Vol. 65.] GLACIAL EROSION IN NORTH WALES. 305 



glaciers sometimes act, like rivers, as depositing agents. But there 

 are two serious difficulties in the application of this method to the 

 problem in hand . first, only the ends and edges of existing glaciers 

 are accessible to observation ; there the ice is thin and the pressure 

 is small ; hence the processes there observed may give no proper 

 indication of what went on at the bottom of the enormous glaciers 

 of the Glacial Period. Second, no one knows the duration of the 

 Glacial Period ; hence, even if the rate at which large glaciers can 

 erode were determined, the amount of work done by the ancient 

 glaciers would remain, under this method, indeterminate. 



(b) Deductions from the physical properties of 

 glaciers. — It has been sometimes attempted, by studying the 

 physical properties of glacial ice, the rate of motion of glaciers, and 

 so on, to determine whether glaciers can erode or not ; but no 

 generally satisfactory results have been reached in this way. More- 

 over, as in the previous method, the total work done could not be 

 determined by this method so long as the duration of the Glacial 

 Period remains unknown. 



(c) Estimates of the volume of glacial drift. — Several 

 attempts have been made to estimate the volume of glacial drift 

 laid down in moraines, and thus to determine the average depth of 

 glacial erosion of the surface whence the drift was derived. This 

 method cannot discriminate between the drift supplied by glacial 

 erosion proper and that supplied by superglacial weathering ; and 

 it fails to include the fine-textured drift washed forward from the 

 moraines by streams. In any case it is inapplicable in Wales, 

 because most of the drift from the Snowdon district is now 

 inaccessible on the sea-bottom. 



(d) Restoration of the pre-Glacial form of a glaciated 

 district. — It is sometimes possible to make a rough restoration of 

 pre-Glacial form in a district that has heen glaciated. Then the 

 difference between pre-Glacial form and present form must be 

 equal to the glacial erosion (with the associated weathering above 

 the ice-surface) plus post-Glacial changes. The latter can generally 

 be allowed for without difficulty, and thus a useful determination 

 of the value of glacial erosion in affecting topographic form can be 

 gained. It was with a view to the application of this method that 

 the preceding discussion of the pre-Glacial form of the Snowdon 

 district was written. 



(e) Comparison of non-glaciated and glaciated dis- 

 tricts. — If a comparison be made between districts, otherwise 

 similar, one of which has been glaciated while the other has not,, 

 the differences may be plausibly ascribed to glacial action. Such 

 a comparison has been instituted in the foregoing pages. It is 

 evident that this method should be associated with the preceding. 

 At the same time, it must be noted that no safe conclusion can be 

 thus reached, until it is shown that the peculiar features of glaciated 

 mountains are such as glaciers would produce if they were eroding 

 agents. Hence the next method must also be employed. 



