198 Proceedings of the Royal Irish Academy. 



with denudation-grooves at fairly regular intervals, reminds one of the arid lands 

 of Arizona (PL XII., figs. 7, 8). When the fringe of snow upon the plateau- 

 edge has melted back, the grooves run dry. Snow lies for a time in some of 

 them, but is got rid of in many cases by a few sudden slides. Blocks loosened 

 from the heights by frost, or even wind, occasionally fall down these grooves, 

 raising clouds of dust, as if an explosion had occurred. It was delightful to 

 find on an exposed plateau al:)ove Cape Wijk typical wind-etched stones, the 

 " dreikanter " of desert lands. The surfaces of the talus-cones that reach back 

 from the shore become also absolutely dry. The same is true of the surface of 

 the tundra-land formed by the raised beaches, though the ground may be 

 frozen, including plenty of solid water, at a depth of half a metre. The 

 abundant vegetation on the cracked surface, including Scdix polaris about 

 half an inch high, and numerous flowering plants, obtains its moisture from 

 the intermediate layer, into which a stick can be easily thrust, and along which 

 a good deal of fluvio-glacial water finds its way in the neighbourhood of the 

 delta-fans. 



Here and there, on the arid mountain-walls, the coalescence of a few 

 grooves has allowed of the formation of a larger hollow, the debris from which, 

 as Eichter pointed out in the ease of cirques in Norway, can be discharged 

 over the surface of the neve in its floor.' Some of these, like the great hollow 

 in Mt. Pyramid, may be clear from snow in summer (PI. XL, fig. 5). If the 

 hollow has been primarily excavated by the destructive action of a snow patch 

 (p. 196), it may develop into a true cirque. On the seaward face of the Spits- 

 bergen plateaus, such cirqiies resemble hanging valleys, not because the main 

 fjord has been overdeepened, but because they have been worked back by frost- 

 action faster than their floors could be eroded. As has been pointed out, they 

 are still in process of formation, though the streams that run out of them in 

 summer obviously tend to reduce their steep outer lips on the fjord-wall to 

 ordinary valley-slopes, and to bring down their mouths to the local base-level 

 of erosion. Some of these cirques, occupied by glaciers in the Ice-Age, retain 

 traces of ice-smoothing in the forms of their floors and outer lips, despite the 

 action of the frost (PI. XIII., fig. 9). As a rule, however, atmospheric 

 weathering in Spitsbergen speedily removes such evidence. = 



Comparison of the Conditions of Erosion in Spitsbergen with those 



IN Ireland. 



The conditions under which glacial and atmospheric erosion are taking 

 place in Spitsbergen — a region where the Ice-Age is still passing — are very 



' Compare G. H. Kinahan, " Tbe Recent Irish Glaciers," Irisli Naturalist, vol. iii. (1894), p. 236. 

 = Compare E. J. Garwood, "Glacial Phenomena of Spitsbergen," Quart. Journ. Geol. Soc. 

 London, vol. Iv (1899), p. 6SS, aud W. Salomon, op. cit., p. 307. 



