The fleet, on weighing from the haven, ran a course of 60 stadia, and anchored under shelter of a desert island called Domai. Next day, with a run of 300 stadia, it reached Saranga, and on the following day anchored at a desert place called Sakala. Another run of 300 stadia brought it on the morrow to Morontobara or Women’s Haven. This haven had a narrow entrance, but was deep, capacious, and well-sheltered. The fleet, before gaining the entrance, had passed through between two islets, which lay so close to each other that the oars grazed the rocks on each side. On leaving this harbour next day it had on the left a tree-covered island 70 stadia long which sheltered it from the violence of the sea. As the channel, however, which separated the island from the mainland was narrow, and shoal with ebb-tide, the passage through it was difficult and tedious, and it was not till near the dawn of the following day that the fleet succeeded in clearing it. A course of 120 stadia brought it to a good harbour at the mouth of the Arabios. Not far from this harbour lay an island described as being high and bare. The island of Domai Colonel Holdich and others would identify with Manora. Manora, however, Haig points out, is even now 4 to 5 miles off from the nearest mainland, and must have been further in Alexander’s time. He would, therefore, place Domai rather more than 4 miles due west of the town of Karâchî, or perhaps further north. The fleet, in its course to Saranga, must have rounded Cape Monze or Râs Muâri, but this projection is not mentioned by Arrian. The position of Saranga, to judge from the recorded length of the run, must have been near the mouth of the Hub river, which is 26 miles distant from the position assigned to Domai. The Hub mouth has been silted up, and this led, last century, to its port being abandoned. Our author points out that if Κ were substituted for Σ in Saranga, we would then have in Karanga a very fair representation of Kharok, the name of the Hub port. However this may be, he adds, there can be no doubt that the Saranga of Nearchos was either at the Hub mouth or a few miles further north. He then corrects a mistake into which Dr. Vincent and myself had both of us fallen in our respective translations of the record of the next part of the voyage—that from Saranga to Sakala, and thence to Morontobara. Our versions represented the two rocky islets, between which the fleet passed instead of taking a circuitous course out in the open sea, as being in the neighbourhood of Sakala instead of that of Morontobara. Sakala, Haig thinks, may be placed a little east of Bidok Lak—a place 24 miles distant from Saranga, if Saranga be taken to lie a few miles north of the Hub mouth. Between these two places the fleet must have passed the island of Gadâni, which is now a part of the mainland, and was probably the Kodanê of Ptolemy.