588 THE ICE AGE IN NORTH AMERICA. 



finer part of the glacial debris carried southward by the gla- 

 cial streams — so that, upon any theory, we should expect a 

 much larger accumulation of loess over the southern portion 

 of the area. 



Mr. Leverett relies largely on the great erosion of the Kansas 

 sheet till as an indication of its age. He estimates, for example, 

 that in northern Missouri not over thirty per cent of the origi- 

 nal plain is left upon the retreat of the ice, in the narrow tabu- 

 lar remnant remaining upon the divides. "The streams are 

 flowing in valleys that have broad slopes and bottoms, the 

 slopes being so toned down as to fall generally below 5° and 

 not uncommonly to 3° or even less. The slopes of the valley, 

 twenty-five meters in depth, often have a breadth of about a 

 kilometer, and the bottoms of small drainage lines often 

 exceed a kilometer in width. Topographic sheets of the 

 United States Survey, which well illustrate the post-Kansan 

 erosion, are the Atlanta, Edina, and Kahoka quadrangles of 

 northern Missouri.' ' (See comparison of North American 

 Glacial., Deposits, from "Zeitschrift fur Gletscherkunde," vol. 

 iv, p. ; 258). 



With this he compares a part of the Belleville, 111., topo- 

 graphic sheet which shows 60 per cent or more of the original 

 glacial plain untouched by erosion; but again in the Iowan 

 drift (which Mr. Leverett would now identify with the Illi- 

 noisan epoch) , the portion of the glacial plain which is undis- 

 sected is not greatly in excess of the Kansan plain from 

 Missouri. 



It should be noted furthermore that the blanket of Kansan 

 till is comparatively uniform over its whole area. There are 

 no moraines in it, and there never were any. Moreover, the 

 deposit was rarely thick enough to disguise the preglacial 

 topography. Much of the supposed evidence of post-glacial 

 erosion, is probably the result of this failure. of the glacial 

 deposits to fill the valleys and channels of the original topog- 

 raphy. ! 



