432 J. D. Dana — SovtInr,r r </ I)i.s/-/t<i/'</e of Lake Winnipeg. 



gravel of the nature of "karnes," but, on the whole, remarkably 

 uniform; on the 49th parallel it rises gently eastward toward 

 the Lake of the Woods, 90 feet in the 77 miles. On the upper 

 part of the Minnesota the deposits are largely the Glacial drift 

 (General Warren, and Professor Winchell), with also portions 

 that are bedded. 



Conclusion. — Taking the { 



s of a barrier of ice or of any other kind. For 

 had been no great change of slope over the region, the 

 ;s of the great lake should be approximately horizontal, 

 3 outlet at Lake Traverse, and if horizontal, they would 



a height in the vicinity of the pn 

 ibove the 1 * 



feet above the lake, supposing the waters just up to the Lake 

 Traverse level, and 300 feet if the water at this place of dis- 

 charge was merely 40 feet deep. Instead of this condition, the 



,_>],:<,;, -rr,i sjmrr line liOs ntilrhj >':■ •? of the SU)'- 



fare ; and, further, tin- .slo/,r <f the hike-bottom prairie is not much 

 different from that <>i the bordering phiteiw on either side. 

 ' We have thence the conclusion, since the present outline of 



approximated the shoreline and once horizontal, that there 

 has been a -rear change in the level of the land, as General 

 Warren urge. 1. The idea of a change in the position of the 

 earth's center of gravity sufficient to change the slope of the 



We may infer also, from the near correspondence between 



the northward slope of the lake-bottom prairie and that of the 

 bordering plateaus, that the prairie and plateaus were affected 

 alike by the conditions as to level. And we may deduce from 



much greater undetermined 



profoundly on questions as to the 

 erior, and "the origin of changes of 

 ice, that it is greatly to be desired 



