ANTARCTIC GLACIOLOGY 339 



between air temperature and exchange of radiation to and from the 

 earth. When outward-directed radiation is in excess, as is usually 

 the case, the intensity of the glacial anticyclone may be expected to 

 decrease when the area of the loose snow surface is at a minimum. 

 The conditions are rendered more complicated by the important 

 effect of the downward-blowing winds in sweeping the loose snow 

 surface towards the edge of the continent and in "packing" the snow 

 during its passage. Nothing definite is known, either, of the relation 

 between evaporation and the intensity of the glacial anticyclone, 

 though there is good reason to expect that evaporation will be greatest 

 when the intensity is greatest, in otherwise similar conditions. Suf- 

 ficient has been said to emphasize the importance of a more complete 

 knowledge of the meteorological and physical conditions associated 

 with the glacial anticyclone. To some extent we may say the chief 

 importance of such knowledge will lie in the help it will afford towards 

 determining to what extent the glacial anticyclone which must have 

 been associated with any large ice mass in the past may have carried 

 with it conditions that tended to limit or control the dimensions of 

 the ice mass to which it owed its origin. 



Small-Scale Ice Erosion: Cwms and Snowdrift Hollows 



A problem on a smaller scale, but one that has exercised much 

 thought and caused much controversy in the past, is that concerning 

 the mode of growth of the smallest of the permanent ice formations 

 of the land. How do the permanent or semi-permanent snowdrifts 

 of glacierized lands deepen the hollows in which they lie until the 

 latter can be dignified by the title of cwms? What, again, are the 

 processes by which these cwms become deepened to such an extent 

 as to give rise to glaciers capable of eroding their beds in the normal 

 manner? For the latter, movement is required, and it seems certain 

 that several hundred feet of ice must accumulate before movement 

 at any pace can take place. Nivation and bergschrund erosion as 

 explained in papers and textbooks are certainly not sufficient to 

 account for these phenomena in Antarctica today. What, then, is 

 the truth? Are the present Antarctic snowdrift hollows and cwms 

 relics of former different climatic conditions but now the containers of 

 stagnant masses of ice that exert a wholly conservative influence? 

 If so, many of them should be gradually disappearing through the 

 action of subaerial erosion, which is active on all exposed land masses 

 in Antarctica today. Such truncated cwms have been described 

 from more than one region of the continent. Bergschrunds do exist; 

 nivation does operate in summer around the edges of the snowdrifts 

 which nestle in the lee of all exposed rock faces. Nevertheless, it 

 seems certain that in the normal Antarctic summer no melting at 



