The classical geographers, the author of the Periplus and Ptolemy the geographer, that date respectively about A. D. 80 and A. D, 150, exhibit knowledge of a division of the country that we derive from the Tamil classical literature. The author of the Periplus2 begins his account of the west of India with the Indus (Sinthus). He says that the river had seven mouths, shallow and marshy, and therefore not navigable. On the shore of the central channel was the sea-port Barbaricum with a capital in the interior of the Scythians called Minnagara (the city of the Min, Scythians); the port Barbaricum has not satisfactorily been identified. It seems to be the Sanskrit Barbaraka (belonging to the country of the Barbara, perhaps the same as the Gk. barbarian). Passing down from there, the Peri-plus comes down the Surashtra coast (Syras-trene), and the Rann of Cutch (Eirinon); sailing across what is the Gulf of Kambay, he takes us to Barygaza (Sans. Brgukachcha, Mod. Broach). With this is supposed to begin Ariaca "which is the beginning of the kingdom of Nambanus and of all India." In regard to the divisions of that part of the country both Ptolemy and the Periplus agree except for the omission of some in the latter. The southern limit of the coast of Ariaca is Tindis according to both. The corresponding portion of the country inland is described in the Periplus as Abhira, the coast portion being Surashtra, as was already stated. This part is described as a fertile country producing wheat, rice, sesame oil and clarified butter; cotton and coarser sorts of cloth made thereform. Pasturing of cattle seems an important occupation and the people are described as of great stature and dark in colour.1 The chief point to note here in

1 Ll. 214-15; also Silappadhikaram 11. 9-12, The term Gavanar is rendered Sonagar by the earlier and mlecchar by the latter of the two commentators.

2 The Periplus of the Erythraean Sea, translated and edited by W. H. Schoff, Secs. 42 to 66.

1 Note the tradition that Agaetya took with him a large colony of people from here in hie southward migration above connection with this statement of the Periplus is that the coast under reference is described as the beginning "of the kingdom of Nambanus and of all India." The latter expression indicates clearly that whoever Nambanus was, he was, at the time that the author of the Periplus got his information, known to the outside world as the king of India. In other words, it seems to have been the days of the Andhra empire of Magadha. The name Nambanus itself is a correction of the text which has Mambarus. This latter might well be the Lambodara of the pauranic list of the Satavahanas or the Andhras of the Dakhan. The chronology of the early rulers of these Satavahanas cannot yet be regarded as being definitely settled, and at any rate the expression in the text seems of very doubtful application to identify Nambanus with Nahapana, the Ksaharata ruler. After describing the difficulties of navigating up to the port of Broach and the arrangement made by the ruler for piloting the vessels safely into the port, the Periplus proceeds to give the countries inland set over against that coast between Barbaricum, at the mouth of the Indus obviously, and Broach. He notes among them the Arattas of the Punjab, the Arachosii of Southern Afghanistan, the Gandaraei (Sanskrit, Gandhara), and the people of Pocalais (Sans. Pushkalavati) both in the region between the Kabul and the

Indus in Northern Afghanistan including also the Northern portions of the Punjab where was also the city of Alexandria which Bucephalus located very near the Jhelum. Beyond these he says were the warlike Bactrians. He gives an interesting fact that in his day coins bearing Greek in-scriptions or Greek legends were circulating in the country round Broach, and they contained, according to the Periplus, the devices of the Greek rulers, among them, Apollodotus and Menander. Oomingfurther east from these countries he speaks of Ozene (Ujjain), and refers to it as the former royal capital. Passing over all that he says about the trade of Broach which is not to our present purpose, we come, in Sec. 50, to another statement which is of immediate interest to us. He says "beyond Barigaza the adjoining coast extends in a straight line from north to south and so this region is called Dachinabades, for Dachan in the language of the natives means ' south.' The inland country back from the coast towards the east comprises many desert regions and great mountains; and all kinds of wild beasts, leopards, tigers, elephants; enormous serpents, hyenas, and baboons of many sorts, and many populous nations as far as the Ganges." This clearly indicates that he describes the whole of the region known as the Dakshinapatha or the Dakhan, and the Dandakaranyam of the Sanskrit writers; the central region of India corresponding to our modern division of the Dakhan. He then describes the interior marts of Paitan and Tagara, and of the sea-ports along the coast till he reached Naura and Tindis, the first marts of Damirica as he calls them (Sanskrit Dramidaca, the correct equivalent of the Greek), and the Tamilakam of the Tamil classics. Damirica, sometimes written by error Lymirica, is the Sanskrit Dramidaka which the author must have heard in contradistinction to Aryaka. It is perhaps a little far-fetched to see in it Tamilakam except through the Sanskritised Dramidaka. With Tindis began according to both Ptolemy and the Periplus, the kingdom of Cherabothra (Cheraputra or Keralaputra). The next port of importance we come to, is 50 miles from Tindis again at the mouth of a river; the port called Muziris (Muyiri or Musiri of the Tamils, the modern Cranganore). Fifty miles further south was the sea-port of Nelcynda which the late Mr. Kanakasabhai Pillai correctly identified with Nirkunram in the country of the Pandyas. This place was situated about ten or twelve miles in the interior with an out-port at the mouth of the river, the village Bakare, Vaikkarai as we know it now. The kings of both these market towns, the Periplus says "live in the interior." The imports into Muziris are given "as a great quantity of coin; topaz, thin clothing not much, figured linens, antimony, coral, crude glass, copper, tin, lead, wine not much but as much as at Barigaza; realgar and orpiment, and wheat only for the sailors, for this is not dealt in by the merchants there." The exports from this place are the "pepper" coming from "Kottanora" (Kutta Nadu in the interior) "great quantities of fine pearls," ivory, silk cloth, spikenard from the Ganges, mala-bathrum from the interior, transparent stones of all kinds, diamonds and sapphires and tortoise-shell, "that from the Chryse (golden) island and that taken from among the islands along the coast of Damirica." One may so far compare this statement with the following two extracts from Tamil Literature:-