I now proceed to consider the sources of the information which are the classical writers, Indian literature, Tamil and Sanskrit, and the

1 Pattinappalai, 134-6, Silappadhikaram, Canto. VI, 11 2-39.

2 Manimekhalai, Canto XIX, 1. 107 ff.

3 Ibid, XVII, 1. 145. This has reference to the small temple of Champapati the Guardian-deity of Jambudvipa. The Tamil kuchchara can have a number of equivalents in Sanskrit and Prakrit one of which of course is Gurjjara. If it is proved that the Gnrjjaras were unknown in India before the end of the fifth century A. D. this equation with Gurjjara will have to be given up. Apart from this it is possible we get a more satisfactory equivalent. Either way this cannot be held to be a decisive test of chronology. 4 Silappadhikaram, Canto XV, 11. 207-17,

Ceylonese chronicle. Of the first group, Strabo wrote in the reigns of Augustus and Tiberius; Pliny published his geography in A. D. 77; the Periplus of the Erythraean Sea was written in the first century A. D., probably A. D. 60; Ptolemy wrote his geography about A. D. 150; the Peutingerian Tables were composed in A. D. 232. There were other writers who wrote later, but we are not concerned with them directly. I would draw attention to three points, taken from the works of classical writers. Pliny remarks : At the present day voyages are made to India every year, and companies of archers are carried on board, because the Indian seas are infested by pirates. Later on he says: It (Muziris) is not a desirable place of call, pirates being in the neighbourhood, who occupy a place called Nitrias, and besides it is not well supplied with wares for traffic. This was before A. D. 77. Ptolemy regarded this port Muziris as an emporium, and places the country of Aioi south of Bakarai. Though Ptolemy does mark the division of the Konkan coast extending northwards of Nitra (Nitrias of Pliny) and up to the port of Mandagara which is identified with some place not yet definitely accepted in the southern Maharatta country north of Goa, as Ariake Andron Peira-tion, meaning the Ariake of the pirates in his time, says no more of pirates at all, meaning there was no piracy at the time to which his work relates, a period not far from him. The Periplus on the contrary does make mention of the piratic character of this coast and gives a straightforward account of its active prevalence at the time in regard to the ports in the neighbourhood. The bearing of this we shall see presently.1

1 The following account from Marco Polo of this coast is worth noting: - There go forth every year more than a hundred corsair vessels on cruise. These pirates take with them their wives and children and stay out the whole summer. Their method is to join in fleets of 20 or 30 of these pirate vessels together, and then they form what they call a sea cordon, that is they drop off till there is an interval of 5 or 6 miles between ship and ship, so that they cover something like a hndred miles of sea, and no merchant ship can escape them. For when any one corsair sights a vessel a signal is made by fire or smoke and then the whole of them make for this and seize the merchants and plunder them. After they have plundered them they let them go saying go along with you and get more gain, and that may-hap will fall to us also.

He also notes in respect of the kingdom of Eli the following :-

the commerce was with the south-west coast only, and not with the interior. He differs from those who find an explanation of this fluctuation in the political and social condition of India itself, and the facilities or their absence for navigating the seas; and considers that the cause is to be sought for in the political and social condition of Rome.

If any ship enters their estuary and anchors there having been bound for some other port, they seize her and plunder the cargo. For they say, you were bound for somewhere else, and it is God has sent you hither to us, so we have a right to all your goods. And they think it is no sin to act thus. And this naughty custom prevails all over the provinces of India, to wit, that if a ship is driven by stress of weather into some other port than that to which it was bound, it was sure to be plundered. But if a ship came bound originally to the place they receive it with all honour and give it due protection.

It would be interesting to note, as Yule remarks, that it was in this neighbourhood that Ibn Batuta fell into the hands of pirates and was stripped to the very drawers." That region continued to be piratical up to the days of Clive and Watson as we know. In the days of Sivaji it continued to be piratical also, as he is said to have replied to an English embassy protesting against this piracy that " it was

The Peutinger an Tables state clearly that two Roman Cohorts were maintained in the same town for the protection of Roman commerce.

Mr. Sewell who has made an elaborate study of the Roman coins found in India, considers that an examination of the coin finds leads to the following conclusions1:-

1. There was hardly any commerce between Rome and India during the Consulate.

2. With Augustus began an intercourse which enabling the Romans to obtain oriental luxuries during the early days of the empire, culminated about the time of Nero, who died A.D. 68.

3. Prom this time forward the trade declined till the death of Caracalla, A.D. 217.

4. From the death of Caracalla it almost entirely ceased.

5. It revived again, though slightly, under the Byzantine emperors.

He also infers that the trade under the early emperors was in luxuries; under the later ones in industrial products, and under the Byzantines against the laws of Conchon, to restore any ship or goods that were driven ashore. " The Central Asian Ambassador Abd-er-Razzak has something to say of pirates near the Calient coast. Marco-Polo, Yule and Cordier. (3rd Edn.) 1ll Chap. XXIV and XXV, pp. 385.392. 1 J. R. A. S. 1904, p. 591.