6 president's address. 



temperature ' declined snow would begin to accumulate in inequalities 

 on the upper slopes. This, by melting and freezing, would soften and 

 corrode the underlying material, which would then be removed by rain 

 and wind, gravitation and avalanche. In course of time the hollow 

 thus formed would assume more and more the outlines of a corrie or a 

 cirque by eating into the hillside. With an increasing diameter it would 

 be occupied, as the temperature fell, first by a permanent snowfield, 

 then by the nevi of a glacier. Another process now becomes important, 

 that called ' sapping.' While ordinary glacier-scour tends, as we are - 

 told, to produce ' sweeping curves and eventually a graded slope,' 

 ' sapping ' produces ' benches and cliffs, its action being horizontal and 

 backwards,' and often dominant over scour. The author of this hypo- 

 thesis 2 convinced himself of its truth in the Sierra Nevada by descending 

 a bergschrund 150 feet in depth, which opened out, as is so common, 

 beneath the walls of a cirque. Beginning in the n&v&, it ultimately reached 

 the cliff, so that for the last thirty feet the bold investigator found rock on 

 the one hand and ice on the other. The former was traversed by fracture 

 planes, and was in all stages of displacement and dislodgment; some 

 blocks having fallen to the bottom, others bridging the narrow chasm, 

 and others frozen into the n&v6. Clear ice had formed in the fissures of 

 the cliff ; it hung down in great stalactites ; it had accumulated in stalag- 

 mitic masses on the floor. Beneath the n&v& the temperature would be 

 uniform, so its action would be protective, except where it set up 

 another kind of erosion, presently to be noticed ; but in the chasm, we 

 are informed, there would be, at any rate for a considerable part of the 

 year, a daily alternation of freezing and thawing. Thus the cliff would 

 be rapidly undermined and be carried back into the mountain slope, so 

 that before long the glacier would nestle in a shelter of its own making. 

 Further down the valley the moving ice would become more effective 

 than sub-glacial streams in deepening its bed ; but since the n^-fiow is 

 almost imperceptible near the head, another agency must be invoked, 

 that of ' plucking.' The ice grips, like a forceps, any loose or projecting 

 fragment in its rocky bed, wrenches that from its place, and carries it 

 away. The extraction of one tooth weakens the hold of its neighbours, 

 and thus the glen is deepened by ' plucking, ' while it is carried back 

 by ' sapping. ' Streams from melting snows on the slopes above 'the 

 amphitheatre might have been expected to co-operate vigorously in 

 making it, but of them little account seems to be taken, and we are even 

 told that in some cases the winds probably prevented snow from resting 

 on the rounded surface between two cirque-heads. 3 As these receded, 



1 In the remainder of this Address • temperature ' is to be understood as mean 

 temperature. The Fahrenheit scale is used. 



2 W. D. Johnson, Science, N.S., ix. (1899), pp. 106, 112. 

 * This does not appear to have occurred in the Alps. 



