1 18 
PROCEEDINGS COTTESWOLD CLUB 
1919 
likely to have been derived from North Wales, the Lake District, 
or the South of Scotland. 
The age of the deposit is probably later than Mid Glacial, 
as is indicated by the occurrence of bones of Mammoth and 
Rhinoceros at the base (188, p. 31). 
The conditions under which this deposit was formed are 
difficult to define with precision, but the general appearance 
of the beds of fine sand indicates ordinary stream action, 
while the seams of coarse and fine gravel are suggestive of 
torrential streams from the melting ice and snow. Some of 
the irregularly shaped masses of sand appear to have been 
enclosed in a frozen condition. 
Among the Drift constituents of other Severn Valley gravels 
there are erratics from Galloway, the Lake District, and the 
northern counties of England and other districts over which the 
ice-sheets moved in their advance towards the Midlands, to- 
gether with waterwom fragments of marine shells from the 
Irish Sea ice, and flints and Quartzose pebbles from various 
sources. Most of these materials were transported by stream 
action from the melting ice fronts or perhaps from the overflow 
of a large ice -lake in the Midlands ( 84 , p. 20 ; 83 , p. 94). The 
base of the gravels often contains large angular fragments 
of local rocks released by the thawing of the frozen bed of the 
river. 
Theories of marine or lacustrine submergence and differential 
earth movements have been advanced in the attempt to account 
for the various elevations at which these and other deposits 
occur. Neither of the explanations is in all respects satis- 
factory, and it is doubtful whether they would have been put 
forward except to explain the presence of the Drift at the few r 
exceptional elevations. The occurrence of waterwom Glacial 
shells in the Severn Valley is regarded as the main support of 
the marine submergence hypothesis, but their origin has now 
been traced to shelly deposits ploughed up by the ice-sheets 
from the bed of the Irish Sea and spread over a part of the 
Severn Basin ( 104 , p. 304). As similar shell fragments are 
found only in the path of the Irish Sea ice or streams flowing 
from it, and as no complete shells or other signs of marine 
