1873.] 



DREW UPPER-INDUS BASIN. 



453 



towards e marks the new or secondary fan projecting in front of the 

 cliff' b. 



As it is always satisfactory to refer to actual cases, I annex a 

 sketch, fig. 9, taken in the Changchenmo valley (the altitude of 



Fig. 9. — Fan cut into by the River, ivith secondary Fan formed in 

 front (four miles south of Pamzalan, Changchenmo, Ladakh). 



which is 15,000 feet), of a compound fan which is a fair example of 

 the action which I have been describing. 



It is likely that the cutting of a notch in the cliff begins early in 

 the formation of the cliff; but it is not likely to become the per- 

 manent bed of the stream until the cliff has been worked a good 

 way back. One often sees many such notches at the end of gullies 

 now dry that once held a stream : but we must also remember that 

 there is a tendency of the waters to separate; so that only part may 

 have gone to produce them ; it is not until the denudation of the 

 gully has gone back to the gorge that the course and the holding 

 together of the waters are determined for the future. It may be 

 remarked that there is no instance, as far as I know, of more than 

 one deep gully leading from the cliff to the mouth of the ravine, 

 although, as in some of the fans opposite Pitak, near Leh, small 

 secondary fans project from several of the radiating water-courses 

 without the gully having been denuded far back. 



Thus far we have dealt with cases in which neither the fan- 

 stream nor the mountain has had any tendency of itself to lower its 

 bed ; for in the last instances the fan-stream would not have cut 

 through its bed but for the accident of the river, by its side action, 

 making a cliff which necessitated a readjustment of the slope of the 

 bed of the ravine. But we often enough meet with evidence of the 



