GEOLOGICAL AND NATURAL HISTORY SURVEY. 21 



that this chano'e of front could not liave taken place in a brief time. 

 In connection with this c'hange of front there appears a re- 

 markable increase in the number of surface bowlders, a feature 

 whic'h may prove to have great significance. The causes for vari- 

 ations in the number of surface bowlders are so little understood 

 as yet that we scarcely are warranted even in conjecturing the im- 

 port of such bowlder belts. They may signify a great recession 

 of the ice sheet, followed by a fresh advance across the Canadian 

 highlands, by which a new supply of bowlders was gathered up 

 and brought down to the extreme limit reached by that advance. 

 Such bowlders in some cases, as recently suggested by Mr. F. B. 

 Taylor (i) may be derived from accumulations in interglacial val- 

 leys. This suggestion, however, can scarcely apply to bowlder 

 belts of such length as are displayed in Western Indiana and East- 

 ern Illinois. 



LATE WISCONSIN DKIPT SHEETS. 



The moraines of the Late Wisconsin series contain generally 

 a larger amount of assorted material with the till than is found in 

 the moraines and accompanying till sheets of the Early Wisconsin 

 series. The topography also is more closely allied to the knob 

 and basin type than to the gently undulating broad ridged type 

 of the Early Wisconsin series. 



Some attention has been given the subject of comparative 

 amounts of surface leaching and erosion on the Marseilles mo- 

 raine and the moraines of the Late Wisconsin series, but no strik- 

 ing contrast has been observed. A more refined investigation, 

 however, may bring to light essential dififerences. 



Great Bowlder Belts and Accompanying Moraines. — The 

 bowlder belts of Northeastern Illinois and Western Indiana have 

 attracted attention from the earliest days of exploration, as have 

 also their continuation in Central and Eastern Indiana and West- 

 ern Ohio. In the early explorations they were observed only as 

 detached belts, hut their connections are now well . established. 

 Occurring, as the bowlders usually do, in narrow, well-defined 

 belts, they are readily traced from section to section, even where 

 unaccompanied by peculiar topographic features. Throughout 

 much of their course in Oihio and Indiana these bowlder belts are 

 simply an accompaniment of well-delined belts of morainic drift. 

 But in Western Indiana and Northeastern Illinois the bowldcr 



[1] In a paper presented at the Washington meeting of the Geological Society 

 of America, December, 1896. 



