336 PROF. W. MORRIS DAVIS ON [Aug. I9O9, 



of Gwarfai Valley, fig. 30 (see also fig. 7, p. 308), where the narrow- 

 gauge railway picks its way among the crags. Such crags on the 

 channel (valley)-sides correspond to the hummocks on the slope 

 of the widened glacial channel east of Llyn Dinas. They are 

 strangely unlike the simple slopes of normally subdued mountains ; 

 they must impress one who, having gained his general impressions 

 in a non-glaciated district, sees them for the first time, as in some 

 way ' out of order ' ; for they do not appear to have been formed 

 by the down-hill streaming of water and the down-hill creeping of 



Fig. 30. — Sketch of crags near the head of Afon Givarfai Valley, 

 looking eastwards. 



waste. It is, however, quite possible that the rocks of normally 

 subdued mountain-masses may be so unequally penetrated by 

 weathering and decay, that a moderate amount of glacial scouring 

 would suffice to transform their sweeping slopes into the disorderly 

 craggy forms of glaciated mountains ; and perhaps such an amount 

 of erosion is not beyond the allowance of destructive work 

 attributed to ice by those who regard glaciers, on the whole, 

 as protective agents. Hence these features, whether on the floor or 

 on the sides of glacial channels, are not of critical value in making 

 choice between the alternative theories. 



Several examples of valley-side cliffs may be instanced. One 

 occurs near the head of Gwynant Valley, just under the south side of 

 Cwm Dyli, fig. 31 (p. 337). A remarkably fine cliff occurs on the east 

 side of the dome-like Moel y Cynghorion [so named on the Ffestiniog 

 (N.E.) quarter of Sheet LXXV of the Geological Survey map, but 

 without a name on the 1-inch Ordnance Survey Snowdon sheet, 

 3rd ed. ; its southern slope is shown in the left part of fig. 7, p. 308]. 

 Still other cliffs occur on the north-east and south sides of Mynydd 

 Mawr ; the first of these is roughly shown in fig. 3 (to left of centre) ; 

 the second, known as the Craig y Bere, is sketched in fig. 32 

 (p. 338). Under the theory of ice-protection, all these cliffs must 

 be referred to the work of normal erosion while the ice stood at 

 the next up-stream rock-step ; but it is unreasonable to think 

 that cliffs of this kind could be so produced, for there is no suffi- 

 cient peculiarity of structure to cause them. The first example 

 mentioned above occurs in a mass of volcanic ash, which is reduced to 



