Vol. 65.] GLACIAL EROSION IN NORTH WALES. 345 



evolution of that area. The. three following features were con- 

 sidered by the speaker as characteristic of glacial erosion: — (1) the 

 cirques ; (2) the steps interrupting the valley-floors ; and (3) the 

 lake-basins. Besides these three features, there were two other 

 kinds of land-forms of equally frequent occurrence in glaciated 

 regions, and, as the speaker had been often puzzled by their origin, 

 he would like to take that opportunity of asking the Author to 

 apply his deductive method to the explanation of the origin of the 

 following two classes of abnormal features : — (1) The isolated 

 knobs occurring here and there in glaciated valleys had already 

 been referred to by the Author as truncated spurs of a former 

 river-valley. This ingenious explanation, although correct in many 

 cases, did not seem applicable to all the instances. Would it be 

 correct to consider some of the knobs as the result of the remark- 

 able way in which the ice eroded differentially according to the 

 hardness of the rocks ? If this were the case, some of these knobs 

 would be merely the giants of the ' roches moutonnees,' or, to put 

 it in a more up-to-date way, ' monadnocks of glacial origin.' 

 (2) Transverse rock-barriers or ' riegel ' were known to stretch 

 frequently across glaciated valleys. In Switzerland some of the 

 most notable of these barriers had a tendency to occur near the 

 junction of two different formations : for instance, both the trans- 

 verse barrier near Meiringen and that near St. Maurice (Valais) 

 were formed of Mesozoic limestones, and stood close to the junction 

 with the crystalline rocks. Both were cut by a narrow gorge which 

 seemed to be the result of subglacial water-erosion, because no 

 terraces and no other indications of a temporary lake were to be 

 found above these barriers. At all events, the Author was perhaps 

 inclined to carry a little too far ' the indifference of topographic 

 form to the trend of formation-boundaries,' in order to reduce 

 every type of scenery to a simple formula where the three ratios 

 worth considering according to him were : (i) the structure ; 

 (ii) the agencies at work ; and (iii) the stage of the cycle. The 

 speaker was tempted to continue to think that the nature of 

 the rocks was worthy of equal consideration, even from a mere 

 geographical standpoint, as introducing into the landscape a specific 

 characteristic of its own. 



With regard to the term ' geographical cycle,' frequently used by 

 the Author, ' cycle ' always suggested to the speaker the idea of a 

 complete circle, of a return to the primitive state of things. As 

 this was not exactly the case in geomorphogeny, where the final 

 peneplain was not entirely comparable with the original structural 

 surface, perhaps it would be more accurate to speak of the ' evolutive 

 stages of erosion.' 



The Rev. E. C. Spicer, referring to glacial cirque-formation, 

 quoted the ' Journal of Geology ' (Chicago) for 1904, vol. xii, p. 569, 

 where Dr. Willard D. Johnson gave an apparently satisfactory 

 explanation of the phenomena. 



Dr. A. P. Young said that it might be interesting to recall the 



