1848.] through Afghanistan and India. 35 



course by about 20 miles ; and that the ancient Pataliputra must have 

 stood at the same distance to the N. of the present Patna. It is only 

 by a supposition of this kind that the recorded distances of Fa Hian 

 and Hwan Thsang can be reconciled with the truth. The very fact 

 that the town, which Fa Hian had seen flourishing in A. D. 399-415, 

 was in ruins in A. D. 629-645, seems to point to its desertion from the 

 encroachments of the river to the south. Since then 1200 years have 

 elapsed ; a period much more than sufficient for the production of the 

 supposed change by the gradual and successive alterations of channel 

 towards the south, a process which is still going on. I do not however 

 attribute this change of course entirely to the gradual alteration of the 

 channel of the Ganges ; for it is probable that the mention by Ma- 

 twan-lin, that about A. D. 756 "the bank of theHo-LANG or Ganges 

 gave way and disappeared," refers to some sudden change in the course 

 of the river. An extraordinary flood of the Gogra river would have been 

 sufficient to have caused the whole amount of southing here contended 

 for ; in proof of which I will only cite the much greater change in the 

 course of the Satlaj which took place about A. D. 1790. This was 

 caused by a cataclysm of the river, which having been dammed up by 

 a landslip near the hot springs of Seoni, 18 miles to the N. of Simla, 

 suddenly burst through the obstruction, and swept irresistibly over the 

 plains until it was stopped by the high bank of the Byas at Hari-ki- 

 patan. The new channel became a permanent one, and the junction of 

 the Byas and Satlaj, which was formerly at Ferozpur, has since then 

 been at Hari-ki-patan, upwards of 30 miles from the old place of con- 

 fluence.) 



(From Pataliputra Hwan Thsang proceeds to Gaya, of which he gives 

 many minute details, that could only be verified by personal inspection 

 or by a very good map on a large scale. Some of them however may 

 easily be identified: Such as the river Ni-lian-chen-na, to the E. of 

 Gaya, which is clearly the Nilajni river of the Government litho- 

 graphed map of the new road. Also the river Ma-ho to the E. of 

 which was a great forest, is certainly the Makona river, on the E. of 

 which Rennell places "Woods" extending for more than 20 miles. 

 After some further details Hwan Thsang mentions the town of Kolo- 

 che-ku-li-sse, " demeure royale," which is undoubtedly the ancient 

 Rajagriha, or " royal residence." I remark here, as in No. 46, the occur- 



f 2 



