202 Wisconsin Academy of Sciences , Arts and Letters. 
have plowed their way nearly at right angles to the range, and conse¬ 
quently directly in the face of the great Laurentian ice sheet for many 
miles out on to the plains. The trough-shaped depressions made by 
these alpine glaciers are now sometimes occupied by small lakes walled 
in on all sides except towards the mountains by the old moraine. Some¬ 
times the water has cut through the morainic walls and completely 
drained the lake. In such cases it usually happens that the swaying of 
the stream from side to side has cut away the whole front portion of 
the moraine, leaving two long parallel ridges on either side extend¬ 
ing back along the course of the stream to the mountains. In still 
other cases the troughs once occupied by the ice are dry and empty. 
Measured on the plains these ice tongues varied from live to forty miles 
in length and from half a mile to five miles or more in width. Of their 
mountain extensions I shall speak later. Their thickness can in some 
cases be inferred from the appearance of the moraines. These are much 
higher on the ice side than on the outer or plains side, as would be ex¬ 
pected, since the ice came down upon the soft Laramie beds at a steep 
angle. The glacier which plowed out the beds of St. Mary’s Lakes was 
the largest of all in this region and must have been of no mean propor¬ 
tions. These lakes, two in number, lie at nearly the same level, the one 
just outside the narrow belt of foot-hills, the other extending through 
the foot-hills reaches the very base of the mountains. The lower lake 
is about five miles long and two wide; the upper is seven or eight miles 
long, but not more than two miles wide. The stream discharging them 
is 150 feet wide, three feet deep, and very swift. From our camp at the 
foot of the lower lake the stream seemed to have cut through a moranic 
wall about 300 feet high a few hundred yards below the foot of the lake. 
I found, however, on climbing up this slope that the ice had ridden over 
the seeming wall, which was not a moraine at all, but a terrace marking 
the depth of post-glacial erosion of the stream. The topography of this 
upper level was unmistakably that of a ground moraine, yet there was 
something unfamiliar about it which I was at first unable to interpret. 
There are three glacial lakes on this terrace, one a mile in diameter, and 
the whole region showed unmistakable evidence of having been covered 
with water since the retreat of the ice. On the sides jf the Milk River 
ridge, which forms the eastern walls of St. Mary’s valley, terraces, a 
dozen or more, extend up to a height of at least 800 feet above the pres¬ 
ent level of the lakes. These, with the peculiar outline given to the 
knobs and hills, constitute the unfamiliar features spoken of. I fol¬ 
lowed down the valley five miles or more, but saw no signs of a terminal 
moraine or barrier ridge of any sort that could have held the waters at 
such a level as is indicated by the terraces. I concluded, therefore, to 
adopt tentatively the hypothesis that the Laurentian ice sheet was 
the barrier; that it had sent a tongue up the St. Mary’s valley far 
