Vol. 65.] 



GLACIAL EKOSION IN NOKTH WALES. 



337 



a graded slope a little farther down the valley. The second occurs 

 in a body of uniform slates, which are elsewhere, except in cwm- 

 heads, reduced to graded slopes. The two cliffs of Mynydd Mawr 

 might, at first sight, be referred to the action of normal erosion on 

 a body of 'intrusive hornblende-porphyry,' which constitutes a 

 large part of the mountain-mass ; but a closer examination must 

 lead to a different conclusion. The cliffs do not, according to the 

 geological map, follow the border of the porphyry and overlook the 

 neighbouring slates, as would be the case if the resistant porphyry 

 and the weaker slates had been acted on by normal erosion ; hence 

 these cliffs, like the others, are suggestive of some abnormal process. 

 Under the theory of ice-protection, no adequate explanation for 



Pig. 31. — Sketch of ragged cliffs under Ctvm Dyli, looking 

 northwards from near Lake Gtuynant. 



cliffs such as these has been given. Under the theory of ice-erosion, 

 they may be accounted for in two ways, according to their position. 

 The rugged cliffs beneath Cwm Dyli and on the north-east side of 

 Mynydd Mawr rise directly from their valley-floors, and represent 

 channel and valley-sides in an unfinished stage ; longer duration 

 of ice-action would have made them smoother. The first of these 

 two examples, sketched in fig. 31 (above), is at a sharp turn in the 

 path of the Cwm Dyli glacier, and is not high enough to have 

 reached up to the ice-surface. Here glacial rasping would have had 

 to do much more work than has been done, before an easy turn could 

 have been made ; the cliff has a steep and ragged rock-face, because 



