50 NEW YORK STATE MUSEUM 



velopment of the Mississippi drainage system, considers that the 

 flow of the Laurentian drainage system was southward into the pre- 

 decessor of the Ohio river. As has already been shown, Spencer's 

 eastward flowing river system can be originated only by a com- 

 plete readjustment of the drainage, resulting from a great relative 

 depression of the eastern uplands. Such a system could only come 

 into existence after the valleys had been formed for it, and hence, 

 as far as the history of the lake basins is considered, no such river 

 system is required, and, unless positive proof of its former existence- 

 is forthcoming, it may be dismissed as hypothetic. One of the most 

 important theories of the origin of the Ontario and other lake basins, 

 and one which has had, and still has many prominent advocates, is 

 that of glacial erosion, either entire or preceded by river erosion.. 

 This explanation was first most strongly urged by Prof. Newberry, 

 and it has found its most recent able supporter in Prof. Tarr. It is 

 impossible to do full justice to this view in the present limited space. 

 Ice erosion is a factor the potency of which has often been over- 

 looked, but of the importance of which there can be no question. 

 We may however question whether a valley which, like that of On- 

 tario, lies transverse to the general direction of ice movement in 

 this region, can owe much of its depth to this agent. The following 

 considerations will be helpful in understanding the influence of 

 glacial erosion on preexisting topography. If a valley like that of 

 Lake Ontario is occupied by a glacier the motion of which is parallel 

 to the trend of the valley, the topographic relief is likely to be 

 accentuated by ice erosion. If the motion of the ice is transverse- 

 to the direction of the valley, the erosion tends to obliterate or at 

 least reduce the relief features. If, however, a mass of ice remains 

 stagnant in the valley, the upper strata of ice may override it, and 

 the amount of glacial erosion is reduced to a minimum. The striae- 

 in this region, together with the direction of slopes from the old- 

 land, point to a southward movement of the ice, and Gilbert has 

 shown that the amount of erosion on the edge of the escarpment in 

 western New York is comparatively slight. 1 Hence we may assume 

 that the basin of Ontario was mainly occupied by ice during the 



"Bui. geol. soc. Am. 11:121. 



