1919.] History of the Drainage of Northern India. 99 
as the writer can see, it cannot be explained on any hypothesis, 
which involves a system of drainage in Siwalik times essentially 
the same as that of to-day, without a serious violation of the 
observed or inferred data. 
SUMMARY. 
The author considers that the peculiar character and 
distribution of the Pleistocene Boulder beds of the Siwalik 
cros rse. Since in Jammu and Kangra these 
boulder beds attain the enormous thickness of 5,000 feet and 
disappear quite suddenly to the north-west of this area, while 
to the south-east they gradually diminish in thickness and are 
feebly represented as far as Bhutan, it follows that such a dam 
must have been situated north-west of Jammu and that the 
river flowed along the foot of the Himalayas from south-east 
to north-west. In Eocene times when sea covered the whole 
of the Western Himalayas such a river must have risen on a 
system and breadth of the Mahanadi so disproportionate to 
what is now the Bay of Bengal from the Eocene onward, 
rising on the same watershed which is mentioned above. 
lift on a more colossal scale than any that had preceded it 
actually reversed the flow of the river in the basin of which 
the boulder conglomerates were deposited, the water flowed 
into the channels of the southerly flowing rivers which were 
ready to receive it. The V-shape of many of the Himalayan 
rivers along a certain portion of their course (the point of 
the V facing north-west), is significant as evidence that the 
northern arms of the Vs represent tributaries flowing in the 
normal direction which they would take to join a great north- 
westerly flowing river. The Gangetic alluvium, thick though 
it is, has all been deposited later than this period in the 
valley of rivers with a normal gradient as the result of annual 
floods, depression continuing simultaneously with the addition 
of flood material and sediments. 
