146 DK. E. II. PASCOE 0^ THE EABLT HISTORY [vol. lxxv r 



(4) This Siwalik belt is succeeded on the side towards the peninsula, and 

 remote from the moxmtain-building. by a similar and even thicker belt of 

 recent river -deposits, to be mentioned later. 



(5) Many of the streams draining the Himalaya hare north-westward 

 pointing V's in their course where they cross the Tertiary foothills, or where 

 they enter the plain. These are thought to indicate, according to Dr. Pilgrim'.- 

 suggestion, a former north-westward-flowing main river, of which these 

 streams were tributaries (see below). 



(6) The similarity of the river-fauna in the Ganges and in the Indus points 

 to a connexion best explained on the assumption of a large main river uniting 

 what are now the Ganges and Indus basins (see below). 



(7) The similarity in nature and strike between the Shillong Plateau and 

 the northern region of the Indian peninsula makes it probable that they 

 represent what was once a continuous feature, the flank of a long river-valley. 

 It is this continuity of the physiographical and geological features along this 

 line from Assam to the Punjab that is so striking. There is a continuous 

 regular mountain-arc of ancient rocks on the north ; there is a parallel 

 upland of similar ancient rocks on the south, continuous beneath the presum- 

 ably shallow gap between the Kajmahal and Garo Hills ; between this 

 mountain arc and the upland is a continuous outcrop of Alluvium of great 

 maximum thickness, occupying a continuous trough, and succeeding what 

 there is no reason to doubt is a continuous Tertiary river deposit. The 

 sub-alluvial barrier deduced north-north-east of Delhi interrupted neither- 

 the continuity of the Siwalik deposits nor that of the Alluvium. There is no 

 rock-barrier in the plains at the present day between the basins of the Indus 

 and the Ganges, the watershed being a scarcely perceptible one. 



(8) The two hypotheses of single westward- flowing rivers flanking the 

 Himalayas throughout their length, one on the Indian side and one on the 

 Tibetan side, mutually support each other, and are strengthened by the 

 curious parallelism between the histories of these two hydrograpbic lines (see 

 below). 



The Siwalik Indobrahm either extended back into Assam from 

 the beginning, or very soon cut its way back into that province- 

 There is a short break in the continuity of the Siwalik exposure 

 south of Darjiling, a result possibly of excessive exposure to the 

 south-south-westerly monsoon sweeping through the Kajmahal- 

 G-aro Hills gap and causing a transgression of the Alluvium north- 

 "wards sufficient to overlap locally the Siwaliks ; the latter are, in 

 all probability, continuous beneath the Alluvium. Eastwards 

 Siwalik beds are seen fringing the plain up to the extreme point of 

 the Assam valley. A glance at a geological map will show that 

 the Shillong Plateau appears as an interrupted extension of Bihar, 

 and consists of much the same kind of rocks (PL X ). It is, in 

 fact, part of the more central tract of the old Grondwana continent 

 of which peninsular India consists, and geodetic observations 

 confirm the surmise that the intervening alluvial gap made by the 

 Ganges and Brahmaputra is of no great depth. That is to say, 

 the southern boundary of the great trough already mentioned 

 extends beneath this gap, and continues as the Shillong Plateau 

 and the Mikir Hills in Assam. It seems highly probable, therefore, 

 that the Siwalik Indobrahm extended back between Bhutan and 

 the Shillong Plateau into Upper Assam. Here it must have come 

 into conflict with another river which at that time flowed south- 



