3$ 
KENDALL : THE GLACIER LAKE8 OF CLICVELAM). 
materials, mingled ^^ith lateral moraine, would be washed into 
the space between the margin of the ice-lolje and the hills 
to form a species of deformed delta. . The stream would thus 
turn towards the only point of escape, Lake Pickering, and 
would spread its gravels to form a bench along tlie ice falling 
from east to w^est. Each valley descending tlie ice-free slopes 
west of Forge Valle}^ would behave in the same way as the main 
stream and turn westwards along the gravel bench. As the 
lobe receded, successive avenues of escape for the streams would 
be afforded along the ice-front, and thus every step of the retreat 
would leave a segment of the main channel liigh and dry upon 
the top of the gravel terrace. 
As the ice slowty retreated, its lateral dwindling would 
allow the river to flow between the elevated terrace and the ice, 
w^hich would result in some scarping of the gravel bank, and a 
partial redistribution of its materials, until the retreat left the 
mouth of Forge Valley open and the waters of the Derw^ent 
debouched directly into Lake Pickering. 
After the glacier-lakes were formed by the invasion of the 
ice-sheets, Lake Pickering would receive practically all the 
drainage of the Cleveland area, from the Ingleby (rreenhow 
Lake by Bilsdale, from all the Xorth Cleveland Lakes and Lake 
Eskdale by Newton Dale, and from the East Coast Lakes by 
Forge Valley. Lake Pickering in its turn discharged its waters 
b}^ Kirkham overflow into Lake Humber, a great glacier-lake held 
up in the great central valley of Yorkshire l^y an ice-dam at 
each of its exits. What were the southern limits of tliis lake 
cannot yet be certainly defined. That the mouth of the Humber 
and the Wash were obstructed by the Xorth Sea ice seems 
pretty certain, and the distribution of the chalky boulder clay 
indicates a great barrier of ice far w^estwards along the Trent 
valley. This proposition is supported by the existence of fine 
laminated muds of identical character wdth those of Eskdale 
in the Vale of York, and far southwards in the Midlands, but no 
signs of deltas or beaches have been noted. The line of the 
overflow of this great lake southwards has not yet been iden- 
tified, but when the glacial evidence of South Yorkshire, Derbj^- 
shire, and the mid -English plain is carefulh' worked out, we may 
