No. 4.] CLAYPOLE PRE-GLACIAL GEOGRAPHY. 221 



The evidence here given completely fails to support the con- 

 clusion drawn. Instead of inferring from the presence of glacial 

 grooves in the bed of Lake Ontario that that lake-bed had been 

 entirely formed by the action of ice, the only logical inference is 

 that it was occupied by ice long enough to allow time for the 

 production of these marks. It would be as correct to argue from 

 the presence of the scratches made by sandpaper upon an orna- 

 mental moulding that the whole of it had been worked out by 

 the carpenter by that means. 



This reply may be carried a step farther. Admitting just for 

 the sake of argument that the beds of Lakes Erie and Ontario 

 were filled for a lono-er or shorter time towards the end of the 

 great ice age by a local glacier, we may fairly ask, " What 

 caused these local glaciers to exist there and move from east to 

 west ? A narrow gorge such as that at Niagara would not de- 

 flect a "lacier. It would be filled with o'lacial drift, and the ice 

 would then pass over it as if it did not exist. This has happened 

 in numerous instances both in east and west and in north and 

 south gorges. To deflect a glacier the depression must be wide 

 enough to allow the ice to sweep or scrape its bottom clear of 

 deposit. We need in fact a wide open valley not a deep ravine 

 for this purpose, and if such deflection of the margin of the con- 

 tinental glacier really occurred, the valleys of Erie and Ontario 

 must have been valleys of this kind, in fact nearly what they are 

 at present. It would have been more logical to assume the 

 existence of these valleys as the cause of the diversion of the 

 local glacier, otherwise the fact of this diversion remains without 

 apparent cause. The excavation, to whatever cause due, was 

 earlier than the glacier which filled it. To suppose otherwise is 

 to make the glacier produce its own cause. But further: ''Is 

 it possible to admit the existence of these local glaciers in the 

 beds of Lakes Ontario, Erie, Huron and Michigan ? " Without 

 the pre-existence of the lake beds this question may be answered 

 promptly and decidedly in the negative. That the edge of the 

 great continental icesheet was not regular we may consider cer- 

 tain. That it stretched farther southward on low ground than 

 on high ground, may be assumed, it being quite in accord with 

 the phenomena of recent glaciers. But that it was capable of 

 throwing out long narrow tongues of ice where scarcely any de- 

 pression of surface existed, is incredible. It is contrary to what 

 we know of the physics of ice to believe in the existence of an 



