1849.] On the Physical Geography of the Himalaya. 769 



Konki the Bagmatti, the Gumti, the Cosilla and the Ramganga. Here 

 is both positive and negative evidence in favour of the doctrine, I 

 advocate as furnishing the key to the aqueous system and natural di- 

 visions of the Himalaya ; for, the upper rivers do, and the lower rivers 

 do not, stand exposed to the influence of the great peaks. 



The petty streams of the lower region or that next the plains, which 

 water the Dhuns vel Maris, traverse those valleys lengthwise ; and, as 

 the valleys themselves run usually parallel to the ghat line of the 

 snows, such is also the direction of these petty streams. In the cen- 

 tral, as in the western,* hills they usually disembogue into the rivers 

 of the first class. 



I have observed that the three great river basins of the Karnali, 

 Gandak and Cosi extend throughout Nepal ; and truly so ; for a river 

 basin, includes the widest space drained by its feeders. But, it results 

 necessarily from the manner in which the deltic basins of the Himalay- 

 an rivers are formed, that there should be intervals between the plain- 

 ward apices of these deltic basins. Of these intervals the most con- 

 spicuous in Nepal, is that which intervenes between the Cosi and Gan- 

 dak. This tract, watered by the Bagmatti, deserves separate mention 

 on many accounts, and it may be conveniently styled the valley region, 

 since it contains not only the great valley of Nepal proper, but also the 

 subordinate vales of Chitlong, Banepa, and Panouti. 



It has been already remarked that the classifications of physical geo- 

 graphy, as of the other sciences, do not constitute a perfect " open 

 sesame" to the mysteries of nature, but only a material help to their 

 study. This observation I will illustrate by a few comments on the 

 basin of the Tishta, lest the somewhat anomalous instance of that basin, 

 should be captiously quoted to impugn the doctrine I contend for ; but 

 contend for, not as exhibiting in every instance an absolute conformity 

 with natural arrangements, but as doing all that can be reasonably ex- 

 pected in that way, and as furnishing, upon the whole, a generally 

 truthful, causally significant, and practically useful, indication of those 

 arrangements. 



I have stated above that the basin of the Tishta extends from the 

 peak of Kangchang to that of Chumalari. But an inspection of the 

 accompanying map will show that between these two peaks there occurs 



* J. A, S. No. 126, p. xxxin. 



5 G 



