Prof. A. E. Nordensh'idld — Geology of Spitzbergen. 17 



fiords, the way in which the channels referred to hranch out towards 

 the interior, to terminate abruptly in an old glacier moraine, or with 

 a glacier still in action, we see clearly that the common theory of the 

 formation of valleys, that is to say, the supposition that valleys have 

 been formed in tlie Plutonic way through a sinking of the earth's 

 strata, is not applicable here, but that the fiords of Spitzbergen, the 

 water-filled valley-depressions in question, were formed after the 

 close of the Miocene period through the denuding action of ice- 

 streams from an inland ice-field, which probably was considerably more 

 extensive than that which now occupies the interior of Spitzbergen. 



Although it is probable that Spitzbergen was not covered by 

 glaciers before the end of the Miocene period, it can be proved that 

 on that part of the world also glaciers occupied a greater extent in 

 former times than now. If we examine the rocks at the sides of 

 the fiords and the small islands which environ the coast itself, 

 numerous proofs of this are to be found. The rocks in these regions 

 are in most cases indeed so split up by frost that the original surface 

 has been completely destroyed, and generally the rock is too loose 

 and easily crumbled down to permit any glacial scratches to con- 

 tinue for a length of time upon a surface exposed to the air. But 

 sometimes there occur yet harder rocks, gneiss, quartzite, diabase, 

 etc., which are better adapted to resist the destructive action of the 

 atmosphere ; and on such rocks, when they lie near the water's 

 edge, there are nearly always found, on a close examination, beauti- 

 ful strise and furrows, which run in the same direction as the fiord. 

 From this we draw the conclusion that the now existing fiord was 

 formerly filled with a glacier or ice-stream, and that the glaciers 

 on the west coast at a former period extended at least to that garland 

 of rocks and small islands by which the land is now environed. 

 Judging from the absence of deep fiords on the east coast of Spitz- 

 bergen, and the shallowness of the sea between Spitzbergen and 

 Nova Zembla, we infer that the inland mer de glace in former times 

 extended considerably further in that direction. Probably during the 

 Glacial Period the luest coast of Spitzbergen tvas the west coast, not 

 merely of a large island, but of a considerable Arctic continent, which 

 towards the south ivas connected with Scandinavia, and towards the 

 east with continental Siberia. 



I have never seen any striated rocks at a height exceeding 1000 

 feet upon the mountains of West Spitzbergen. The reason probably 

 is the destructive action of frost upon the rock-surface. 



But it is worthy of notice that while the gneiss and granite moun- 

 tains nearly 1800 feet high, on Parry's, Phipps's, Marten's, and 

 Castren's Islands, all lying on the north coast of North East Land, 

 upon one side, chiefly the south, exhibit rounded forms, which clearly 

 show that the mountains were completely covered by glaciers; the 

 rocks of the same varieties on the north-west corner of Spitzbergen, 

 on the contrary, do not give any certain evidence of having been 

 entirely covered by ice. 



I cannot adduce any conclusive proof that the glaciers are con- 

 tinuously receding. On the contrary, judging by the last centuries' 



DECADE II. — VOL III.— NO. I. 2 



