280 Canadian Record of Science. 
eastern and western sides of the ridge, but this is, presently, 
mere conjecture. The height of the marine terraces on 
Montreal Mountain and elsewhere, as compared with the 
level of Lake Ontario, the absence of the Leda clays with 
their marine shells and fish farther west than Packenham, 
and the direction of the ice grooves which have a trend to 
the west of south on the Lake Ontario side, and, generally 
speaking, to the east of north or of south, on the St. Lawrence 
and Ottawa River sides, all tend to suggest this former 
higher altitude of the Laurentian ridge at the Thousand 
Islands. In this connection it may be noted that whilst it is 
usual to refer to the direction of the ice grooves as being 
either to the east or west of south, it is quite in consonance 
with the direction of the St. Lawrence Valley that these 
grooves should sometimes be referred to as having a course 
to the east of north. 
With the elevation of the Niagara escarpment came the 
first record we have in the history of Lakes Ontario, Huron 
and Michigan as independent basins with the contours of 
to-day. Previous to and after this elevation, the present 
basins of these lakes were the seat of a great river system, 
with probably lake expansions smaller and different in out- 
line from those now existing. Profs. Spencer and Claypole 
suggest that Lakes Ontario and Erie in part formed the 
valleys of a great pre-glacial river which, Spencer thinks, 
crossed from Lake Huron through the counties of Lambton, 
Middlesex and Elgin, and swerving around Long Point to 
the deepest portion of Lake Erie, trended thence northward 
to the Dundas Valley. Through this valley it entered the 
present basin of Lake Ontario, the line of deepest depression 
in which it formed by cutting down into the Hudson River 
shales, along the escarpment of which it flowed. There is 
much in the features of the lake floors and of the superficial 
deposits to support some such view, if more recent local 
warpings in the strata are considered. The great fracture 
in the strata at Dundas would give the required direction to 
the river there, and would be greatly enlarged by its eroding 
action. The outlet of this river by way of the Mohawk 
