Cole — Glacial Features in Spitslergen. 197 



the wall of the plateau ; and alluvium is carried in summer into the valley- 

 floor. The heads of three of these streams have notched the plateau a little 

 more conspicuously than their neighbours. Nests of snow have been able to 

 linger in them, and have worked out recesses year by year. The edge of the 

 plateau is here viewed from a height of about 500 metres and a distance of 

 4 kilometres (2i miles). These incipient cirques are already hollows of some 

 magnitude. Behind them a much older cirque has penetrated the plateau, 

 and has effectually reduced the area on which snow can gather during 

 winter. As this erosive process goes on, aretes and peaks will develop ; and 

 the plateau-landscape will give place to one of the Snowdonian or ordinary 

 Alpine type. 



A. Penck/ in his masterly survey of Alpine glaciation, looks back to a time 

 when some regions were covered by continiious iirn (neve) or glacier-ice, and 

 here cirques have not developed. He urges that cirque-building requires a 

 steep mass of rock aliove the snow-line, and rising high above the general field 

 of n^ve. The neve, moreover, must be sharply divided by a marginal cleft, 

 the bergschrund, from the wall of rock behind it. Penck, however, must have 

 also had in mind the possibility of the development of a cirque-landscape, or 

 " karling," on the edge of a great snow-dome or plateau — in fact, lielow the 

 general snow-line of the district. The shelter given to snow as the cirque 

 increases in depth, and the continued drift of snow into it from abo-\'e, conspire 

 to carry down tlie snow-line locally. 



Kichter- has contrasted the erosion-form of a niche originated by running- 

 water with that of one into which snow falls. He recognizes that true cirques 

 arise " in der Schneeregion oder ihr sehr nahe," and that the snow which collects 

 in such hollows protects their floors from becoming funnel-shaped, like those of 

 hollows eroded by running water. But he does not seem to consider the 

 possibiUty that a funnel formed by summer-rills from melting snow may be 

 converted into a cirque when snow slips into it from above, and that its whiter 

 type may ultimately prevail throughout the year, without the occurrence of a 

 clhnatic change. 



This seems to me to have been the case in Spitsbergen ; and snow eveii now 

 remains in hollows, when it has melted off the overlying plateau. Such snow 

 is stiU developing the cirques. 



Arid Type of Weathering. 



Where the plateau has not been dissected back from the margins of the Ice 

 Fjord by any promment stream of ice or water, the uniform type of weathering, 



1 Die Alpen iiii Eiszeilalter (1909), pp. 284-287. 



-Op. cit., Sitzungsber. k. AkaJ. "Wiss. Wien, Bd. cv., Abt. 1, pp. 156 and 158. 



