part 3] OF THE IMUS, BEAU MAL'l'TKA, AM) GANGES. 140 



of the portion captured by the Sutlej, due to the advance westwards of the 

 invading Tsangpo, is perhaps responsible for the insignificance of the present 

 stream at the bottom of the deep canons cut in the thick alluvial valley of 

 Ngari Khorsum. 



(5) That the Attock tributary of the Indobrahm, perhaps at an early date, 

 captured the Tibetan River in Gilgit. 



It will be seen that the parallelism between the old Tibetan 

 River and the Himalayan part of the Indobrahm is not confined to 

 their position and direction, but also applies to their age, origin, 

 and history. They both appear to have supplanted marine gulfs 

 along the western part of their courses at the same period ; both 

 seem to have been the result of the Himalayan movement, to which 

 their courses are at right angles ; in both cases the eastern half of 

 the stream has been completely reversed in direction by capture on 

 the part of an invading river. 



As an indication of a north-westward-flowing main river along 

 the Himalayan foot, Dr. Pilgrim cites the local V-shaped course 

 of many of the present rivers from the mountains as they cross 

 the Siwalik belt, the angle- of the V pointing north-westwards and 

 occurring usually close to the boundary-line between the Siwaliks 

 and the Alluvium. The northern limb of each V may be regarded 

 as the remnant of a right-bank Indobrahm tributary, which has 

 persisted in its old westerly direction, and has become more deeply 

 impressed and permanent owing to the upheaval of the Siwalik 

 deposits over which it flows. The southern limb of the V would 

 represent the final position which the capturing stream has taken 

 up : for, if the deduction that the actual capture happened sub- 

 sequent to Siwalik times be correct, it must have taken place 

 within the outcrop of the Alluvial belt — that is to say, the point 

 of the V at the time of capture was west of where it now is, but 

 has since worked eastwards until it met the more stable topography 

 of the Siwalik belt, where its regression was in most cases held up. 

 Figs. 1-4 (pp. 150-51) illustrate this point in some of the larger 

 rivers, showing the general westward course through the Siwalik 

 belt and the more southerly tendency in the Alluvium ; the Sutlej 

 seems to be an exception, unless the point of the V has worked 

 back into the centre of the Siwalik belt. 1 



At the close of the Siwalik and during the Recent period the 

 left-bank affluents of the lower Indobrahm, the Jhilam, Chinab, 

 Ravi, Beas, Sutlej, and Grhaggar (assisted perhaps by the 

 continued earth-movements) cut back across the plains of the 

 Punjab in a north-easterly direction, and captured various portions 

 of the Indobrahm, perhaps approximately in the order named: 

 not only because the distances to be cut back were in the same 

 order, but also because the suggested geosyuclines along which 

 these rivers cut would probably be initiated in this order. 

 Whatever the precise order was, the Ghaggar must have become 



1 The higher parts of the courses of these mountain rivers the portions 

 traversing the older pre-Tertiary rocks — are not relevant to this question. 



