330 R. D. Oldbam — Ou probable Chanrjes in the Geography [No. 4, 



certain tliat the Indus, or one branch of it, must have at one time flowed 

 through the gap near Aror, it is equally certain that nothing but the 

 flood waters then or since have flowed eastwards past Aror to the Narra, 

 and that for the last 1100 years at least the Indus flowed west of the low 

 range of hills running southwards from Sukkur and Aror. The tract 

 of country between these hills and the range to the west is on the map 

 a simple network of deserted river channels, and it will be hopeless to 

 attempt to determine with accuracy which of these wms the river course 

 at any one particular period. 



§ 4. Yet, though the Eastern Narra is not a deserted bed of tha 

 Indus, it seems probable that as late as the eleventh century it was 

 occupied by a flowing river. My witnesses to prove this are, 1st, the 

 Chachnama, and, 2nd, the Beglarnama, both translated in the first 

 volume of Prof. Dowson's edition of Sir H. Elliot's History of India. 



When Muhammad Kasim invaded Sindh, he sent his mangonels up 

 the river to Niran, and, after receiving the submission of that place, he 

 determined to go against Sehwan, and after its capture to " recross the 

 river "* and proceed against Dahir ; from this it is evident that he must 

 have crossed one of the main branches of the Indus, thus confirming 

 other accounts Avhich place Niran between the two main branches of 

 the Indus. After the capture of Sehwan, he returned to I^iran, where 

 he crossed the Mihran by a bridge of boats, and went against Dahir ; 

 after crossing the river and defeating Dahir's troops, whom they pursued 



* as far as the gates of Jham,' the Arab army marched on till it reached 



* the fort of Bait,' where an entrenched camp was formed. Muhammad 

 Kasim then advanced towards Rawar and came to a " lake," but, as this 

 had to be crossed by a boat, it was probably a branch of the river ; after 

 crossing he advanced a day's march ^nd came to " Jewar on the banks of 

 the Wadhawah (or according to another MS. Dadhawah").t After his 

 defeat by Muhammad Kasim, Dahir took refuge in the fort of Rawar, 

 which was but a day's march from ' Jewar on the Wadhawah,' and which 

 seems itself to have been on the Wadhawah, for, among the administra- 

 tive arrangements made by Muhammad Kasim before he marched north- 

 wards, it is stated that he placed " Nuba, son of Daras, in the fort of 

 Rawar and directed him to hold the place fast and keep the boats ready. 

 If any boat coming up or down stream was loaded with men or arms of 

 war, he was to take and bring them to the fort of Rawar. "J From this 

 it is evident that Rawar was on a navigable stream, and it remains to 

 identify this if possible. 



Elphinstone has placed Rawar on the Indus, but tliis was clearly 



* Elliot, op. cit, 1, 158. t Op. cit., 1, 1G8. t Op. cit., I, 189. 



