Ptolemy in his map places Gerrha, the mart of ancient Indian trade and the starting-point for caravans on the great road across Arabia, on the coast just opposite the islands, near where the town of El Katif now is, and accepts Strabo's and Pliny's names for the Bahrein Islands, calling them Tharros, Tylos or Tyros, and Arados. The fact is that all our information on the islands prior to the Portuguese occupation comes from the Periplus of Nearchus. Eratosthenes, a naval officer of Alexander's, states that the Gulf was 10,000 stadia long from Cape Armozum, _i.e._ Hormuz, to Teredon (Koweit), and the mouth of the Euphrates. Androsthenes of Thasos, who was of the company of Nearchus, made an independent geographical survey of the Gulf on the Arabian side, and his statements are, that on an island called Ikaros, now Peludji, just off Koweit, he saw a temple of Apollo. Southwards, at a distance of 2,400 stadia, or 43 nautical leagues, he came on Gerrha, and, close to it, the islands of Tyros and Arados, 'which have temples like those of the Phoenicians,' who were (the inhabitants told him) colonists from these parts. From Nearchus, too, we learn that the Phoenicians had a town called Sidon or Sidodona in the Gulf, which he visited, and on an island called Tyrine was shown the tomb of Erythras, which he describes as 'an elevated hillock covered with palms,' just like our mounds, and Erythras was the king who gave his name to the Gulf. Justin accepts the migration of the Phoenicians from the Persian Gulf as certain; and M. Renan says, 'The primitive abode of the Phoenicians must be placed on the Lower Euphrates, in the centre of the great commercial and maritime establishments of the Persian Gulf.'[5] As for the temples, there are no traces of them left, and this is also the case in Syrian Phoenicia; doubtless they were all built of wood, which will account for their disappearance.