obstacle which Nearchos had encountered. It was not a rock, however, but probably an oyster-bank, for when search was made for it afterwards during a survey it was no longer to be found. The island of Krokala, which General Cunningham erroneously identified with the island of Kîâmârî, which lies in front of Karâchî, no longer exists as an island, but forms part of the mainland. It lay at the mouth of the Gisri Creek, by which the Malîr river pours its waters into the sea. The headland which Arrian calls Eiros is to be identified with the eminence called “Clifton,” the eastern headland of Karâchî Bay, the “narrow creek” which the fleet entered on leaving Krokala, is Chinî Creek, which leads into Karâchî Bay and harbour. Kîâmârî thus corresponds with the “low, flat island” of the Greek narrative, while Manora (mistaken by Cunningham for Eiros), exactly corresponds with Bibakta. We must now briefly notice what is said regarding the eastern portion of the Delta. Here the most important of all the forsaken channels of the Indus is the Purân, which can still be clearly traced from two different starting-points in Central Sindh, one 24, the other 36 miles north-east of Haidarâbâd. The two head channels run south-east for about 50 miles, and unite at a spot 45 miles east by south from Haidarâbâd. The single channel has then a course of over 140 miles to the head of the Korî Creek, the last 50 miles being through the Ran of Kuchchha. The eastern arm of the Indus, which Alexander in person explored, was probably some channel running into the Purân not far above the point where it enters the Ran. On reaching the sea by this eastern branch, Alexander, as Arrian informs us, landed, and with some cavalry proceeded three marches along the coast. This statement the _Memoir_ declares to be a fabrication, since such a march would be an utter impossibility. At the same time, the notion of wells being dug in the locality is scouted as an absurdity.