lxxxvi PROCEEDINGS OF THE GEOLOGICAL SOCIE1T. L-^ a 3' I 9°4» 



continued creeping-backward of the ice, their contents were 

 drained off to lower levels. A multitude of records of old water- 

 levels or ' strand-lines ' was thus left over the surface of the 

 country. It is the opinion of Scandinavian geologists that all 

 the terraces not of marine origin lie within that area. 1 



As one of the distinctive characters of the shore-lines left by the 

 glacier-lakes, the author of the ' Antlitz der Erde ' cites the occur- 

 rence of the rock-shelves or platforms (seter) eroded out of the 

 solid rock, and he refers the origin of these common features of the 

 fjords to the daily oscillations of temperature at the surface of the 

 lakes.- I shall try to show, by a reference to the abundant examples 

 of such rock-shelves in our own islands, that this explanation is 

 at least inadequate. If, however, for a moment, we grant that the 

 strand-lines, including the seter of the Norwegian fjords, do mark 

 levels of former freshwater lakes, it is obvious that, in order to 

 pond the drainage back and produce these lakes, the mouths of 

 the fjords must have been in some way blocked up by a barrier 

 which has disappeared. If this barrier were land-ice, as Prof. 

 Suess appears to assume, the water would rise behind it, until, 

 if the overflow found no escape into the Atlantic, it would pass 

 over the watershed, and joining the various bodies of water that 

 were there intercepted by the great Swedish ice-sheet, would 

 eventually find its way into the Gulf of Bothnia. There would 

 thus be two huge bodies of ice, between which the drainage was 

 accumulated. 3 AVe must remember, however, that the strand-lines 

 are not confined to the fjords, but sweep round the coast on either 



1 See two important papers by A. M. Hansen in the Christiania ' Archiv foi- 

 Mathemat ik og Naturvidenskaberne.' The first of these, toI. x (1886) pp. 329-52, 

 deals with the occurrence of seter or strand-lines in connection with ice- 

 dammed lakes at great heights above the sea, ranging from 652 to 1090 metres. 

 The second, vol. xiv (1890) pp. 254-343, & vol. xv (1892) pp. 1-96, contains a 

 full discussion of the character, distribution, and origin of the strand-lines of 

 Norway. See also G. de Geer, Sveriges Geol. Undersokn. Ser. C, No. 161, 1896 ; 

 & G. Andersson, ibid. No. 166, 1897, p. 5. Although tbe largest and most 

 abundant lakes, formed during the retreat of the ice-sheet, undoubtedly lie on 

 the eastern or Swedish side of the watershed, it is not improbable that others 

 were produced also on the western side by the irregular way in which the ice 

 disappeared. Dr. C. Sandler has suggested that the mouths of the Norwegian 

 fjords may have been blocked up by a succession of -vast moraines, which kept 

 back the sea and turned these sea-lochs into inland lakes. But the difficulties in 

 the way of the acceptance of this explanation are insuperable. See Petermann's 

 Mittheil. vol. xxxvi (1890) pp. 209, 235. 



2 ' Das Antlitz der Erde ? vol. ii (1888) p. 431. 



8 Dr. Sandler, in the paper already quoted, has considered the possibility 

 of such a flanking ice-dam, but has dismissed the idea as untenable. 



