part 3] OF THE IN I) IS, BR All MAPI TEA, AND GANGES. 1 1"> 



of the Punjab and the North- West Frontier, where the first stage 

 was its transference from the Murree zone southwards to the Soan 

 valley or Siwalik zone. In post-Tertiary times, as we shall see, 

 this line was pushed still farther towards the peninsula. The 

 relationship of the Zhob- Valley Tertiaries in Baluchistan awaits 

 further investigation. 



The Siwalik Indobrahm, therefore, rose probably in Assam, 

 flowed along the southern border of Bhutan, traversed the northern 

 angle of Bengal below Darjiling, skirted the southern margins of 

 Nepal and Kumaon, proceeded north-westwards through Dehra 

 Dun and the Sirmur State, up through the Kangra district into 

 Jannnu, thence along the Soan valley south of Rawalpindi, past 

 Makhad, across Kohat to the Bannu plain, where it turned south 

 past the Shirani Hills and Bugti Hills, debouching ultimately into 

 the Sind estuary, which silted up in later Siwalik times. On its 

 right bank it received young vigorous tributaries born in the 

 Tertiary Era ; on its left bank it received as tributaries the relics of 

 very ancient rivers which drained Gondwanaland, such as the Son, 

 the Chambal and its branches, and others. The Punjab Soan, as 

 indicated above, is to be regarded as a relic of the old Tertiary 

 Indobrahm, and is in conspicuously puny proportion to the great 

 thickness and extent of the Tertiary fluviatile sediments through 

 which it Hows. The reasons for believing in the presence of this 

 Siwalik river may now be enumerated : — 



(1) A marine gulf originally occupied most of the same line, and during the 

 silting process the oil and coal of this region were produced and gj r psum was 

 deposited ; this hydrocarbon, coal, and gypsum belt follows the Nummulitic 

 •outcrop from Kumaon to the Mekran coast. Such a filled-up gulf is naturally 

 followed by a river — for instance, the Burma gulf by the Irrawadi, or the 

 Mesopotamian gulf by the Tigris and Euphrates, — which would be pushed 

 further towards the centre of the peninsula by the persistent movement. 



(2) The supposition that the Siwalik deposits were laid down as enormous 

 talus-fans by the mountain-streams which issued into the plain, has always 

 been difficult to accept in view of the great thickness of the deposits — for they 

 .average over 16,000 feet — and their great extent. An almost unbroken belt 

 •of these beds can be traced, usually in great thickness, from Assam to the 

 Soan valley and thence to Baluchistan, bounded externally by continuous 

 mountain country consisting, where known, of igneous and very ancient rocks, 

 with or without an intervening belt of Murree rocks. It seems far more 

 reasonable to deduce a single river rather than a number of transverse streams. 

 The mountain-streams, in any case, must have had an outlet to the sea in 

 •one direction or the other, and must have joined a main river in the plains. 



(3) There are no grounds for assuming that the northward- flowing drainage 

 •of the old Gondwana continent was reversed in Siwalik times. In fact, the 

 depression of the trough would have invigorated this drainage, which, it is 

 highly probable, persists to this day. It seems almost certain, therefore, that 

 this northerly drainage must have met the southerly Himalayan drainage in a 

 large river flowing either south-eastwards or north-westwards. Since the 

 Siwalik outcrop is not continuous to the sea south-eastwards, but is prac- 

 tically continuous north-westwards and subsequently southwards along the 

 line of an old gulf, it seems reasonable to assirme that the river followed this 

 •direction. 



