part 3] OF THE [NDUS, BRAHMAPUTRA, AM) GANGES. 1 1/ 



westwards between the Shillong Plateau and Mikir Hills on the 

 one side and the Naga Hills on the other, over what is now the 

 hill-section of the Assam-Bengal Railway, into the Bay of Bengal. 

 The present Meghna may he the remains of this river, which seems 

 to have been beheaded by the Indohrahm. Prom the size of the 

 Siwalik deposits in this railway hill-section, this old Meghna 

 appears to have been a river of some magnitude, and may possibly 

 have derived its importance from the capture of the Tsangpo, now 

 the Tibetan part of the Brahmaputra. The Tsangpo was then 

 much shorter than it is now. and did not reach far into Tibet : it is 

 thought to have formed at that time the head- waters of the 

 Errawadi, connected therewith by what is now a tributary of 

 the latter, the Chindwin. 1 The upper waters of the Tsangpo- 

 Meghna were then apparently captured in turn by the Indohrahm. 

 The Kapili tributary of the Brahmaputra seems to have cut back 

 at sonic time between the Mikir Hills and the Shillong Plateau 

 and captured a part of the Meghna, but whether this capture was 

 effected before that made by the Indobrahm is not clear. Whether 

 the Chindwin- Irrawadi or the Meghna were intermediaries or not. 

 the Indobrahm — or Brahmaputra, as it may then have become— - 

 eventually captured the head- waters of a Tibetan river, the history 

 of which is- a curious one. 



Sir S. (J. Burrard & 'Dr. Hayden have pointed out the strange 

 backward direction of many of the present Tsangpo tributaries, 

 which is explained by the assumption that they originally fed a 

 westward and not an eastward (lowing river: — 



'The most recent maps show that shortly before their junctions with the 

 Brahmaputra, these tributaries are beginning to bend in their courses, and to 

 turn towards the present direction of the Brahmaputra's flow." a 



The most important tributaries having a backward directed 

 course are the Kyi or Lhasa, the Nyang, the Rang, and the Shang. 

 but there are many smaller ones that exhibit the same peculiarity. 

 The eastern end of the Himalayan chain is supposed to have begun 

 to rise earlier than the western. This would initiate in early 

 Tertiarv times not only the Indobrahm line, but also another line 

 of drainage in the same direction on the Tibetan side. Sir S. (J. 

 Burrard it Dr. Hayden suggest three hypotheses worthy of con- 

 sideration, regarding the former outlet of this old westward-flowing 

 Tibetan river : — 



(a) It may have flowed over the Pliotu P;iss and through the defile ol 



the Kali Gandak on to the Gauges plain; 

 (6) It may have passed through the basin of the Karnali with a similar 



destination ; or 

 (r) It may have followed the present Himalayan course of either the 



Sutlej or the Indus. 



1 Until it was shown to be otherwise, the old Chinese surveyors believed 



that the Tsangpo of Tibet still flowed into the Irrawadi (S. G. Burrard & 

 H. H. Hayden. 'A Sketch of the Geography & Geology of the Himalaya 

 Mountains & Tibet' 1007 100S, p. 157). 

 - Op. ji'ni tit, pp. 1 ">.') .")»;. 



