KAMFS. 353 



etored to their natural course, and the drainage-line of the 

 Mohawk was resumed, these kam.es of the Finger Lake re- 

 gion would be naturally formed. 



Westward from Xew York the conditions are not favor- 

 able for the formation of this class of glacial deposits. Still, 

 there are numerous short series of kames in Ohio at the low 

 place in the water-shed between the lake and the tributaries 

 to the Ohio River. Through these the glacial torrents poured 

 over into the southern streams during all that period when 

 the outlet through the AT abash was closed up and the dam 

 across the Mohawk Valley was at its height. Kames are 

 abundant in Summit county, between Ravenna and Akron, 

 and southward, and in the neighborhood of Seville, in Me- 

 dina county. 



Occasional kame-like ridges are reported farther west, as 

 in Pipestone county, Minn.* But, accordiug to the testi- 

 mony of Mr. Upham, whose experience is wider than that 

 of any one else in these investigations, ki prolonged kames, 

 comparable with those of Sweden and Scotland, and with 

 those described in Maine, Xew Hampshire, and Massachu- 

 setts, have not been found." + 



A summary of the last three chapters may be helpful here : 

 The extreme length of preglacial as compared with post- 

 glacial time is evident from the enormous extent of pregla- 

 cial erosion. Outside the glaciated region all the rivers oc- 

 cupy deeply eroded valleys, showing v the great length of the 

 time through which eroding agencies have been at work. 

 The post-glacial gorge of Niagara is but seven miles long, 

 whereas the preglacial gorge of Ohio is both wider and 

 deeper than that, and is more than a thousand miles in 

 length. 



It is easy to see that all the northerly-flowing streams of 

 preglacial drainage would be dammed up by ice during the 



* " Geological and Natural History Survey of Minnesota," vol. i of the final 

 report, p. 545. 



f Ibid., " Ninth Annual Report," p. 290, 



