54 Mr Thomson on the Parallel Boads of Lochaber. 



position true, the shelf should certainly be marked in that 

 situation, and also round Glen Glaster ; but, according to Mr 

 Milne, it is to be found in neither of these places. The middle 

 shelf of Glen Roy, according to Agassiz, should also occur in 

 Glen Spean, between the two supposed glaciers ; but Mr 

 Milne asserts that, in fact, it does not. Thus, then, the gla- 

 ciers supposed by Agassiz will not satisfy the conditions of 

 the question ; nor will any other system of blockage do so, 

 except one, according to which Glen Glaster would, for a cer- 

 tain period, have been separated from Glen Roy. 



"We are, therefore, if we proceed on the supposition of the 

 agency of glaciers, led to the conclusion, that the one which 

 stopped up the mouth of Glen Roy to form its highest shelf, 

 must have extended up that glen beyond the mouth of Glen 

 Glaster. It must also have blocked up Glen Collarig nearly 

 to the place named the Gap. Then, to explain the formation 

 of the middle shelf, it is only necessai-y to suppose that the 

 glacier retired a little, so as to connect Glen Glaster with 

 Glen Roy. The water in the latter would immediately begin 

 to discharge itself by the ancient river-course before men- 

 tioned, and its surface would thus be lowered to the level of 

 the middle shelf. Lastly, the lowermost shelf of all would 

 be formed when the glacier retired to near the mouth of Glen 

 Spean. 



Mr Milne, however, asserts that, on account of the character 

 of the mouth of Glen Roy. in regard to levels, direction, and 

 distance from Ben Nevis, such a glacier as I have described 

 could not have existed ; but there does not appear to me to 

 be any real difficulty in the supposition. 



The following considerations will, I think, tend to render 

 this clear. Of all climates capable of generating glaciers, 

 there are two extremes which must produce two correspond- 

 ing extremes in the mode of distribution of the ice on the 

 surfixce of the earth. The one of these extremes of cli- 

 mate may be instanced as occuiTing in Switzerland, and the 

 other in the Antarctic Continent recently discovered by Sir 

 James Clarke Ross. In Switzerland the mean temperature of 

 the comparatively low and flat land is so much above the 

 freezing point, that the ice no sooner descends from the 



