

GLACIAL PHENOMENA AND SUPERFICIAL DEPOSITS. 207 
the rock coincides, is S. 23° W.; superimposed on which, at one place, are 
scratches S. 55° W., or N. 55° E. Near this a direction of S. 60° W., 
or N. 60° E. is seen, on which is superimposed striation 8. 25° W., a 
direction closely agreeing with the general one, and probably indicating 
_a brief resumption of the original force after a short interval. Another 
surface showed scratches 8. 20° E., or N. 20° W. A part of this marking 
‘may be accounted for by the packing of ice on the shores of this open 
reach of the lake, but much of it is not of this nature, and appears to 
indicate the combined action of glacier and floating ice in the lower part 
_of the region, toward the close of the glacial epoch. 
471. On the eastern side of Flag Island, a small point composed of 
hard hornblendic rock, was observed to have its nearly perpendicular 
eastern front planed off perfectly smooth, and beautifully striated and 
polished, while its summit was only gently rounded, and showed com- 
paratively little evidence of glaciation, and its southern front remained 
rough; the whole evidencing a strong movement of propulsion from 
the north, and intense local and lateral pressure. (Plate X., Fig. 3.) 
Not far south of this, on a surface of granitoid gneiss, the ice action was 
seen to have shaped and rounded previously existing asperities of the 
rock without obliterating them. One instance was especially remark- 
able, a little hollow in the rock, obliquely transverse to the general 
direction of the force, had originally a sloping northern and perpen- 
dicular south-western side. The ice had not touched the weathered 
surface of the former, while it had bevelled off and polished the latter in 
the most perfect manner. Such phenomena as this, appear to show 
great preponderance of onward movement over downward pressure 
in the action of the ice. (Plate X., Fig. 1.) 
472. The passages by which the lake finds its way over the junction 
of the Laurentian and green slate series to the Winnipeg River, do not 
depend on any evident conformation of the rocks. They cut across the 
hard ridge which marks the position of the fault bringing these forma- 
tions together, nearly at right angles, and in a direction a few degrees 
west of north. They are probably due to some fault or crack structure 
traversing the rocks with that bearing, a course which agrees almost 
exactly with that of the fracture containing the large intrusive diorite of 
the south end of Lacrosse Island (§ 84), and may tend to prove some con- 
nection between it and the similar rock which appears near the line of 
junction at Rat Portage. The gorge of the fall is certainly pre-glacial in 
date, and has probably arisen from subaerial weathering along some such 



