228 MR ALFRED HARKER ON 



essential to some of the considerations to be adduced below to note that the pre-Glacial 

 drainage-system of Skye was a fully matured one. The igneous rocks are of Eocene age, 

 and perhaps in part later, but the}^ had undergone an enormous amount of erosion 

 during the latter half of Tertiary time, prior to the advent of the Glacial epoch. Bear- 

 ing in mind that, under the conditions now existing, erosion is practically at a stand- 

 still, it is difficult to resist the inference that during the carving out of the pre-Glacial. 

 valleys the land stood somewhat higher than it does to-day. At the close of the 

 glaciation it stood about 1 00 feet lower than at present. These and other considera- 

 tions may suffice to assure us that at the time when glacial conditions were initiated the 

 land had not experienced any recent elevation : the drainage-system was a fully 

 established one, with valleys adapted to the streams which flowed in them and with the 

 normal adjustment of tributaries to principal valleys. The point is of importance in 

 comparing the types of topography due to ice- and water-erosion respectively, for some 

 of the most instructive points of contrast do not hold good in their full degree unless 

 this condition is realised. 



It is part of the plan of the present contribution to confine attention to the 

 special phenomena of the selected area, which, as has been indicated, has peculiar 

 claims to be regarded as a type, and accordingly few references will be made to 

 other districts. Still less is it within our province to discuss in its generality the 

 much-vexed question of the degree of importance to be attached to ice as an erosive 

 agent. Having regard to the mechanical element in erosion only, it is manifest 

 that a sand-grain gripped in the sole of a glacier, or of an ice-sheet thousands of feet 

 in thickness, must be incomparably more efficient as a graving-tool than the same 

 grain rolled along the bed of a stream. The question, therefore, from the a priori 

 point of view, turns upon the rate of working and the duration of the requisite 

 conditions. As regards the former point, it is to be observed that, where a groove 

 lias been cut in a rock-surface by abrasion by an individual sand-grain dragged along 

 it, the time required to cut the groove was obviously the time taken by the grain 

 in travelling the length of the groove. Assigning even a low rate of flow to the 

 ice, and allowing for the sand-grain lagging behind the ice in its movement, this 

 consideration still suggests that the removal of material thus effected by ice well 

 supplied with rock-debris in its lower layers must be a rapid process as compared 

 with anything that can be effected by the agency of running water. The duration 

 of the Ice Age is a question upon which we cannot enter in this place. 



It is beyond doubt that the carving out of the mountain and valley system of the 

 Uuillins is, as regards its broad outlines, the result of aqueous erosion during the 

 latter half of Tertiary time ; but it is no less certain that the actual details of the 

 relief, as we now see them, are to be credited to the action of ice and frost during 

 the Glacial period. If we have regard to the total amount of material removed, we 

 must recognise water-erosion as the chief factor in the result ; but from the point of 

 view of earth-sculpture it is to glacial erosion that we must assign the more important 



