29 



the land was laid bare and the usual erosive action began; the nunie- 

 rous torrents then formod by the dissolving glaciers were larger and 

 more powerful than the existing strcams so that they carried away 

 immense quantities of detritus. Rain and frost acled upon the basalt 

 — which in spite ofconsiderable hardness is soon elTected by weather — 

 and broke it down, as it continues to do to the present day. 

 Marine erosion, undoubtedly very powerful, also must not be for- 

 gotten. The Islands were once upon a time much larger, and the 

 sea must have played a prominent part in reducing their area. The 



Fig. 17. Osteri). View from >^Slattaratinder'' over the lofty mountain tops which f'oini a crest with 

 low curves. Débris lynig in strips on the hill-sides. (After a photograph hy the General StafTj. 



precipitous sea-cliffs bear distinct evidence of its erosive action, 

 and the faet of the dip of the strata being towards N. E. and S. E. 

 aids the sea from the west in its work of destruction, the tuff-layers 

 being thus more easily acted upon and demolished which in turn 

 brings about the fall of the basalt. The sea seized upon ever}"^ 

 weakness in the structure of the clifTs and undermined caves, and 

 as it gradually widened the latter the roofs above them grew unsafe 

 and at last gave waj', which must account for the vertical faces 

 of the shores and isolated rocks. 



The entire structure of the land points to its having been much 

 larger; thus the same beds may be traced from one island to an- 

 other, either across a sound or from one side of the fjord to the 

 other. On the other band, there is no strong geologicai evidence 

 to prove the theory, maintained, e. g. by James Geikie, that in post- 

 glacial times Scotland, the Færoes and Iceland were connected, 



