120 MIDDLBM1SS PHYSICAL GEOLOGY OF SUB-HIMALAYA. 



tor a long distance both U. and M. Siwaliks are absent. Beyond the 

 Kho M. Siwaliks set in. As we near the R£mganga the narrow 

 band of Siwalik conglomerate running from the Kotri stream to the 

 former river indicates the very opposite of energetic torrent and 

 river action in accumulating thick beds of coarse detrital material. 

 On the other hand, at the debouchure of the R^mganga and the 

 Kosi, we have once more a great thickening of the conglomerate ; 

 the wandering of the Ra*mganga to the west along the P£tli dun 

 leaves in its train all that thick low range of conglomerate to the 

 south as an indelible mark of its former activity. Again, the Kotah 

 dun conglomerates are so palpably an ancient river fan spreading 

 out from the present Kosi river-bed that they need no remark. Their 

 eastern boundary at Ka*ladha*ngi in a similar manner marks the 

 limit of the Kosi's former activity in that direction : for there are no 

 more conglomerates at the foot of the high Naini Tcil hills where 

 large rivers are absent. 



Nothing is more clearly demonstrated in the whole range of Sub- 

 Himalayan geology than the connection between the position of the 

 debouchure of the present large rivers and the deposits of Siwalik con- 

 glomerate. That being so, we are bound to believe that these depo- 

 sits were formed by the direct parents of those rivers, in the places 

 where they are now found ; and it would be as impossible to credit 

 the belief that the conglomerates could once have extended far into 

 the hills as it would be to find the Bha*bar deposits in a similar 

 locality. 



Apart from these considerations, I have shown that the northern 

 boundary of the conglomerate becomes undoubtedly an approximate 

 limit of deposition in the section south of the Sanguri s6t, and ac- 

 tually a real limit on the north side of the Kotah dun. Thus, the 

 configuration of the southern margin of the Himalaya must have been 

 very much what it is now in Upper Siwalik times. The absence of 

 outliers to the north of the boundary is negative evidence favouring 

 the same conclusion ; but too obvious to be more than mentioned. 



If we now take the fault separating the sand-rock from the 

 ( «78 ) 



