36 PROCEEDINGS COTTESWOLD CLUB 1910 



Johnson discovered that the great gaping crevasse which 

 generally comes between the head-walls of the cwm and 

 the glacier-ice went right down to the rock beneath the neve, 

 and that 



" the bed of the glacier, elsewhere protected from frost-work, was 

 here subjected to exceptionally rapid weathering. By maintaining 

 the rock wall continually wet, and by admitting the warm air from 

 the surface during the day, diurnal changes of temperature have 

 resulted in very appreciable mechanical effects, whereas above the 

 neve only the seasonable effects were important."' 



Indeed, Johnson's observation indicates that the maximum 

 excavation of the cwm-wall takes place not above, but below, 

 the neve, and shows " how a nearly perpendicular cirque wall 

 is steadily cut backwards through basal sapping at the bottom 

 of the bergschrund." 



Prof. Davis's theory may well account for the cwms that 

 appear to have been developed out of stream valley-heads ; 

 and Mr Johnson's, for the way in which the steep head-walls 

 steadily receded ; but Mr Johnson does not attempt to account 

 for the actual origin of the cwms. I confess to having experi- 

 enced some difficulty in accepting Prof. Davis's theory for the 

 origin of all the cwms — especially those in the rounded hills 

 now technically called " moels." Mr F. E. Matthes's very 

 interesting observations on how snow-banks, without move- 

 ment, steadily deepen the often very slight depression within 

 which they lie by a process which he calls " nivation " (that is, 

 excessive frost-work about the receding margins of the drifts 

 during the summer season), however, certainly help matters.^ 

 On Cleeve Hill, near Cheltenham, for example, there are 

 hollows that are very likely indeed being widened, and in 

 places deepened by this process of nivation. 



But for the present purpose nothing more need be added 

 on the probable origin of cwms, for those with which we are 

 really concerned have most likely originated in the way Prof. 

 Davis has suggested. In a mountain-mass like Snowdon, as 

 the contiguous cwms were developed, the hill-spurs were 

 sharpened into aretes ; the central summit became sculptured 

 to form a broken tooth-like projection ; and doubtless the final 

 products of the hanging-glaciers were moraine-heaps that now 

 ofttimes enclose the well-known little glacier-tarns. 



1 See Science, N.S., vol. i.x. (iSgt)), p. 106 ; and ibid., pp. 112, 113. 



2 Twenty-first Ann. Rep., U.S. Geol. Sur., 1809-1900, pp. 167-190. 



I 



