. Evidence of non-Glaciation. 157 



much as it appears at present, only somewhat larger, its waters 

 being held by the dam of silt which had been laid down in front 

 of the ice. 



Having in mind the conclusions of Dawson, McConnell and 

 Russell as to the northern limit of glaciation in the Yukon basin, 

 evidence on that point was carefully sought in the plateau region 

 southwest of Selkirk. For the first one hundred and twenty- 

 five miles the evidence was wholly negative. No sign of glaci- 

 ation was seen, and this too in a country well calculated to retain 

 the marks of ice action. The stream gravels consist of a very 

 small number of rock species, and on following a stream to its 

 head the source of each was usually found, showing that no 

 foreign material had been brought into their basine. While in 

 general the surface contours are smooth and flowing, this is the 

 result of long-continued subaerial rock disintegration, and gen- 

 erally the surface rock is deeply buried beneath great accumu- 

 lations of fragmental debris, though occasional sharp pinnacles 

 and towers of rock project from the smooth talus slopes. Had 

 this region been subjected to the action of an ice sheet during 

 the glacial epoch, not only would the greater part of the rock 

 debris have been removed but the projecting pinnacles would 

 have been planed down to rounded knobs which would still- 

 retain polished and striated surfaces. 



Where Nisling river was crossed its broad valley is filled with 

 a deposit of coarse gravel and bowlders, and from their great 

 quantity and variety it was inferred that the stream had its 

 source in a drift-covered region. The first undoubted evidence 

 of ice, however, was found on the divide between Nisling and 

 Kluantu rivers, where the northern edge of a sheet of bowlder 

 clay was passed. From this point southward the character of 

 the surface suffers a marked change. It is no longer composed 

 of the fragments of one or two kinds of rock occurring in place 

 near at hand, but rather of many varieties confusedly mingled 

 with clay and sand. The drainage system is imperfectly adjusted 

 to the toi:)Ographic surface, so that wide valleys carry small 

 streams, and large streams like the Kluantu and Donjek flow, for 

 considerable distances at least, through narrow valleys. 



The ice which has left its records in this sheet of bowlder clay 

 was probably a confluent glacier formed by streams coming from 

 the south through narrow valleys now occupied by Kluantu 

 and Donjek rivers. These valleys do not appear to have been 



22— Nat. Geog. Mag., vol. IV, 1892. 



