576 
later, as the climate became less severe, the ice was confined 
to, and kept working down, in the small subordinate valleys 
of Crummack Beck and the Kibble, where, as in Chapel-le- 
dale, the moutonneed surface of the rocks in the valley, the 
rock barrier at the end, so well insisted upon by Professor 
Ramsay as evidence of glacial action, and the half- obsolete 
moraines, all point to the continuance or repetition of similar 
action on a smaller scale. 
One question is raised by an examination of the drift to 
which I think the above explanation of the glacial phenomena 
of our district will enable us to give at any rate a plausible 
answer. Why are there so many scratched stones in the 
older clay drift, when we find so few in the moraines of the 
modern glacier. Now to examine this question fairly, let us 
consider why there are few in the modern moraine. The 
stones are scratched by being crushed along against the 
rocky bed and one another in or under the glacier. There- 
fore, as a large proportion of the stones are carried down on 
the glacier, and never get in or under it, they never get 
scratched at all ; and, in the next place, as there are always 
streams in, and issuing from, the end of the glaciers, which 
roll the stones and obliterate the scratches, a large proportion 
once scratched get their scratches worn away. Now, under 
what conditions should we get least of these two causes which 
reduce the proportion of scratched stones? If we have an 
ice sheet, and no land above the glacier, there can be no 
debris to fall on to it ; or, if the land is so far that all the 
debris has fallen into the crevasses, all the detrital matter 
being in or under must be liable to be scratched. Also, in 
such a case, the water which flowed from the end of the ice 
sheet, or of glaciers so large as to be almost continuous across 
the country, would collect and flow in the hollows and valleys 
and the moraine matter left at the end of the small hills, or 
on their flanks, would escape its action. The nearer the 
