SUMMARY AND DISCUSSION. 65 



places result in their occasional discovery even farther to the west than 

 they have yet been observed. 



Another fact of importance, and one which impressed itself on the 

 writer in the course of the recent examination, is the following : Except 

 in the case of the moraines evidently referable to glaciers of the Rocky 

 mountains, which we have found reason to assign to a very early period 

 and which save in the case of Bow valley are closely confined to the base 

 of the mountains, the more obvious evidences of the work of glaciers are 

 conspicuously absent in this entire region of the foothills and Porcupine 

 hills. The highest and farthest limits of the drift are not marked by 

 moraines, and moraines, drumlins, kames, and eskers are, with the above 

 exceptions, entirely wanting. This is very striking when comparison is 

 made between this region and that of British Columbia or the Laurentian 

 plateau, both of which are known to have been overridden by vast 

 glaciers. 



Within the past year Professor T. C. Chamberlin has formulated and 

 named a series of stages in the glacial history of North America, and 

 although the author of the classification would probably be the first to 

 admit its provisional character, it has undoubtedly already been of con- 

 siderable service in suggesting a basis of arrangement and in fixing the 

 direction of future work. Thus it will be appropriate briefly to note 

 here in conclusion what appear to the writer to be the probable relations 

 of the glacial deposits of Alberta to this general classification. 



The " lower '' boulder-clay may, it is believed, be regarded as repre- 

 senting the Kansan formation, while the interglacial deposits, best de- 

 veloped along the Belly river, are supposed to be contemporaneous with 

 the post-Kansan interval. The " upper " boulder-clay of the western 

 plains may then be identified with the lowan formation and like it is 

 associated with abundant silty beds. The Wisconsin formation is in all 

 probability not met with in the extreme west, but its limit in this direc- 

 tion maybe marked by the Missouri Coteau, which in Canadian territory 

 extends from the forty-ninth parallel to the North Saskatchewan and in- 

 definitely beyond in the farther north. The post-Iowan interval, in this 

 case, appears here, as in the region farther east, to be marked by the 

 erosion of important interglacial valleys, which find their limit at the 

 Coteau and its systems of drift ridges and hills.^ No deposits like the 

 Coteau occur in connection with the western terminations of the " lower " 

 or " upper " boulder-clays. 



Reverting now, on the basis of the above correlation, to the Saskatche- 

 wan gravels and the " western " boulder-clay, it will be apparent that 

 these must represent an antecedent and unnamed stage of glaciation in 



* Geology and Resources of the Forty-ninth Parallel, p. 230. 



