Geology 



that to the right there was a small ice-fall which, 

 however, only occupied half the width of the glacier, 

 terminating at the rock-train, while the other half con- 

 sisted of smooth ice. From this we concluded that to 

 the right of the train was a concealed mass of rock over 

 which the ice was breaking and from the end of which 

 the material of the rock-train was being torn. 



Near the rock-train on the side of the glacier remote 

 from the small ice-fall was a stream of water flowing in 

 a channel which it had cut in the ice. The channel 

 contained a few boulders and here and there a little 

 sand, but there could be no doubt that it was cut owing 

 to the water being slightly above the freezing-point 

 owing to absorption of solar heat, and on that account 

 melting the ice of its bed. The channel was strikingly 

 similar in form to some of those which we had observed 

 in the plateau limestones near Lyell, which, it will be 

 remembered, owed their origin to solution. Like the 

 streams of the plateau, this river soon plunged below 

 the surface and became lost to view, doubtless following 

 some tunnel in the ice, or perhaps reaching its base and 

 joining the sub-glacial river which emerged in the gorge 

 near our camp. 



At the point where the stream disappeared there was 

 a funnel-shaped opening giving access to a vertical shaft 

 whose depth it was impossible to determine with the 

 means at our disposal. 



There is this difference between the moulins, as 

 these glacial funnels are called, and the sink holes of 

 the limestone area ; in the former case the medium in 

 which they are cut is in motion, while the limestone is, 

 of course, stationary. It might be thought that on this 



272 



