278 



G. M. DAWSON OIST THE SUPERFICIAL GEOLOGY OF 



Dundas Island and Cape Pox at the southern extremity of Alaska? 

 shows heavy grooving from IN". 50° E. to S. 50° W., proving that 

 this strait must have heen filled with ice. 



Sketch Map of part of British Columbia, showing the supposed exten- 

 sion and general direction of flow of the glacier-ice when near its 

 maximum limit. 



■ 32° _I30^ 125° 



124° " ~ 122° 



The arrows indicate the direction of flow of the ice. 



The dotted line shows the seaward margin of the confluent glacier. 



The scarcity of examples of w r ell-marked terraces on the coast, 

 and the comparatively small elevations at which they are found, has 

 been remarked previously. At Port Simpson, however, in lat. 54° 34', 

 the surface bears a considerable thickness of detrital matter, and from 

 a distance this appears to form an ill- defined terrace at a height of 

 somewhat over 100 feet. A few . miles further southward, at Metla- 

 katla, there is a well-marked terrace, flat, with an elevation, baro- 

 metrically determined, of 95 feet abo^e high-water mark. 



In the previous paper, already several times referred to, evidence 

 ■was brought forward in favour of a belief that during a part of the 

 glacial period a vast glacier filled the entire Strait of Georgia, which 

 separates the south-eastern part of Vancouver Island from the main- 

 land, and that the ice swept across the south-eastern extremity of the 



