36 THE ROYAL SOCIETY OF CANADA 



collected there was one which had a flat striated base, while the upper 

 part was distinctly a "dreikanter" with polished triangular faces 

 caused by wind-blown sand. 



It was very noticeable, however, that till was confined to the 

 lower levels. Several hills which were climbed, one reaching 360 feet, 

 showed no suggestion of ice action. The higher ones are "Demoiselle" 

 hills, as described by Clarke, and consist of an agglomerate of some 

 greatly weathered basic eruptive, probably basalt, only loosely 

 coherent and yet with steep slopes, even reaching 30° or 40°. The soil 

 upon them passes down into broken and weathered rock and is clearly 

 residual. It seems improbable that these small hills of loose frag- 

 ments, with fairly steep slopes, could have been crossed by an ice 

 sheet without any visible effects; and the statements of the earlier 

 observers seem quite justified when the higher parts of the island alone 

 are considered. If an ice sheet ever crossed the island it must have 

 touched it very lightly, or else have done its work so long ago that 

 weathering has had time to remove completely the evidences of its 

 work. 



There is, however, another Pleistocene feature to be considered. 

 The older observers all mention foreign boulders transported by ice 

 when the sea stood at aTiigher level, and these may be found, in rather 

 small numbers, along the present shore and at various points up to 

 160 or 170 feet above sea on the hill sides. On the south flank of the 

 Demoiselle hill near the harbour, a number of small boulders were 

 found, including undoubted foreign stones, such as quartzite, granite 

 and porphyritic granite. The granites at the higher levels are very 

 few in number. As these blocks rest upon a typical residual soil 

 they must have been transported by floe ice during a higher stage of 

 the sea. This higher sea level seems to have lasted only a brief time, 

 however, since no well-marked beach was formed in spite of the 

 abundant loose materials at hand for the formation of gravel. When 

 one observes the lofty cliffs and miles of sandbars at the present sea 

 level the lack of distinct beaches at these higher levels is very striking. 



There still remains the problem of accounting for till at low 

 levels with no indication of ice action on hills of loose volcanic materials 

 rising only a few hundred feet higher. It is hard to imagine these hills 

 of only 250 to 350 feet rising as nunataks above a continental ice sheet 

 depositing boulder clay up to 105 feet; and the only alternative seems 

 to be floating ice as the cause of the low lying till. Goldthwait's 

 objection that the striae on the stones in the till are generally in the 

 direction of the longest axis is not necessarily fatal to this theory, 

 since icebergs are merely floating portions of glaciers which originate 



