259 



pointed out, but it only occurs in the more pronounced moun- 

 tains in the N., in the highest parts of the land, close upon 

 1000 m above the sea, and in some deep chasm valleys. The 

 plateau itself is in the summer, even at a height of 800 — 900 m, 

 almost free from snow. A good picture of this remarkable 

 country is shown in Og. 23. A satisfactory explanation of the 

 absence of snow has not yet been offered. Very probably the 

 plateau character has been unfavourable to the formation of 

 ice, both because it discourages atmospheric precipitation, and 

 because the snow that falls here is easily carried away by the 

 wind; and it is possible that the precipitation at the level that 

 the plateau reaches here within the fjord is, apart from this, 

 less than out by the coast. But this does not seem to be a 

 sufficient explanation. Polar explorations have now shown us 

 that in different regions tracts of land, whose rocky bed consists 

 of horizontally lying, loose sedimentary rocks, are unfavourable 

 to ice formation, even if they lie on an open coast. A short 

 resumé of the observations connected with this has been 

 furnished by J. G. Andersson^), who has promised to return 

 to the question. Without, therefore, going into the matter 

 here. I will merely suggest as a possibility — applicable to 

 the present case — that the freedom from snow is directly 

 attributable to the loose nature of the soil, to its porosity. 

 On projecting rocks of sandstone specks of snow can still be 

 found, though not nearly to the same extent as in the areas 

 of primary rock; on a soil of gravel or sand there is not a 

 trace of ice, save a few snow-drifts down in the valleys, which 

 probably melt before the arrival of winter. Why the nature 

 of this soil should have this effect is harder to explain. Per- 

 haps, by letting the thawed water from the snow above sink 

 down, it prevents the formation of a crust of ice on the ground, 

 or in some other way prevents the snow turning into ice, which, 



M Bull. Geo). Inst. IJpsala, Vol. VII, p. 24. 



IT 



