230 MR ALFRED HARKER ON 



action. In the writer's opinion, such analysis is to a great extent practicable, and an 

 essay towards it is offered in the following divisions of this section. The mechanics 

 of ice in bulk presents in its entirety a difficult physical problem, as yet unsolved. 

 It follows that we have no basis for a full a priori discussion of the mechanism of 

 glacial erosion, and the attempts hitherto made on this line are necessarily inconclusive. 

 M' Gee's investigation,* for example, involves in many parts conflicting elements, the 

 relative magnitudes of which are not known. But although we have no working 

 theory of ice-erosion, we have a theory of water-erosion which is complete in most 

 essentials and is amply justified by the results to which it leads. It supplies us with 

 most important laws built up on a few simple principles. These fundamental 

 principles are proper to water but alien to ice, and this must be true of the con- 

 sequences deduced from them. Thus, although it may be difficult to lay down a priori 

 the laws of ice-erosion, it is not difficult to see in many cases how they must differ 

 from the laws of water-erosion ; and different laws will find their expression in different 

 topographic forms. This is the point of view to be adopted here. The differences in 

 question fall conveniently under five heads. 



(v.) Independence of Physical Features and Geological Structure. 



A study of the actual topography of the Cuillins shows that, under the conditions 

 that there prevailed, ice-erosion is controlled in a much less degree than water-erosion 

 by lithological differences and geological structure. 



One consideration which would lead us to anticipate this difference is sufficiently 

 evident ; while in the case of water-action the process is effected by the co-operation 

 of chemical with mechanical disintegration, in ice-action the chemical factor is 

 minimised or wholly in abeyance. We shall have to notice below, other circumstances 

 which conduce to simplicity of form on glaciated land-surfaces, and so tend to overrule 

 the expression of geological constitution in surface-relief. 



The general principle propounded is beautifully illustrated in the Cuillins. This 

 group of mountains, remarkably simple in constitution in a broad view, is in detail 

 highly complex. On the summit-ridges and on many parts of the higher slopes this 

 complexity of structure expresses itself in the form of the ground. Many of the 

 dykes give rise to gullies and notches, and the parallel intrusive sheets of dolerite 

 impart something of a step-like character to the slopes. All this becomes most marked 

 in those places which have suffered most from frost-weathering after the epoch of ice- 

 moulding. On the floors of the corries, and in the main valleys, i.e. in places where 

 the maximum effects of glacial erosion have been experienced and the resulting surface 

 remains intact, the appearance is very different. Here we see gabbro, basaltic lavas, 



* W. J. M'Gke, "Glacial Canons," Joum. of Geol., vol. ii. pp. 350-304, 1894. 



