Chapter XLV: Book VI: , it follows that it was from Megasthenês that he drew his (1)

accounts of the forces of the Gangaridæ, Modogalinga, Andaræ, Prasi, Megallæ, Asmagi, Oratæ, Suarataratæ, Automula, Charmæ, and Pandæ (VI. 19), names which, as will be shown below, betray a knowledge of all parts of India. It is a fair inference that the remaining names mentioned by Pliny were taken by him from Megasthenês, perhaps through the medium of Seneca's work. The corruption of Pliny's text and the fact that Megasthenês learnt the tribal names in their Prakrit forms, make it extremely difficult to identify many of the races referred to.

That part of Pliny's account of India which may with some certainty be traced back to Megasthenês begins with a statement of the stages of the royal road from the Hypasis (Biás) to Palibothra (Patna) (Nat. Hist. VI. 17). The next chapter gives an account of the Ganges and its tributaries and mentions the Gangaridæ of Kalinga with their capital Pertalis as the most distant nation on its banks. In the 19th chapter, after an account of the forces of the Gangaridæ, Pliny gives a list of thirteen tribes, of which the only ones that can be said to be satisfactorily identified are Modogalinga (the three Calingas: Caldwell Drav. Gr.), Molindæ (compare Mount Mâlindya of Varâha Mihira Br. S. XIV.), and Thalutæ (McCrindle reads Taluctæ and identifies with the Tâmraliptakas of Tamluk on the lower Ganges). He next mentions the Andaræ (Andhras of Telingana) with thirty cities 100,000 foot 2000 horse and 1000 elephants. He then digresses to speak of the Dardæ (Dards of the Upper Indus) as rich in gold and the Setæ (of Mêwâr, Lassen) in silver, and next introduces the Prasi (Prâcyas) of Palibothra (Pâtaliputra) as the most famous and powerful of all the tribes, having 600,000 foot 30,000 horse and 8000 elephants. Inland from these he names the Monædes (Munda of Singbhúm) and Suari (Savaras of Central India) among whom is Mount Maleus (Mahendra Male?). Then after some account of the Iomanes (Yamunâ) running between Methora (Mathurâ) and Chrysobora (McCrindle reads Carisobora, Arrian Ind. VIII. Kleisobora = Krishnapura?) he turns to the Indus, of some of whose nineteen tributaries he gives some account in chapter 20. He then digresses to give an account of the coast of India, starting from the mouth of the Ganges, whence to Point Calingon (Point Godâvari) and the city of Dandaguda (Cunningham's Râja Mahendri, but more probably the Dhanakataka or Dhenukâkata of the Western cave inscriptions) he reckons 625 miles. The distance thence to Tropina (Tirupanatara near Kochin according to Burgess) is 1225 miles. Next at a distance of 750 miles is the cape of Perimula, where is the most famous mart of India. Further on in the same chapter is mentioned a city named Automula on the sea shore among the Arabastræ (or Salabastræ and Oratæ, McCrindle) a noble mart where five rivers together flow into the sea. There can hardly be a doubt that the two places are the same, the two names being taken from different authorities, and that the place meant is Chemula or Cheul (Ptolemy's Simulla) the five rivers being those that flow into Bombay Harbour northward of Cheul. The distance from Perimula to the Island of Patala in the Indus is 620 miles. Pliny next enumerates as hill tribes between the Indus and Jamna, shut in a ring of mountains and deserts for a space of 625 miles, the Cæsi (the Kekioi of Arr. Ind. IV. and Kêkayas of the Purânas, about the head waters of the Sutlej), the Cetriboni of the woods (... Vana?), the Megallæ (Mêkalas) with 500 elephants and unknown numbers of horse and foot, the Chrysei (Karûsha) Parasangæ (Pârasava, corrupted by the likeness of its first three syllables to the word parasanga, the Asmagi (Asmaka of Varâha Mihira) with 30,000 foot 300 elephants and 800 horse. These are shut in by the Indus and surrounded by a circle of mountains and deserts for 625 miles. Next come the Dari and Suræ and then deserts again for 187 miles. Whether these are or are not correctly identified with the Dhars and Saurs of Sindh, they must be placed somewhere to the north of the Ran. Below them come five kingless tribes living in the hills along the sea-coast--the Maltecoræ, Singhæ, Marohæ, Rarungæ, and Moruni--none of whom are satisfactorily identified, but who may be placed in Kachh. Next follow the Nareæ, enclosed by Mount Capitalia (Âbu) the highest mountain in India, on the other side of which are mines of gold and silver. The identification of Capitalia with Âbu is probable enough, but the name given to the mountain must be connected with the Kapishthala of the Purânas, who have given their name to one of the recensions of the Yajur Veda, though Kaithal, their modern representative, lies far away from Âbu in the Karnâl district of the Panjâb, and Arrian places his kambistholoi (Ind. IV) about the head waters of the Hydraôtês (Râvî). After Capitalia and the Nareæ come the Oratæ with but ten elephants but numerous infantry. These must be the Aparântakas of the inscriptions and purânas, Megasthenês having learnt the name in a Prâkrit form (Avarâta, Orâta). The name of the next tribe, who have no elephants but horse and foot only, is commonly read Suarataratæ (Nobbe) but the preferable reading is Varetatæ (McCrindle) which when corrected to Varelatæ represents Varalatta, the sixth of the seven Konkans in the purânic lists (Wilson As. Res. XV. 47), which occupied the centre of the Thána district and the country of the wild tribe of the Varlîs. Next are the Odonbæores, whose name is connected with the udumbara Ficus glomerata tree, and who are not the Audumbari Sâlvas of Pânini (IV. i. 173) but must be placed in Southern Thána. Next come the Arabastræ Oratæ (so read for Arabastræ Thorace of Nobbe, and Salabastræ Horatæ of McCrindle) or Arabastra division of the Oratæ or Konkanîs. Arabastra may be connected with the Ârava of Varâha-Mihira's South-Western Division (Br. S. XIV. 17) where they are mentioned along with Barbara (the seventh or northernmost Konkan). This tribe had a fine city in a marsh infested by crocodiles and also the great mart of Automula (Cheul) at the confluence of five rivers, and the king had 1600 elephants 150,000 foot and 5000 horse, and must therefore have held a large part of the Dakhan as well as of the sea coast. Next to this kingdom is that of the Charmæ, whose forces are small, and next to them the Pandæ (Pândya of Travancor) with 300 cities 150,000 foot and 500 elephants. Next follows a list of thirteen tribes, some of which St. Martin has identified with modern Râjput tribes about the Indus, because the last name of the thirteen is Orostræ, "who reach to the island of Patala," and may be confidently identified with the Saurâshtra of Kâthiâvâda. We must however assume that Megasthenês after naming the tribes of the west coast enumerates the inland tribes of the Dakhan until he arrives at the point from which he started. But the only identification that seems plausible is that of the Derangæ with the Telingas or Telugus. Next to the Orostræ follows a list of tribes on the east of the Indus from south to north--the Mathoæ (compare Mânthava, a Bâhîka town Pân. IV. ii. 117), Bolingæ (Bhâulingi, a Sâlva tribe Pân. IV. i. 173), Gallitalutæ (perhaps a corruption of Tâilakhali, another Sâlva tribe, ib.), Dimuri, Megari, Ardabæ, Mesæ (Matsya of Jaipur?), Abi, Suri, (v. 1. Abhis Uri), Silæ, and then deserts for 250 miles. Next come three more tribes and then again deserts, then four or five (according to the reading) more tribes, and the Asini whose capital is Bucephala (Jalâlpur) (Cunningham Anc. Geog. 177). Megasthenês then gives two mountain tribes and ten beyond the Indus including the Orsi (Urasâ) Taxilæ (Takshasilâ) and Peucolitæ (people of Pushkâlavatî). Of the work of Dêïmachos, who went on an embassy to Allitrochadês (Bindusâra) son of Candragupta, nothing is known except that it was in two books and was reckoned the most untrustworthy of all accounts of India (Strabo, II. i. 9).

[Ptolemy II.] Ptolemy II. Philadelphos (died 247 B.C.) interested himself in the trade with India and opened a caravan road from Koptos on the Nile to Berenikê on the Red Sea (Strabo, XVII. i. 45) and for centuries the Indian trade resorted either to this port or to the neighbouring Myos Hormos. He also sent to India (apparently to Asoka) an envoy named Dionysius, who is said by Pliny (VI. 17) to have written an account of things Indian of which no certain fragments appear to remain. But we know from the fragments of [Agatharkhides.] Agatharkhides (born c. 250 A.D.) who wrote in old age an account of the Red Sea of which we have considerable extracts in Diodôros (III. 12-48) and Phôtios (Müller's Geogr. Gr. Min. I. 111ff), states that in his time the Indian trade with Potana (Patala) was in the hands of the Sabæans of Yemen. (Müller, I. 191.) In fact it was not until the voyages of Eudoxos (see below) that any direct trade sprang up between India and Egypt. The mention of Patala as the mart resorted to by the Arabs shows that we are still in Pliny's first period (see below).

[The Baktrian Greeks.] The Baktrian Greeks extended their power into India after the fall of the Mâurya empire (c. 180 B.C.) their leader being Dêmêtrios son of Euthydêmos, whose conquests are referred to by Justin (XLI. 6) and Strabo (XI. ii. 1). But the most extensive conquests to the east and south were made by Menandros (c. 110 B.C.) who advanced to the Jumna and conquered the whole coast from Pattalênê (lower Sindh) to the kingdoms of Saraostos (Surâshtra) and Sigertis (Pliny's Sigerus?) (Strabo, XI. ii. 1). These statements of Strabo are confirmed by the author of the Periplus (c. 250 A.D.) who says that in his time drakhmai with Greek inscriptions of Menandros and Apollodotos were still current at Barygaza (Per. 47). Apollodotos is now generally thought to have been the successor of Menandros (C. 100 B.C.) (Brit. Museum Cat. of Bactrian Coins page xxxiii.). Plutarch (Reip. Ger. Princ.) tells us that Menandros' rule was so mild, that on his death his towns disputed the possession of his ashes and finally divided them.

[Eudoxos of Cyzicus.] Eudoxos of Cyzicus (c. 117 B.C.) made in company with others two very successful voyages to India, in the first of which the company were guided by an Indian who had been shipwrecked on the Egyptian coast. Strabo (II. iii. 4), in quoting the story of his doings from Poseidônios, lays more stress upon his attempt to circumnavigate Africa than upon these two Indian voyages, but they are of very great importance as the beginnings of the direct trade with India.

[Eratosthenês.] The Geographers down to Ptolemy drew their knowledge of India almost entirely from the works of Megasthenês and of the companions of Alexander. Among them Eratosthenês (c. 275-194 B.C.), the founder of scientific geography, deserves mention as having first given wide currency to the notion that the width of India from west to east was greater than its length from north to south, an error which lies at the root of Ptolemy's distortion of the map. Eratosthenês' critic Hipparkhos (c. 130 B.C.) on this point followed the more correct account of Megasthenês, and is otherwise notable as the first to make use of astronomy for the determination of the geographical position of places.

[Strabo.] Strabo (c. 63 B.C.-23 A.D.) drew his knowledge of India, like his predecessors, chiefly from Megasthenês and from Alexander's followers, but adds (XV. i. 72) on the authority of Nikolaos of Damascus (tutor to the children of Antony and Cleopatra, and envoy of Herod) (an account of three Indian envoys from a certain king Pôros to Augustus (ob. A.D. 14), who brought presents consisting of an armless man, snakes, a huge turtle and a large partridge, with a letter in Greek written on parchment offering free passage and traffic through his dominions to the emperor's subjects. With these envoys came a certain Zarmanokhêgas (Sramanâcârya, Lassen) from Bargosê (Broach, the earliest mention of the name) who afterwards burnt himself at Athens, "according to the ancestral custom of the Indians." The fact that the embassy came from Broach and passed through Antioch shows that they took the route by the Persian Gulf, which long remained one of the chief lines of trade (Per. chap. 36). If the embassy was not a purely commercial speculation on the part of merchants of Broach, it is hard to see how king Pôros, who had 600 under-kings, can be other than the Indo-Skythian Kozolakadaphes, who held Pôros' old kingdom as well as much other territory in North-West India. This if correct would show that as early as the beginning of our era the Indo-Skythian power reached as far south as Broach. The fact that the embassy took the Persian Gulf route and that their object was to open commercial relations with the Roman empire seems to show that at this period there was no direct trade between Broach and the Egyptian ports of the Red Sea. Strabo however mentions that in his time Arabian and Indian wares were carried on camels from Myos Hormos (near Râs Abu Somer) on the Red Sea to Koptos on the Nile (XVII. i. 45 and XVI. iv. 24) and dilates upon the increase of the Indian trade since the days of the Ptolemies when not so many as twenty ships dared pass through the Red Sea "to peer out of the Straits," whereas in his time whole fleets of as many as 120 vessels voyaged to India and the headlands of Ethiopia from Myos Hormos (II. v. 12 and XV. i. 13). It would seem that we have here to do with Pliny's second period of Indian trade, when Sigerus (probably Janjira) was the goal of the Egyptian shipmasters (see below). Strabo learnt these particulars during his stay in Egypt with Aelius Gallus, but they were unknown to his contemporary Diodôros who drew his account of India entirely from Megasthenês (Diod. II. 31-42) and had no knowledge of the East beyond the stories told by Jamboulos a person of uncertain date of an island in the Indian Archipelago (Bali, according to Lassen) (Diod. II. 57-60). Pomponius Mela (A.D. 43) also had no recent information as regards India.

[Pliny.] Pliny (A.D. 23-79) who published his Natural History in A.D. 77 gives a fairly full account of India, chiefly drawn from Megasthenês (see above). He also gives two valuable pieces of contemporary information:

(i) An account of Ceylon (Taprobanê) to which a freedman of Annius Plocamus, farmer of the Red Sea tribute, was carried by stress of weather in the reign of Claudius (A.D. 41-54). On his return the king sent to the emperor four envoys, headed by one Rachias (VI. 22).

(ii) An account of the voyage from Alexandria to India by a course which had only lately been made known (VI. 23). Pliny divides the history of navigation from the time of Nearchus to his own age into three periods:

(a) the period of sailing from Syagrus (Râs Fartak) in Arabia to Patalê (Indus delta) by the south-west wind called Hippalus, 1332 miles;

(b) the period of sailing from Syagrus (Râs Fartak) to Sigerus (Ptol. Milizêgyris, Peripl. Melizeigara, probably Janjíra, and perhaps the same as Strabo's Sigertis);

(c) the modern period, when traffic went on from Alexandria to Koptos up the Nile, and thence by camels across the desert to Berenice (in Foul Bay), 257 miles. Thence the merchants start in the middle of summer before the rising of the dogstar and in thirty days reach Okelis (Ghalla) or Cane (Hisn Ghorab), the former port being most frequented by the Indian trade. From Okelis it is a forty days' voyage to Muziris (Muyyiri, Kranganur) which is dangerous on account of the neighbouring pirates of Nitrias (Mangalor) and inconvenient by reason of the distance of the roads from the shore. Another better port is Becare (Kallada, Yule) belonging to the tribe Neacyndon (Ptol. Melkynda, Peripl. Nelkynda) of the kingdom of Pandion (Pândya) whose capital is Modura (Madura). Here pepper is brought in canoes from Cottonara (Kadattanâdu). The ships return to the Red Sea in December or January.

It is clear that the modern improvement in navigation on which Pliny lays so much stress consisted, not in making use of the monsoon wind, but in striking straight across the Indian ocean to the Malabar coast. The fact that the ships which took this course carried a guard of archers in Pliny's time, but not in that of the Periplus, is another indication that the direct route to Malabar was new and unfamiliar in the first century A.D. The name Hippalus given to the monsoon wind will be discussed below in dealing with the Periplus.

[Dionysios Periégétés.] Dionysios Periégétés who has lately been proved to have written under Hadrian (A.D. 117-138) (Christ's Griech. Litteratur Gesch., page 507) gives a very superficial description of India but has a valuable notice of the Southern Skythians who live along the river Indus to the east of the Gedrôsoi (I. 1087-88).

[Klaudios Ptolemaios.] Klaudios Ptolemaios of Alexandria lived according to Suidas under Marcus Aurelius Antoninus (A.D. 161-180). He compiled his account of India as part of a geographical description of the then known world, and drew much of his materials from Marinos of Tyre, whose work is lost, but who must have written about A.D. 130. Ptolemy (or Marinos before him) had a very wide knowledge of India, drawn partly from the relations of shipmasters and traders and partly from Indian lists similar to those of the Purânas but drawn up in Prâkrit. He seems to have made little if any use of Megasthenês and the companions of Alexander. But his map of India is distorted by the erroneous idea, which he took from Eratosthenês, that the width of India from west to east greatly exceeded its length from north to south. Ptolemy begins his description of India with the first chapter of his seventh book, which deals with India within the Ganges. He gives first the names of rivers, countries, towns, and capes along the whole coast of India from the westernmost mouth of the Indus to the easternmost mouth of the Ganges. He next mentions in detail the mountains and the rivers with their tributaries, and then proceeds to enumerate the various nations of India and the cities belonging to each, beginning with the north-west and working southwards: and he finally gives a list of the islands lying off the coast. In dealing with his account of western India it will be convenient to notice together the cities of each nation which he mentions separately under the heads of coast and inland towns.

He gives the name of Indo-Skythia to the whole country on both sides of the lower course of the Indus from its junction with the Koa (Kábul river), and gives its three divisions as Patalênê (lower Sindh) Abiria (read Sabiria, that is Sauvîra or upper Sindh and Multân) and Surastrênê (Surâshtra or Kâthiâvâda). We have seen that Dionysios knew the southern Skythians of the Indus, and we shall meet with them again in the Periplus (chapter 38ff).

He enumerates seven mouths of the Indus, but the river is so constantly changing its course that it is hopeless to expect to identify all the names given by him (Sagapa, Sinthôn, Khariphron, Sapara, Sabalaessa, and Lônibare) with the existing channels. Only it may be noted that Sinthôn preserves the Indian name of the river (Sindhu) and that the easternmost mouth (Lônibare) probably represents both the present Korî or Launî and the Lûnî river of Mârwâr, a fact which goes some way to explain why Ptolemy had no idea of the existence of Kachh, though he knows the Ran as the gulf of Kanthi. Hence he misplaces Surastrênê (Surâshtra or Kâthiâvâda) in the Indus delta instead of south of the Ran. Ptolemy enumerates a group of five towns in the north-western part of Indo-Skythia (Kohat, Bannu, and Dera Ismail Khân) of which Cunningham (Anc. Geog. pages 84ff) has identified Banagara with Bannu, and Andrapana with Daraban, while the sites of Artoarta, Sabana, and Kodrana are unknown. Ptolemy next gives a list of twelve towns along the western bank of the Indus to the sea. Of these Embolima has been identified by Cunningham (Anc. Geog. page 52) with Amb sixty miles above Attok, and Pasipêda is identified by St. Martin with the Besmaid of the Arab geographers and placed near Mithankot at the junction of the Chenab with the Indus. Sousikana, which comes next in the list to Pasipêda, is generally thought to be a corruption of Mousikanos, and is placed by the latest authority (General Haig, The Indus Delta Country, page 130) in Bahâwalpur, though Cunningham (Anc. Geog. page 257) puts it at Alor, which is somewhat more in accordance with Ptolemy's distances. Kôlaka the most southerly town of the list, cannot well be the Krôkala of Arrian (Karâchi) as McCrindle supposes, for Ptolemy puts it nearly a degree north of the western mouth of the Indus.

The two great towns of the delta which Ptolemy next mentions, are placed by General Haig, Patala at a point thirty-five miles south-east of Haidarâbâd (op. cit. page 19) and Barbarei near Shâh Bandar (op. cit. page 31). Barbarei is mentioned again in the Periplus (chapter 38) under the name of Barbarikon. Ptolemy gives the names of nine towns on the left bank of the Indus from the confluence to the sea, but very few of them can be satisfactorily identified. Panasa can only be Osanpur (St. Martin) on Fluellen's principles. Boudaia must represent the Budhîya of the Arabs, though it is on the wrong side of the river (see Haig, op. cit. page 57ff). Naagramma may with Yule be placed at Naushahro. Kamigara cannot be Aror (McCrindle), if that place represents Sousikana. Binagara is commonly thought to be a corrupt reading of Minnagara (compare Periplus chapter 38). Haig (op. cit. page 32 note 47) refers to the Tuhfatu'l Kirâm as mentioning a Minnagar in pargana Shâhdâdpur (north-east of Haidarâbâd). Parabali, Sydros, and Epitausa have not been identified, but must be looked for either in Haidarâbâd or in Thar and Pârkar. Xoana may with Yule be identified with Siwana in the bend of the Lûnî and gives another indication that Ptolemy confounded the Lûnî with the eastern mouth of the Indus.

On the coast of Surastrênê (Kâthiâvâda) Ptolemy mentions, first, the island of Barakê (Dvârakâ Bêt): then the city Bardaxêma which must be Porbandar (Yule), in front of the Barada hills: then the village of Surastra, which perhaps represents Verâval, though it is placed too far north. Surastra cannot well be Junâgad (Lassen) which is not on the coast and in Ptolemy's time was not a village, but a city, though it is certainly strange that Ptolemy does not anywhere mention it. Further south Ptolemy places the mart of Monoglôsson (Mangrol). The eastern boundary of the coast of Indo-Skythia seems to have been the mouth of the Môphis (Mahî). Ptolemy's account of Indo-Skythia may be completed by mentioning the list of places, which he puts to the east of the Indus (i. e. the Lûnî) and at some distance from it.

These are: Xodrakê, which has not been identified, but which must be placed somewhere in Mewâr, perhaps at the old city of Pûr, seventy-two miles north-east of Udaipur, or possibly at the old city of Ahar, two miles from Udaipur itself (Tod's Râjasthân, I. 677-78).

Sarbana, which is marked in Ptolemy's map at the head-waters of the Mahî in the Apokopa mountains (Aravallis), must be identified with Sarwan about ten miles north-west of Ratlâm. There is also a place called Sarwanio close to Nimach, which Ptolemy may have confused with Sarwan.

Auxoamis, which St. Martin identifies with Sûmî, and Yule with Ajmir, but neither place suits the distance and direction from Sarwan. If Ptolemy, as above suggested, confused Sarwan and Sarwanio, Auxoamis may be Ahar near Udaipur, Pûr being then Xodrakê: otherwise Auxoamis may be Ídar. The question can only be settled by more exact knowledge of the age of Ahar and of Ídar. Orbadarou may provisionally with Yule be placed at Âbu.

Asinda must be looked for near Sidhpur, though it cannot with St. Martin be identified with that place. Perhaps Vadnagar (formerly Ânandapura and a very old town) may be its modern representative.

Theophila may be Devaliya (Yule) or Thân (Burgess) in north-east Kâthiâvâda.

Astakapra is admitted to be Hastakavapra or Hâthab near Bhâvnagar (Bühler).

Larikê is described by Ptolemy next after Indo-Skythia on his way down the West Coast. The northern limit of its coast was the mouth of the river Môphis (Mahî). Its name is the Lâta of purânas and inscriptions. Ptolemy mentions as on its coast the village of Pakidarê, which may be a misreading for Kâpidarê and represent Kâvî (Kâpikâ of inscriptions) a holy place just south of the Mahî. Next comes Cape Maleô, which Ptolemy both in his text and in his map includes in Larikê, though there is no prominent headland in a suitable position on the east side of the Gulf of Cambay. As he puts it 2 1/4 degrees west of Broach, it may probably be identified with Gopnâth Point in Kâthiâvâda on the other side of the gulf (the Pâpikê of the Periplus), his name for it surviving in the neighbouring shoals known as the Malai banks. It is in agreement with this that Ptolemy puts the mouth of the river Namados (Narmadâ) to the north of Cape Maleô. South of the river is Kamanê which may be identified with the Kamanijja or Karmanêya of inscriptions, that is with Kamlej on the Taptî above Surat. It has been supposed to be the Kammôni of the Periplus (chapter 43), which was the village opposite to the reef called Hêrônê on the right (east) of the gulf of Barygaza: but it is perhaps best to separate the two and to identify Kammôni with Kim, north of Olpâd. The next town mentioned is Nousaripa, which should probably be read Nousarika, being the Navasârikâ of inscriptions and the modern Nausârî. The most southerly town of Larikê is Poulipoula, which has been identified with Phulpâdâ or old Surat, but is too far south. Bilimora is perhaps the most likely position for it, though the names do not correspond (unless Pouli is the Dravidian Puli or poli = a tiger, afterwards replaced by Bili = a cat). Ptolemy begins his list of the inland cities of Larikê with Agrinagara, which may with Yule be identified with Âgar, thirty-five miles north-east of Ujjain, and the Âkara of inscriptions. The next town is Siripalla, which has not been identified, but should be looked for about thirty miles to the south-east of Agar, not far from Shâhjahânpur. The modern name would probably be Shirol. Bammogoura must be identified, not with Pawangad (Yule), but with Hiuen Tsiang's "city of the Brâhmans" (Beal, Si-yu-ki, II. 262), 200 li (about 33 miles) to the north-west of the capital of Mâlava in his time. The distance and direction bring us nearly to Jaora. Sazantion and Zerogerei have not been satisfactorily identified but may provisionally be placed at Ratlâm and Badnawar respectively, or Zêrogerei may be Dhâr as Yule suggested. Ozênê the capital of Tiastanês is Ujjain the capital of the Kshatrapa Cashtana who reigned c. 130 A.D. His kingdom included Western Mâlwâ, West Khándesh, and the whole of Gujarát south of the Mahî. His grandson Rudradâman (A.D. 150) tells us in his Girnâr inscription (I. A. VII. 259) that his own kingdom included also Mârwâr Sindh and the lower Panjâb. Next to Ujjain Ptolemy mentions Minnagara, which must have been somewhere near Mânpur. Then we come to Tiatoura or Chândor (Yule) on the ridge which separates Khándesh from the valley of the Godâvari, and finally on that river itself Nasika the modern Nâsik. It is very doubtful whether Nâsik at any time formed part of the dominions of Cashtana, since we know from the inscriptions in the Nâsik caves that the Kshatrapas were driven out of that part of the country by Gautamîputra Sâtakarni, the father of Ptolemy's contemporary Pulumâyi. Ptolemy probably found Nâsik mentioned in one of his lists as on a road leading from Ujjain southwards and he concluded that they belonged to the same kingdom.

Ariakê of the Sadinoi included the coast of the Konkan as far south as Baltipatna (near Mahâd) and the Deccan between the Godâvari and the Krishna. The name occurs in Varâha Mihira's Brihat Samhitâ XIV. in the form Âryaka. The tribal name Sadinoi is less easy to explain. The suggested connection with the word Sâdhana as meaning an agent (Lassen) and its application to the Kshatrapas of Gujarát, are not tenable. The only authority for this meaning of Sâdhana is Wilson's Sanskrit Dictionary, and at this time it is certain that Ariakê belonged, not to the Kshatrapas of Gujarát, but to the Sâtakarnis of Paithan on the Godâvari. Bhândârkar's identification of the Sadinoi with Varâha Mihira's Sântikas seems also somewhat unsatisfactory. Ptolemy's name may possibly be a corruption of Sâtakarni or Sâtavâhana. The coast towns of this region were Soupara (Supârâ near Bassein), south of which Ptolemy places the river Goaris (Vaitaranî), Dounga (perhaps Dugád ten miles north of Bhiwndi) south of which is the Bênda river (Bhiwndî Creek), Simylla, a mart and a cape, the Automula and Perimula of Pliny and the modern Cheul (Chemula); Milizêgyris an island, the same as the Melizêigara of the Periplus and (probably) as the Sigerus of Pliny and the modern Janjîra; Hippokoura, either Ghodegâon or Kudâ (Yule) in Kolâbâ district; Baltipatna, probably the Palaipatmai of the Periplus and the same as Pâl near Mahâd.

The inland dominions of the Sadinoi were much more extensive than their coast line. Ptolemy gives two lists of cities, one of those lying to the west (i. e. north) of the Bênda, whose course in the Deccan represents the Bhîmâ river, and the other of those between the Bênda and the Pseudostomos (here the Mâlprabhâ and Krishna or possibly the Tungabhadra with its tributaries). The most easterly towns in the first list, Malippala and Sarisabis, are not satisfactorily identified, but must be looked for in the Nizâm's country to the south-east of Haidarâbâd. Next comes Tagara mentioned in the Periplus (chapter 51) as ten days east from Paithan, and therefore about the latitude of Kulbarga, with which it is identified by Yule. The distance and direction make its identification with Deogir (Wilford and others), Junnar (Bhagwanlâl), or Kolhâpur (Fleet) impossible. The best suggestion hitherto made is that it is Dârur or Dhârur (Bhândârkar), but Dârur in the Bhîr district is too far north, so Dhârur fifty miles west of Haidarâbâd must be taken as the most likely site. Next to Tagara Ptolemy mentions Baithana, which is the Paithana of the Periplus and the modern Paithan on the Godâvari. It is called by our author the capital of Siroptolemaios, who is the Srî-Pulumâyi of the Nâsik cave inscriptions. Next to Baithana comes Deopali, which may safely be identified with the modern Deoli in the suburbs of Ahmadnagar. Gamaliba, the next stage, must be placed somewhere on the line between Ahmednagar and Junnar, which latter ancient town is to be identified with Ptolemy's Omênogara, although this name is not easy to explain.

The second list of towns in Ariakê begins with Nagarouris (Nagarapurî) which probably represents Poona which even then must have been a place of importance, being at the head of the great road down the Bhorghat. Tabasô (compare Varâha Mihira's Tâpasâsrâmâh and Ptolemy's own Tabasoi) may be the holy city of Pandharpur. Indê has retained its ancient name (Indî in the north of the Bîjâpur district). Next follows Tiripangalida (Tîkota in the Kurundwâd State ?) and then Hippokoura, the capital of Baleokuros. Dr. Bhândârkar has identified this king with the Vilivâyakûra of coins found in the Kolhápur state. His capital may possibly be Hippargi in the Sindgi taluka of the Bîjâpur district. Soubouttou, the next town on Ptolemy's list, is not identifiable, but the name which follows, Sirimalaga, must be Sirnâl in the Bîjâpur taluka of the same district.

Kalligeris may be identified not with Kanhagiri (McCrindle) but with Galgali at the crossing of the Krishna, and Modogoulla is not Mûdgal (McCrindle) but Mudhol on the Ghâtprabhâ. Petirgala should probably read Penengala, and would then represent the old town of Panangala or Hongal in the Dhârvâd district. The last name on the list is Banaouasei, which is Vanavâsî, about ten miles from Sirsi in Kanara, a very old town where a separate branch of the Sâtakarnis once ruled.

The Pirate Coast is the next division of Western India described by Ptolemy, who mentions five sea-ports but only two inland cities. It is clear that the pirates were hemmed in on the land side by the dominions of the Sâtakarnis, and that they held but little territory above the ghâts, though their capital Mousopallê was in that region. The places on the coast from north to south were Mandagara, the Mandagara of the Periplus (chapter 53) which has been satisfactorily identified with Mandangad to the south of the Bânkot creek.

Byzantion, which, as Dr. Bhândârkar first pointed out, is the Vaijayantî of inscriptions may be placed either at Chiplun or at Dábhol at the mouth of the Vâsishthî river. Chiplun is the only town of great antiquity in this part of the Konkan, and if it is not Vaijayantî Ptolemy has passed over it altogether. The similarity of the names has suggested the identification of Byzantion with Jaygad (Bhândârkar) or Vijayadrug (Vincent), but both these places are comparatively modern. There are indeed no very ancient towns in the Konkan between Sangamêshvar and the Sâvantvâdi border.

Khersonêsos is generally admitted to be the peninsula of Goa.

Armagara is placed a little to the north of the river Nanagouna and may be represented by Cape Ramas in Portuguese territory.

The river Nanagouna here is generally supposed to be the Kâlînadî, though in its upper course it seems to represent the Taptî, and a confusion with the Nânâ pass led Ptolemy to bring it into connection with the rivers Goaris and Bênda (Campbell).

Nitra, the southernmost mart on the pirate coast, is the Nitrias of Pliny, and has been satisfactorily identified by Yule with Mangalor on the Nêtravatî.

The inland cities of the Pirates are Olokhoira and Mousopallê the capital, both of which must be sought for in the rugged country about the sources of the Krishna and may provisionally be identified with the ancient towns of Karâd and Karvîr (Kolhápur) respectively. To complete Ptolemy's account of this coast it is only necessary to mention the islands of Heptanêsia (Burnt Islands ?) Trikadiba and Peperinê. We are not here concerned with his account of the rest of India.

[Bardesanês.] Bardesanês met at Babylon certain envoys sent from India to the emperor Antoninus Pius (A.D. 154-181) and received from Damadamis and Sandanês, who were of their number, accounts of the customs of the Brâhmans and of a rock temple containing a statue of Siva in the Ardhanârî form. Lassen (III. 62 and 348) connects Sandanês with the Sadinoi and places the temple in Western India, but neither of these conclusions is necessary. The object of the embassy is unknown.

[Periplus.] The Periplus of the Erythrean Sea, formerly though wrongly attributed to Arrian (150 A.D.), is an account of the Egyptian trade with East Africa and India, written by a merchant of Alexandria for the use of his fellows. It is preserved in a single manuscript which in some places is very corrupt. The age of this work has been much disputed: the chief views as to this matter are,

(i) that the Periplus was written before Pliny and made
use of by him (Vincent, Schwanbeck, and Glaser). The
arguments of Vincent and Schwanbeck are refuted by Müller
(Geogr. Gr. Min. I. xcviii.) Glaser's case is (Ausland 1891,
page 45) that the Malikhas of the Periplus is Malchos III. of
Nabathæa (A.D. 49-71), that the Periplus knows Meroê as capital
of Ethiopia, while at the time of Nero's expedition to East
Africa (A.D. 68), it had almost vanished, and lastly that the
author of the Periplus is Basilis or Basilês, whom Pliny names
as an authority for his Book VI. It may be replied that Malikhas
is the title Malik and may have been applied to any Arab Sheikh
(Reinaud): that the Periplus does not with certainty mention
Meroê at all: and that Basilis whether or not a contemporary of
Ptolemy Philadelphos was at any rate earlier than Agatharkhidês
(c. 200 B.C.), who quotes him (Geog. Gr. Min. I. 156);
(ii) that the Periplus was written at the same time as Pliny's
work, but neither used the other (Salmasius). This view is
refuted by Müller (op. cit. page 155);
(iii) that the Periplus was written after 161 A.D. (Dodwell);
Müller has shown (ibid.) that Dodwell's arguments are
inconclusive;
(iv) the received view that the Periplus was written between
A.D. 80 and A.D. 89 (Müller);
(v) that the Periplus was written about the middle of the third
century (Reinaud Mém. de l'Ac. des Inscr. XXIV. Pt. ii.
translated in I. A. VIII. pages 330ff).

The only choice lies between the view of Müller and that of Reinaud. Müller argues for a date between A.D. 80 and A.D. 89, because the Periplus knows no more than Pliny of India beyond the Ganges, whereas Ptolemy's knowledge is much greater: because the Periplus calls Ceylon Palaisimoundou, which is to Ptolemy (VII. iv. 1) an old name: because the Nabathæan kingdom, which was destroyed A.D. 105, was still in existence at the time of the Periplus: because the Periplus account of Hippalos shows it to be later than Pliny: and because the Periplus mentions king Zôskalês, who must be the Za Hakalê of the Abyssinian lists who reigned A.D. 77-89. It may be replied that the Periplus is not a geography of Eastern Asia, but a guide book for traders with certain ports only: that Ptolemy must have found in his lists three names for Ceylon, Taprobanê, Palaisimoundou, and Salikê, and that he has wrongly separated Palai from Simoundou, taking it to mean "formerly" and therefore entered Simoundou as the old and Salikê as the modern name, [1467] whereas all three names were in use together: that the Nabathæan king Malikhas was simply the Sheikh of the tribe (Reinaud), and points to no definite date: that the Periplus' account of Hippalos is certainly later than Pliny: and that the Zôskalês of the Periplus is the Za Sâgal or Za Asgal of the Abyssinian lists, who reigned A.D. 246-47 (Reinaud).

It follows that Reinaud's date for the Periplus (A.D. 250) is the only one consistent with the facts and especially with the Indian facts. As will appear below, the growth of the Hippalos legend since Pliny's time, the rival Parthians in Sindh, the mention of Mambaros and the supplanting of Ozênê by Minnagara as his capital since Ptolemy's time, the independence of Baktria, and the notices of Saraganês and Sandanês, are all points strongly in favour of Reinaud's date.

In the time of the Periplus the ships carrying on the Indian trade started from Myos Hormos (near Ras Abu Somer) or Berenikê (in Foul Bay) and sailed down the Red Sea to Mouza (Musa twenty-five miles north of Mokhâ), and thence to the watering place Okêlis (Ghalla) at the Straits. They then followed the Arabian coast as far as Kanê (Hisn Ghurâb in Hadramaut) passing on the way Eudaimôn Arabia (Aden) once a great mart for Indian traders, but lately destroyed by king Elisar (Müller's conjecture for KAISAR of the MS.) From Kanê the routes to India diverge, some ships sailing to the Indus and on to Barygaza, and others direct to the ports of Limyrikê (Malabár Coast). There was also another route to Limyrikê, starting from Arômata (Cape Guardafui). In all three voyages the ships made use of the monsoon, starting from Egypt in July. The monsoon was called Hippalos, according to the Periplus (chapter 57), after the navigator who first discovered the direct course across the sea, and it has been inferred from Pliny's words (VI. 23) that this pilot lived in the middle of the first century A.D. But Pliny's own account shows that, as we should expect, the progress from a coasting to a direct voyage was a gradual one, with several intermediate stages, in all of which the monsoon was more or less made use of. There was therefore no reason for naming the wind from the pilot who merely made the last step. Further though Pliny knows Hippalus as the local name of the monsoon wind in the eastern seas, he says nothing of its having been the name of the inventor of the direct course. The inference seems to be that Hippalos the pilot is the child of a seaman's yarn arising out of the local name of the monsoon wind, and that his presence in the Periplus and not in Pliny shows that the former writer is much later than the latter.

The merchant bound for Skythia (Sindh) before he reaches land, which lies low to the northward, meets the white water from the river Sinthos (Indus) and water snakes (chapter 38). The river has seven mouths, small and marshy all but the middle one, on which is the port of Barbarikon (Shâhbandar, Haig, page 31) whence the merchants' wares are carried up by river to the capital Minnagar (near Shâhdâdpur, Haig, page 32), which is ruled by Parthians who constantly expel one another (chapter 39). These contending Parthians must have been the remnant of the Karên Pahlavs who joined with the Kushâns to attack Ardeshir Pâpakân (Journ. As. [1866] VII. 134). The imports are clothing, flowered cottons, topazes, coral, storax, frankincense, glass vessels, silver plate, specie, and wine: and the exports costus (spice), bdellium (gum), yellow dye, spikenard, emeralds, sapphires, furs from Tibet, cottons, silk thread, and indigo. The list of imports shows that the people of Skythia were a civilised race and by no means wild nomads.

The Periplus next (chapter 40) gives an accurate account of the Ran (Eirinon) which in those days was probably below sea level (Haig, page 22, Burnes' Travels into Bokhara, III. 309ff), and was already divided into the Great and the Little. Both were marshy shallows even out of sight of land and therefore dangerous to navigators. The Ran was then as now bounded to south and west by seven islands, and the headland Barakê (Dvârakâ) a place of special danger of whose neighbourhood ships were warned by meeting with great black water-snakes.

The next chapter (41) describes the gulf of Barygaza (gulf of Cambay) and the adjoining land, but the passage has been much mangled by the copyist of our only MS. and more still by the guesses of editors. According to the simplest correction (hêrostês' Ariakêschôra) our author says that next after Barakê (Dvârakâ) follows the gulf of Barygaza and the country towards Ariakê, being the beginning of the kingdom of Mambaros and of all India. Mambaros may possibly be a corruption of Makhatrapos or some similar Greek form of Mahâkshatrapa, the title of the so-called "Sâh Kings" who ruled here at this period (A.D. 250). According to the reading of the MS. the author goes on to say that "the inland part of this country bordering on the Ibêria (read Sabiria = Sauvîra) district of Skythia is called ... (the name, perhaps Maru, has dropped out of the text), and the sea-coast Syrastrênê (Surâshtra)." The country abounded then as now in cattle, corn, rice, cotton and coarse cotton cloth, and the people were tall and dark. The capital of the country was Minnagara whence much cotton was brought down to Barygaza. This Minnagara is perhaps the city of that name placed by Ptolemy near Mânpur in the Vindhyas, but it has with more probability been identified with Junâgad (Bhagvânlâl) which was once called Manipura (Kath. Gaz. 487). Our author states that in this part of the country were to be found old temples, ruined camps and large wells, relics (he says) of Alexander's march, but more probably the work of Menandros and Apollodotos. This statement certainly points to Kâthiâvâda rather than to Mânpur. The voyage along this coast from Barbarikon to the headland of Pâpikê (Gopnâth) near Astakapra (Hâthab) and opposite to Barygaza (Broach) was one of 3000 stadia = 300 miles, which is roughly correct. The next chapter (42) describes the northern part of the gulf of Cambay as 300 stadia wide and running northward to the river Maïs (Mahî). Ships bound for Barygaza steer first northward past the island Baiônes (Peram) and then eastward towards the mouth of the Namnadios (Narmadâ) the river of Broach. The navigation (chapter 43) is difficult by reason of rocks and shoals such as Hêrônê (perhaps named from some wreck) opposite the village of Kammôni (Kim) on the eastern shore and by reason of the current on the western near Pâpikê (perhaps a sailor's name meaning Unlucky). Hence the government sends out fishermen in long boats called Trappaga or Kotumba (Kotia) to meet the ships (chapter 44) and pilot them into Barygaza, 300 stadia up the river, by towing and taking advantage of the tides. In this connection our author gives a graphic description of the Bore in the Narbadâ (chapter 45) and of the dangers to which strange ships are exposed thereby (chapter 46).

Inland from Barygaza (that is, from the whole kingdom, which, as we have seen, bordered on Sauvîra or Multân) lay (chapter 47) the Aratrioi (Arattas of the Mahâbhârata and Purânas, who lived in the Panjâb), the Arakhôsioi (people of eastern Afghanistan), Gandaraioi (Gandhâra of N.-W. Panjâb), Proklais (near Peshâwar), and beyond them the Baktrianoi (of Balkh) a most warlike race, governed by their own independent sovereigns. These last are probably the Kushâns who, when the Parthian empire fell to pieces in the second quarter of the third century, joined the Karên Pahlavs in attacking Ardeshir. It was from these parts, says our author, that Alexander marched into India as far as the Ganges--an interesting glimpse of the growth of the Alexander legend since the days of Arrian (A.D. 150). Our author found old drakhmai of Menandros and Apollodotos still current in Barygaza.

Eastward in the same kingdom (chapter 48) is the city of Ozênê; which was formerly the capital, whence onyxes, porcelain, muslins, and cottons are brought to Barygaza. From the country beyond Proklais came costus, bdellium, and spikenard of three kinds, the Kattybourine, the Patropapigic, and the Kabalitic (this last from Kábul).

We learn incidentally that besides the regular Egyptian trade Barygaza had commercial relations with Mouza in Arabia (chapter 21) with the East African coast (chapter 14) and with Apologos (Obollah) at the head of the Persian Gulf and with Omana on its eastern shore (chapter 36). The imports of Barygaza were wine, bronze, tin and lead, coral and gold stone (topaz ?), cloth of all sorts, variegated sashes (like the horrible Berlin wool comforters of modern days), storax, sweet clover, white glass, gum sandarac, stibium for the eyes, and gold and silver coin, and unguents. Besides, there were imported for the king costly silver plate, musical instruments (musical boxes are still favoured by Indian royalty), handsome girls for the harem (these are the famous Yavanî handmaids of the Indian drama), high-class wine, apparel and choice unguents, a list which shows that these monarchs lived in considerable luxury. The exports of Barygaza were spikenard, costus, bdellium, ivory, onyxes, porcelain, box-thorn, cottons, silk, silk thread, long pepper (chillies), and other wares from the coast ports.

From Barygaza our author rightly says (chapter 50) that the coast trends southward and the country is called Dakhinabadês (Dakshinâpatha): much of the inland country is waste and infested by wild beasts, while populous tribes inhabit other regions as far as the Ganges. The chief towns in Dakhinabadês (chapter 51) are Paithana (Paithan) twenty days journey south of Barygaza and Tagara (Dhârur) a very large city ten days east of Paithana. From Paithana come onyxes, and from Tagara cottons muslins and other local wares from the (east) coast.

The smaller ports south of Barygaza are Akabarou (perhaps the Khabirun of Mahomedan writers and the modern Kâvêrî the river of Nâusâri) Souppara (Supârâ near Bassein) and Kalliena, which was made a mart by the elder Saraganês, but much injured when Sandanês became its master, for from his time Greek vessels visiting the port are sent under guard to Barygaza. This interesting statement is one of the clearest indications of the date of the Periplus. As Bhândârkar has shown, the elder Saraganês implies also a younger, who can be no other than Yajñasrî Sâtakarni (A.D. 140), and the Periplus must be later than his time. The Sandanês of the text must have been a ruler of Gujarát and may be identified with the Kshatrapa Sanghadâman (A.D. 224).

South of Kalliena (chapter 53) were Sêmylla (Chaul) Mandagora (Mandangad) Palaipatmai (Pâl near Mahâd) Melizeigara (probably Janjîra) and Byzantion (Chiplun). The words which follow probably give another name of Byzantion "which was formerly also called Turannosboas," the name Toparon being a misunderstanding (Müller, Geogr. Gr. Min. I. 296). South of this are the islands of Sêsekreienai (Burnt Islands), Aigidioí (Angediva), Kaineitai (Island of St. George) near the Khersonêsos (Goa), and Leukê (Laccadives ?) all pirate haunts. Next comes Limyrikê (the Tamil country) the first marts of which are Naoura (Cannanor or Tellichery, rather than Honávar, which is too far north) and Tyndis (Kadalundi near Bepur) and south of these Muziris (Kranganur) and Nelkynda (Kallada). Tyndis and Muziris were subject to Kêprobotras (Keralaputra that is the Cera king) and Nelkynda to Pandion (the Pândya king of Madura). Muziris was a very prosperous mart trading with Ariakê (North Konkan) as well as Egypt. Nelkynda was up a river 120 stadia from the sea, ships taking in cargo at the village of Bekarê at the mouth of the river. Our author gives an interesting account of the trade at these ports and further south as well as on the east coast, but we are not concerned with this part of his work.

[Markianos.] Markianos of Hîrakleia about the year 400 A.D. is the leading geographer of the period following Ptolemy, but his work consisted chiefly in corrections of Ptolemy's distances taken from an obscure geographer named Prôtagoras. He adds no new facts to Ptolemy's account of western India.

[Stephanos.] Stephanos of Byzantium wrote about 450 A.D. (or at any rate later than Markianos, whom he quotes) a huge geographical dictionary of which we have an epitome by one Hermolaos. The Indian names he gives are chiefly taken from Hekataios, Arrianos, and especially from a poem called Bassarika on the exploits of Dionysos, by a certain Dionysos. But his geography is far from accurate: he calls Barakê (Dvârakâ) an island, and Barygaza (Broach) a city, of Gedrôsia. Among the cities he names are Argantê (quoted from Hekataios), Barygaza (Broach), Boukephala (Jalâlpur), Byzantion (Chiplun), Gêreia, Gorgippia, Darsania famous for woven cloths, Dionysopolis (Nysa ?), Kathia (Multân ?), Kaspapyros and Kaspeiros (Kasmîr), Margana, Massaka (in Swât), Nysa, Palimbothra (Pâtaliputra), Panaioura near the Indus, Patala (thirty-five miles south-east of Haidarâbâd, Sindh), Rhodoê, Rhôganê, Rhôn in Gandarikê, Saneia, Sesindion, Sinda on the great gulf (perhaps Ptolemy's Asinda, Vadnagar), Sôlimna, and Taxila. He also names a number of tribes, of whom none but the Orbitai (Makrân) the Pandai (Pândya) Bôlingæ (Bhâulingi Sâlvas) and possibly the Salangoi (Sâlankâyana) belong to the western coast.

[Kosmas.] Kosmas Indikopleustes, shipman and monk, who wrote his Topographia Christiana between A.D. 530 and 550, is the last of the ancient writers who shows independent knowledge of India. He says that Sindu (Sindh), is where India begins, the Indus being the boundary between it and Persia. The chief ports of India are Sindu (Debal), which exports musk and nard: Orrhotha (Surâshtra that is Verâval) which had a king of its own: Kalliana (Kalyân) a great port exporting brass, and sîsam (blackwood) logs and cloth having a king of its own and a community of Christians under a Persian bishop: Sibor which also had a king of its own and therefore cannot be Supârâ, which is too close to Kalliana, but must be Goa, the Sindabur of the Arabs: Parti, Mangaruth (Mangalor), Salopatana, Nalopatana, and Pudopatana which are the five marts of Malê the pepper country (Malabâr), where also there are many Christians. Five days' sail south of Malê lay Sielediba or Taprobanê (Ceylon), divided into two kingdoms in one of which is found the hyacinth-stone. The island has many temples, and a church of Persian Christians, and is much resorted to by ships from India Persia and Ethiopia dealing in silk, aloewood, cloves, sandalwood, &c. On the east coast of India is Marallo (Morava opposite Ceylon) whence conch-shells are exported: Then Kaber (Kaveripatam or Pegu. Yule's Cathay Introd. page clxxviii.) which exports Alabandinum; further on is the clove country and furthest of all Tzinista (China) which produces the silk. In India further up the country, that is further north, are the White Ounoi or Hûnas who have a king named Gollas (Mihirakula of inscriptions) who goes forth to war with 1000 elephants and many horsemen and tyrannises over India, exacting tribute from the people. His army is said to be so vast as once to have drunk dry the ditch surrounding a besieged city and marched in dryshod.

In his book XI. Kosmas gives some account of the wild beasts of India, but this part of his work does not require notice here.