called Salghéda; its fituation answers exactly to the defcription given of it by Alexander's hiftorians. The kings of Sangala are known in the Perfian history by the name of Schangal, one of them affifted Afrafab against the famous Caicofru; but to return from this digreffion to Patali putra.

The true name of this famous place is, Patali-purȧ; which means the town of Patali, a form of Devi wor fhipped there. It was the refidence of an adopted fon of the goddess Patali, hence called Patali-putra, or the fon of Patali. Patali-putra and Bali-putra are abfolutely inadmiffable, as Sanferit names of towns and places; they are used in that fenfe, only in the spoken dialects; and this, of itself, is a proof, that the poems in queftion are modern productions. Patali-pura, or the town of Patali, was called fimply Patali, or corruptly Pattiali, on the invafion of the Muffulmans : it is mentioned under that name in Mr. Dow's tranflation of Ferishta's history. It is, I believe, the Patali of Pliny. From a paffage in this author compared with others from Ptolemy, Marcianus, Heracleota, and Arrian in his Periplus, we learn that the merchants, who carried on the trade from the Gangetic Gulph, or Bay of Bengal, to Perimula, or Malacca, and to Bengal, took their departure from fome place of rendezvous in the neighbourhood of Point Godavery, near the mouth of the Ganga Godavery. The fhips ufed in this navigation, of a larger conftruction than common, were called by the Greek and Arabian failors, colandrophonta, or in the Hindustani dialect, coilan-di-pota, coilan boats or fhips: for pota in Sanfcrit, fignifies a boat or a fhip; and di or da, in the western parts of India, is either an adjective form, or the mark of the genitive cafe. Pliny has preferved to us the track of the merchants who traded to Bengal from Point Godavery.

They went to Cape Colinga, now Palmira; thence to Dandagula, now Tentu-gully, almoft oppofite to Fultati*; thence to Tropina, or Triveni and Trebeni, called Tripina by the Portuguese, in the laft century; and, laftly, to Patale, called Patali, Patiali as late as the twelfth century, and now Patna. Pliny, who miftook this Patale for another town of the fame name. fituate at the fummit of the Delta of the Indus, where a form of Devi, under the appellation of Patali is equally worthipped to this day, candidly acknowledges, that he could by no means reconcile the various accounts he had feen about Patale, and the other places mentioned before.

The account tranfmitted to us of Chandra-Gupta, by the hiftorians of Alexander, agrees remarkably well with the abftract I have given in this paper of the Mudra Rácfhafa. By Athenæus, he is called Sandracoptos, by the others Sandracottos, and fometimes Androcottos. He was alfo called Chandra fimply; and, accordingly, Diodorus Siculus calls him Xandrames from Chandra, or Chandram in the accufative cafe ; for in the western parts of India, the spoken dialects from the Sanfcrit do always affect that cafe. According to Plutarch, in his life of Alexander, Chandra-Gupta had been in that prince's camp, and had been heard to fay afterwards, that Alexander would have found no difficulty in the conqueft of Práchi, or the country of the Prafians had he attempted it, as the king was despised, and hated too, on account of his cruelty.

In the Mudra Rácfhafa it is faid, that king Nanda, after a fevere fit of illness, fell into a ftate of imbecillity, which betrayed itself in his difcourfe

*This is the only place in this effay not to be found in Rennell's Atlas.

and

and actions; and that his wicked minifter, Sacatara, ruled with defpotic fway in his name. Diodorus Siculus and Curtius relate, that Chandram was of a low tribe, his father being a barber. That he, and his father Nanda too, were of a low tribe, is declared in the Vishnu-purána and in the Bhagavat Chandram, as well as his brothers, was called Maurya from his mother Mura; and as that word * in Sanfcrit fignifies a barber, it furnished occafion to his enemies to afperfe him as the fpurious offspring of one. The Greek hiftorians fay, the king of the Prafu was affaffinated by his wife's paramour, the mother of Chandra; and that the murderer got poffeffion of the fovereign authority, under the fpecious title of regent and guardian to his mother's children, but with a view to deftroy them. The puránas and other Hindu books, agree in the fame facts, except as to the amours of Sacatara with Mura, the mother of Chandra-Gupta, on which head they are filent. Diodorus and Curtius are mistaken in faying, that Chandram reigned over the Prafu, at the time of Alexander's invafion he was contemporary with Seleucus Nicator.

I fufpect Chandra-Gupta kept his faith with the Greeks or Yavans no better than he had done with his ally, the king of Nepal; and this may be the motive for Seleucus croffing the Indus at the head of a numerous army; but finding Sandro-coptos prepared, he thought it expedient to conclude a treaty with him, by which he yielded up the conquefts he had made; and, to cement the alliance, gave him one of his daughters in marriage. Chandra-Gupta appears to have agreed on his part to furnish

* See the Jutiviveca, where it is faid, the offspring of a barber, begot by ftealth, of a female of the Sudra tribe, is called Maurya: the offspring of a barber and a flave woman is called Maurya.

Strabo, B. 45, p. 724.

S 3

Seleucus

Seleucus annually with fifty elephants; for we read of Antiochus the Great going to India, to renew the alliance with king Sophagafemus, and of his receiving fifty elephants from him. Sophagafemus, I conceive, to be a corruption of Shivaca-Séna, the grandson of Chandra-Gupta. In the puránas this grandfon is called Afecavard-dhana or full of mercy, a word of nearly the fame import as Aféca-féna or Shivaca-féna; the latter fignifying he whofe armies are merciful do not ravage and plunder the country,

The fon of Chandra-Gupta is called Allitrochates and Amitrocates by the Greek hiftorian. Seleucus fent an ambaffador to him; and after his death the fame good intelligence was maintained by Antiochus the fon or the grandfon of Seleucus. This son of Chandra-Gupta is called Varifára in the puránas; according to Parafara, his name was Dafuratha; but neither the one nor the other bear any affinity to Amitrocates this name appears, however, to be derived from the Sanfcrit Mitra-Gupta, which fignifies Javed by Mitra or the Sun, and therefore probably was only a furname.

It may be objected to the foregoing account, the improbability of a Hindu marrying the daughter of a Yavana, or, indeed, of any foreigner. On this difficulty I confulted the Pundits of Benares, and they all gave me the fame anfwer; namely, that in the time of Chandra-Gupta the Yavanas were much refpected, and were even confidered as a fort of Hindus though they afterwards brought upon themfelves the hatred of that nation by their cruelty, avarice, rapacity, and treachery in every tranfaction while they ruled over the western parts of India; but that at any rate the objection did not apply to the cafe, as Chandra-Gupta himself was a Sudra, that is to fay, of the loweft clafs. In the

Vishnu

Fishnu-purána, and in the Bhagawat, it is recorded, that eight Grecian kings reigned over part of India. They are better known to us by the title of the Grecian kings of Bactriana. Arrian in his Periplus, enumerating the exports from Europe to India, fets down as one article beautiful virgins, who were generally fent to the market of Baroche. The Hindus acknowledged that, formerly, they were not fo ftri&t as they are at this day; and this appears from their books to have been the cafe. Strabo does not pofitively say that Chandra-Gupta married a daughter of Seleucus, but that Seleucus cemented the alliance he had made with him by connubial affinity, from which expreffion it might equally be inferred that Seleucus married a daughter of Chandra-Gupta ; but this is not fo likely as the other; and it is probable the daughter of Seleucus was an illegitimate child, born in Perfia after Alexander's conquest of that country.

Before I conclude, it is incumbent on me to account for the extraordinary difference between the line of the Surya Varfus or children of the fun, from Ichfwacu to Dafaratha-Rama, as exhibited in the fecond volume of the Afiatick Researches, from the Vishnu-purana and the Bhagawat, and that fet down in the Table I have given with this Effay. The line of the Surya Varfas, from the Bhagawat being abfolutely irreconcileable with the ancestry of Arjuna and Chrishna, I had at firft rejected it, but, after a long fearch, I found it in the Ramayen, fuch as I have represented it in the table, where it perfectly agrees with the other genealogies. Dafaratha-Rama was contemporary with Parafu Rama, who was, however the eldeft; and as the Ramayen is the hiftory of Dafaratha-Rama, we may reasonably fuppofe, his ancestry was carefully fet down and not wantonly abridged. I fhall now conclude this Effay with the following

remarks:

S 4

I. It