with the towns Gorya and Dionysopolis, or Nagara (now Nagar); and further E., between the Suastus and the Indus, the Gandarae (one, doubtless, of the original seats of the Gandháras). Following the mountain range to the E., we come to Caspiria (now Cashmir, in earlier times known, as we have seen, to Herodotus, under the name of Caspatyrus). Southward of Cashmir was the territory of Varsa, with its capital Taxila, a place of importance so early as the time of Alexander (Arrian, v. 8), and probably indicated now by the extensive remains of Manikyala (Burnes, Travels, vol. i. p. 65), if, indeed, these are not too much to the eastward. A little further S. was the land of Pandous (Hardwov xúpa, doubtless the representative of one of the Pandava dynasties of early Hindú history), during the time of Alexander the territory of the king Porus. Further eastward were the state Cylindrine, with the sources of the Sutledge, Jumna, and Ganges; and the Gangani, whose territory extended into the highest range of the Himalaya.

Many small states and towns are mentioned in the historians of Alexander's campaigns along the upper Panjab, which we cannot here do more than glance at, as Peucelaotis (Puskkalávatí), Nicaea, Bucephala, the Glaucanitae, and the Sibae or Sibi. Following next the course of the Ganges, we meet with the Daetichae, the Nanichae, Prasiaca; and the Mandalae, with its celebrated capital Palibothra (beyond all doubt the present Pataliputra, or Patna), situated at the junction of

the Erannoboas (Hiranjávaha) and the Ganges; with some smaller states, as the Surasenae, and the towns Methora and Clisobra, which were subject to the Prasii. Southward from Palibothra, in the interior of the plain country, dwelt the Cocconagae, on the banks of the Adamas, the Sabârae, the Salaceni, the Drillophyllitae, the Adeisathri, with their capital Sagida (probably the present Sohagpur), situated on the northern spurs of the Vindhya, at no great distance from the sources of the Sonus. Between the Sonus and the Ganges were the Bolingae. In a NW. direction, beyond the Sonus and the Vindhya, we find a territory called Sandrabatis, and the Gymnosophistae, who appear to have occupied the country now called Sirhind, as far as the river Sutledge. The Caspeiraei (at least in the time of Ptolemy; see Ptol. vii. 1. § 47) seem to have extended over a considerable breadth of country, as their sacred town Modura (Módoupa Tv Dewv) was situated, apparently, at no great distance from the Nerbudda, though its exact position has not been identified. The difficulty of identification is much, indeed, increased by the error of reckoning which prevails throughout Ptolemy, who held that the coast of India towards the Indian Ocean was in a straight line E. and W. from Taprobane and the Indus, thereby placing Nanaguna and the Namadus in the same parallel of latitude. On the southern spurs of the Vindhya, between the Namadus and Nanaguna, on the edge of the Deccan, were the Phyllitae and Gondali; and to the E. of them, between the Bittigo M. and the river Chaberus (Cáveri), the nomad Sorae (Zŵpaι voμádes), with a chief town Sora, at the eastern end of M. Bittigo. To the southward of these, on the Chaberus and Solen, were several smaller tribes, the Brachmani Magi, the Ambastae, Bettigi or Bitti, and the Tabassi.

All the above-mentioned districts and towns of any importance are more fully described under their respective names.

The ancients appear to have known but little of the islands which are now considered to form part of the East Indies, with the exception of Taprobane or Ceylon, of which Pliny and Ptolemy have left some considerable notices. The reason is, that it was not till a much later period of the world's history that the Indian Archipelago was fully opened out by its commercial resources to scientific inquiry. Besides Ceylon, however, Ptolemy mentions, in its neighbourhood, a remarkable cluster of small islands, doubtless (as we have remarked before) those now known as the Laccadives and Maldives; the island of Iabadius (Java), below the Chersonesus Aurea; and the Satyrorum Insulae, on the same parallel with the S. end of this Chersonesus, which may perhaps answer to the Anamba or Natuna islands.

Of the government of India, considered as a whole, comparatively little was known to the Greek writers; indeed, with the exception of occasional names of kings, it may be asserted that they knew nothing E. of Palibothra. Nor is this strange; direct connection with the interior of the country ceased with the fall of the Graeco-Bactrian empire; from that period almost all the information about India which found its way to the nations of the West was derived from the merchants and others, who made voyages to the different out-ports of the country. It may be worth while to state briefly here some of the principal rulers mentioned by the Greek and Roman writers; premising that, previous to the advance of Alexander, history is on these subjects

silent. Previous, indeed, to Alexander, we have nothing on which we can rely. There is no evidence that Darius himself invaded any part of India, though a portion of the NW. provinces of Bactria may have paid him tribute, as stated by Herodotus. The expeditions of Dionysus and Hercules, and the wars of Sesostris and Semiramis in India, can be considered as nothing more than fables too credulously recorded by Ctesias. At the time of the invasion of Alexander the Great, there can be no doubt that there was a settled monarchy in the western part of India, and his dealings with it are very clearly to be made out. In the north of the Panjab was the town or district Taxila (probably Manikyála, or very near it), which was ruled by a king named Taxiles; it being a frequent Indian custom to name the king from the place he ruled over. His name in Diodorus is Mophis (xvii. 86), and in Curtius, Omphis (viii. 12), which was probably the real one, and is itself of Indian origin. It appears that Alexander left his country as he found it. (Strab. xv. pp. 698, 699, 716.) The name of Taxiles is not mentioned in any Indian author. The next ruler Alexander met with was Porus (probably Paurava Sanser., a change which Strabo indicates in that of Aapiaúnν into Aapeîov), with whom Taxiles had been at war. (Arrian, v. 21.) Alexander appears to have succeeded in reconciling them, and to have increased the empire of Porus, so as to make his rule comprehend the whole country between the Hydaspes and Acesines. (Arrian, v. 20, 21, 29.) His country is not named in any Indian writer. Shortly afterwards, Alexander received an embassy and presents from Abisaris (no doubt Abhisára), whose territory, as has been shown by Prof. Wilson from the Annals of Cashmir, must have been in the mountains in the southern part of that province. (Asiat. Res. vol. xv. p. 116.) There had been previously a war between this ruler and the Malli, Oxydracae, and the people of the Lower Panjab, which had ended in nothing. Alexander confirmed Abisaris in the possession of his own territory, made Philip satrap of the Malli and Oxydracae, and Pytho of the land between the confluence of the Indus and Acesines and the sea (Arrian, vi. 15); placing, at the same time, Oxyarces over the Paropamisadae. (Arr. vi. 15.) It may be observed that, in the time of Ptolemy, the Cashmirians appear to have held the whole of the Panjab, so far as the Vindhya mountains, a portion of the southern country being, however, in the hands of the Malli and Cathaei.

The same state of things prevailed for some time after the death of Alexander, as appears by a decree of Perdiccas, mentioned in Diodorus (xviii. 3), and with little material change under Antipater. (Diod. xviii. 39.) Indeed, the provinces remained true to the Macedonians till the commencement of the rule of the Prasii, when Sandrocottus took up arms against the Macedonian governors. (Justin. xv. 4.) The origin of this rebellion is clearly traceable. Porus was slain by Eudamus about B.C. 317 (Diod. xix. 14); hence Sandrocottus must have been on the throne about the time that Seleucus took Babylon, B.C. 312. The attempt of the Indians to recover their freedom was probably aided by the fact that Porus had been slainby a Greek. Sandrocottus, as king of the Prasii (Sansc. Prachya) and of the nations on the Ganges, made war with Seleucus Nicator, who penetrated far into India. Plutarch says he ruled over all India, but this is not likely. (Plut. Alex. 62.) It appears

that he crossed the Indus, and obtained by marriage Arachosia, Gedrosia, and the Paropamisadae, from Seleucus. (Strab. xv. p. 724; Appian, Syr. 55.) It was to his court that Megasthenes (as we have before stated) was sent. Sandrocottus was succeeded by Amitrochates (Sansc. Amitraghátas), which is almost certainly the true form of the name, though Strabo calls him Allitrochades. He was the contemporary of Antiochus Soter. (Athen. xiv. 67.) It is clear, from Athenaeus (l. c.), that the same friendship was maintained between the two descendants as between the two fathers. Daimachus was sent as ambassador to Palibothra. (Strab. ii. p. 70.) Then came the wars between the Parthians and Bactrians, and the inore complete establishment of the Graeco-Bactrian kingdom, under Menander, Apollodotus, Eucratides, and their successors, to which we cannot here do more than allude. The effect, however, of these wars was to interrupt communication between the East and the West; hence the meagre nature of the historical records of the period. The expedition of Antiochus the Great to India brought to light the name of another king, Sophagasenus (Polyb. xi. 32), who was, in all probability, king of the Prasii. The Scythians finally put an end to the Bactrian empire about B. c. 136. (De Guignes, Mém. de l'Acad. d. Inscr. xxv. p. 17.) This event is noticed in the Periplus (p. 22), where, however, Parthi must be taken to mean Scythi. (See also Periplus, p. 24; Dionys. Perieg. vv. 1087 -1088.) Eustathius adds, in his commentary on Dionysius:-O kai ’Ivdoσkúlai ovvléтws λeyouéVOL. Minnagara was their chief town, a name, as appears from Isid. Char. ( p. 9), which was partly Scythian and partly Sanscrit. (Cf. also De Guignes, 1. c.)

The Scythians were in their turn driven out of India by Vicrámaditya, about B. c. 56 (Colebrooke, Ind. Algebra, Lond. 1817, p. 43), who established his seat of empire at Oujein (Ujjayini). At the time when the Periplus was compiled, the capital had been again changed, as we there read, 'Ośývn, èv ý кal τὰ Βασιλεία πρότερον ἦν.

It is remarkable that no allusion has been found in any of the early literature of the Hindús to Alexander the Great; but the effect of the later expeditions of the Bactrian kings is apparently indicated under the name of the Yavana. In the astronomical works, the Yavana are barbarians who understood astronomy, whence it has been conjectured by Colebrooke that the Alexandrians are referred to. (Ind. Algebra, p. 80.) Generally, there can be no doubt that the Yarana mean nations to the W. of India. Thus, in the Mahabharata, they make war on the Indians, in conjunction with the Páradi (i. e. Parthi), and the Sacae or Scythians. (Lassen, Pentap. p. 60.) In the Drama of the Mudra-Ráxasa, which refers to the war between Chandragupta and another Indian King, it is stated that Cusumapura (i. e. Palibothra) was surrounded by the Cirratae, Yavani, Cambogi, Persae, Bactrians, and the other forces of Chandragupta, and the king of the Mountain Regions. Lassen thinks, with much reason, that this refers to Seleucus, who, in his war with Chandragupta, reached, as we know, Palibothra. (Plin. vi. 17.)

With regard to the commerce of ancient India, which we have every reason to suppose was very extensive, it is impossible in this place to do more than to indicate a few of the principal facts. Indeed, the commerce of India, including the northern and the southern districts, may be considered as an epitome of the commerce of the world, there being few pro

VOL II.

ductions of any other country which may not be found somewhere within its vast area.

The principal directions in which the commerce of ancient India flowed were, between Western India and Africa, between the interior of the Deccan and the outports of the southern and western coast of the Indian Ocean, between Ceylon and the ports of the Coromandel coast, between the Coromandel coast and the Aurea Chersonesus, and, in the N., along the Ganges and into Tátary and the territory of the Sinae. There appears also to have been a remarkable trade with the opposite coast of Africa, along the district now called Zanguebar, in sesamum, rice, cotton goods, cane-honey (sugar), which was regularly sent from the interior of Ariaca (Concan) to Barygaza (Beroach), and thence westward. (Peripl. p. 8.) Arab sailors are mentioned who lived at Muza (Mocha), and who traded with Barygaza. (Peripl. p. 12.) Banians of India had established themselves on the N. side of Socotra, called the island of Dioscorides (Peripl. p. 17): while, even so early as Agatharchides, there was evidently an active commerce between Western India and Yemen. (Agatharch. p. 66, ed. Hudson.) Again, the rapidity with which Alexander got his fleet together seems to show that there must have been a considerable commerce by boats upon the Indus. At the time of the Periplus there was a chain of ports along the western coast, Barygaza (Beroach), Muziris in Limyrica (Mangalore), Nelkynda (Neliceram), Pattala (once supposed to be Tatta, but much more probably Hydrabúd), and Calliene, now Gallian (Peripl. p. 30): while there were three principal emporia for merchandise, Ozene (Oujein), the chief mart of foreign commerce, (vide an interesting account of its ruins, Asiat. Res. vol. vi. p. 36), and for the transmission of the goods to Barygaza; Tagara, in the interior of the Deccan (almost certainly Deoghir or Devanagari near Elloru), whence the goods were conveyed over difficult roads to Barygaza and Pluthana or Plithana, a place the exact position of which cannot now be determined, but, from the character of the products of the place, must have been somewhere in the Ghats.

Along the Regio Paralia to the S., and on the Coromandel coast, were several ports of consequence; and extensive pearl fisheries in the kingdom of king Pandion, near Colchi, and near the island of Epiodorus, where the πvvikóv (a silky thread spun from the Pinna-fish) was procured. (Peripl. p. 33). Further to the N. were, Masalia (Masulipatam), famous for its cotton goods (Peripl. p. 35); and Gange, a great mart for muslin, betel, pearls, &c., somewhere near the mouth of the Ganges, its exact locality, however, not being now determinable. (Peripl. p. 36.) The commerce of Ceylon (Selandib, i. e. Sinhala-dwipa) was in pearls of the best class, and precious stones of all kinds, especially the ruby and the emerald. The notices in Ptolemy and Pliny shew that its shores were well furnished with commercial towns (Ptol. vii. 4. §§ 3, 4, 5), while we know from the narrative of Cosmas Indicopleustes (ap. Montfaucon, Coll. Nova Bibl. Patr. vol. ii.) that it was, in the sixth century A.D., the centre of Hindu commerce. Besides these places, we learn that there was an emporium upon the Coromandel coast, whence the merchant ships crossed over to Chryse (in all probability Malacca), in the Aurea Chersonesus; the name of it, however, is not specified.

It is probable, however, that the greatest line of commerce was from the N. and W. along the

E