Great Ice-Da/iis. — Taylor. 21 
the glacier extended southeast from Georgian bay to, or nearly 
to, the Oak ridges 20 miles north of Toronto, and it seems 
probable that between this lobe and the northern limb of the 
Albion dam, which then filled the lower part of the Trent val- 
lev as well as the eastern half of the basin of lake Ontario, 
there was a lake covering the upper part of the Trent valley 
and probably part of lake Simcoe also, and having its outlet 
southward to the Ontario basin. 
vSuch were the great ice-dams — immense ice-walls or bar- 
riers hundreds of miles long and standing for much of their 
length in water hundreds of feet deep. The foregoing descrip- 
tion of the positons of the ice-dam where it crossed the pres- 
ent water-filled lake beds is, of course, in each case largely 
conjectural. But in most instances there is a fair basis of fact 
to guide one's estimate, and the waterlaid moraines which have 
been traced across some of the basins in their dry parts furnish 
invaluable precedents and principles. It is possible that in 
some cases the course of the submerged moraine across the 
lake bottom may be traceable by a study of the soundings. In 
the pouthern part of lake Huron there is a submerged ridge 
showing sand and clay and gravel. It is 200 to 300 feet below 
lake level and trends very closely in line with the Hagenville 
moraine produced towards the southeast and may in fact repre- 
sent it. There are probably also rock ledges, however, having 
the same general trend and from which any supposed morainic 
ridges must be carefully distinguished.* 
Solidity of the lee-lobes. 
The reality of ice-dams of such great magnitude might be 
doubted on the ground that the lobes of a glacier could not 
stand in water 200, 500 or 1,000 feet deep, because they would 
break up and float away as icebergs. Mr. Leverett suggests 
this view in both of his papers referred to above.t To the 
writer, however, it seems clear that such a conclusion is not 
warranted by the facts now at hand. If it be examined crit- 
ically this assumption will be found to rest on a supposed simi- 
larity where in reality similarity does not exist. The great 
ice-streams that flow out from the main, ice-sheet of Green- 
*Sce U. S. Lake Survey chart of lake Huron. 
tPage 296 in the first: pages 14-15 in the second. 
