ICE-EROSION IN THE CUILLIN HILLS, SKYE. 243 



ascend usually to heights of 800 to 1000 feet, or sometimes as much as 1300 feet, on 

 the higher moorland hills ; only in places near the coast does the drift-line sink much 

 lower, and even locally to sea-level. In the mountain- valleys the drift, always with 

 diminished thickness, runs up in tongues to altitudes of 1000 to 1500 feet, and 

 exceptionally 1750 feet. 



We are here referring to the tract within which the glaciation was strictly local. 

 The limits of this tract are partly laid down in the small sketch-map given above 

 (fig. 1). Within it, the boulders are wholly derived from the tract itself, thus present- 

 ing a marked contrast to the area outside the line drawn. In the south-eastern part 

 of Skye, Mr Clough has found foreign boulders even up to the highest summit (Sgurr 

 na Coinnich, 2400 feet). The same observer notes along Loch Eishort a certain 

 intermingling of boulders from different sources, which may be taken to indicate some 

 oscillation of the line where the native and foreign ice marched together. Such 

 oscillations, depending on the balance of varying pressures, are to be expected. On the 

 north-eastern side of our area the domain of the native ice is delimited with sufficient 

 closeness. Especially marked is the contrast between Scalpay and the neighbouring 

 part of Skye. On the smaller island, which was overflowed by the great ice-sheet 

 from the mainland, boulders of Scottish extraction occur at all altitudes, and in the 

 boulder-clay no less than on the surface, while further points of difference are apparent 

 in the nature and distribution of the drift and in the form of the ground. Raasay 

 evidently falls under the same head. Our detailed survey of Skye has not progressed 

 far enough northward to trace the boundary-line in that direction, but the breaking 

 in of the Scottish ice over the northern part of the island seems to be sufficiently 

 established. 



A closer examination of the materials of the local drift affords much information 

 relative to the movement of the Skye ice, and also leads to observations which have a 

 very direct bearing upon the amount and the mechanics of glacial erosion. Since in 

 the area chiefly involved the number of local rocks which can be readily identified is 

 not great, it becomes necessary to take note of the relative proportions in which 

 different rocks enter into the composition of the accumulations. The writer has found 

 that for this purpose general impressions are not to be trusted, and he has followed 

 as far as possible the statistical method. As an illustration of the use of this, as well 

 as for some of the results obtained, we will follow the course of what we have already 

 referred to as the mid-stream line of the Glen Sligachan ice (see map below). To lay 

 down this line, about a hundred convenient stations were selected, and at each station 

 from 200 to 500 boulders were taken without selection from the drift and the per- 

 centages of different rocks estimated. 



Looking down Glen Sligachan from the watershed at Loch Dubh, the observer 

 has the Red Hills on his right and the Cuillins on his left. The former are essentially 

 of granite, the latter of gabbro ; the line of junction of the two rocks running for some 

 distance along the bottom of the valley. Accordingly granite boulders preponderate 



