2SS The American Geologist. November, 1898 
basin of lake Erie, the comparatively small Saginaw valley 
lobe in the center, and the western lobe filling the basin of 
lake Michigan. When the upper Grand river was making 
its way along the western front of the Saginaw lobe, the 
edge of the Lake Michigan lobe was not very many miles away. 
The glacial torrent crossed the intervening space and reached 
its margin. Of this we are quite certain, but concerning its 
further course we can only conjecture. Perhaps it passed di- 
rectly beneath the great ice mass which blocked its way, to 
issue again in augmented proportions at its far southern ex- 
tremity. This conclusion seems as reasonable as any. Other- 
wise it must have found an outlet over the higher lands to 
the south along the glacial margin. But a rise sufficient for 
this would have backed the water clear across the area of ex- 
posed land surface, flooding the country on both sides of 
the river channel. There seems to be no evidence of such 
an inundation; on the other hand, the condition of the valley 
as above described indicates the presence of a wide flowing- 
stream. The most reasonable conclusion, then, appears to be 
that the original river actually flowed from beneath one gla- 
cial lobe and disappeared beneath the other, — a circumstance 
perhaps not parallelled in any other part of the country. 
It may be that at a still earlier period of its history this 
capacious and fruitful valley was occupied by an entirely 
subglacial stream, which was first uncovered along the middle 
of its course by the melting away of the ice which pre- 
viously had joined the Lake Erie and Lake Michigan glacial 
Lobes into one continuous ice-field. 
It would be interesting to follow the history of this river 
system back to still remoter times, involving a study of the 
preglacial topography of the Carboniferous area of Michigan, 
but we are not at present prepared to undertake the task. 
[Note. — Further study in the correlation of the many retreatal mo- 
raines of the glacial lobes mentioned in this paper may lead to the con- 
clusion that, when the stratified drift of the lower part of the Grand 
river valley was being deposited, the Lake Michigan lobe had shrunk 
so far as to wholly uncover that part of the valley and to give to the 
glacial lake Michigan an outlet past Chicago to the Des Plaines, Illinois, 
and Mississippi rivers. The lake then had a level at Chicago only some 
20 or 25 feet above its present hight, the lowest place of the watershed 
being about 15 feet above the lake. — Eds.1 
