18 THE ROYAL SOCIETY OF CANADA 
westward to Lake Winnipeg and Lake of the Woods, but when its 
front had retired to within about a hundred miles of the present 
southern shore of Hudson Bay, it was approximately between 400 
and 500 feet below its present level, for about that time the salt water 
of the Bay or of the ocean flowed round in front of it, and in this salt 
water stratified marine sands and clays began to be deposited. It 
is quite possible, however, that before the ice had retired to this extent, 
and when it covered a very much larger area, the land under and around 
it may have been depressed much more than this, in which case 
Lake Agassiz in its later stages may have been at sea level, and may 
have been connected with ocean water in the basin of Hudson Bay 
along the western face of the Labradorean glacier. 
As the front of the glacier continued to retire the land continued 
to rise, though what was the exact relationship between the rate of 
retirement of the glacier and the rate of rise of the land is not known. 
The floor of till on which the glacier rested before its retirement 
was remarkably level throughout most of the country near the Bay, 
but there was a decided valley or depression along the line of the pres- 
ent valley of Hayes river. I do fot know how this old valley was 
formed. Possibly the surface of the till conforms in a general way 
to the contour of the underlying rock floor, or it may have been 
formed in Interglacial times, and the lignite-bearing beds of sand 
between the upper and lower tills may be in some way connected 
with it. Certainly it was not formed by erosion in Postglacial times. 
As the strip of sea between the glacier and the land continued to 
widen, sediment was being discharged by the glacier and streams flowing 
from it, and by streams flowing from the land to the south from 
which the ice had recently retired, and this sediment was spread 
more or less evenly over the sea floor, filling its inequalities and 
reducing it to a level plain which now has a gentle slope northward 
of 5 feet to the mile. At that time mollusca of various northern 
species inhabited this sea, and many of the beds of sand and clay 
contain their shells in abundance. 
These beds are, as far as I am aware, most fully developed on 
the banks of Hayes river, where they were deposited in the old valley 
or narrow bay above referred to. I would, therefore, propose for them 
the name Hayes River Series, merely as a local designation to facilitate 
description, and to distinguish them from similar beds elsewhere. 
They were probably formed at the same time as the Champlain 
Clays and Sands of the St. Lawrence valley. 
Near the mouth of Hayes river, where the scarped bank rises 
20 feet above the top of the beach, the upper 10 feet is composed of 
stratified sand containing a few marine shells, and with a layer of 
