J. M. Wilson — Forms of Valleys and Lake-basins. 483 



and so scraped over tliem instead of tearing them out from their 

 bed. It became perfectly plain that ice acts in this way, detaching 

 a block and pushing it down the valley, and thus can break up and 

 remove large masses of rocks, while the direct effect of attrition is 

 very small. 



Hence the forms of the sides and bed of the valleys seem to depend 

 on these divisional planes ; they have been worked down until they 

 offer a minimum resistance to the passage of the ice, which will slide 

 by a smooth divisional plane, or gradually wear away the edges of 

 masses that dip backward towards the direction from which the ice 

 is moving, but will wrench out a block which presents a face in the 

 direction of the ice stream. 



It may be illustrated, as my friend Mr. Sidgwick suggested to me 

 in one of these valleys, by sheets of cork, a material similarly jointed. 

 In certain positions your hand will slide over them and produce no 

 effect ; but in others, that is when the faces of the blocks meet the 

 hand, the blocks are torn up from the surface, and leave the blocks 

 below in precisely the same disadvantageous position, so that the 

 process will be continued. Or it may be illustrated by a pack of 

 cards. If they are thrown down so as to form a long stream — 

 preparatory to cutting for partners — your hand will glide over them 

 in one direction without displacing them ; in the other direction it 

 catches them and pulls them out of place. 



In one or two cases I found the dip of the planes down the valley ; 

 but these were at points where the slope was so steep that the dip 

 was up the ice stream though down the valley. 



It is impossible not to form the hypothesis that here may be found 

 the origin of lake-basins. For if we admit that ice in the form of a 

 glacier has vast pushing power, even though it has very slight 

 eroding power, it seems plain that wherever the divisional planes 

 are so bent as to form a series of concentric basins in the path 

 of a glacier, it will scoop out the blocks in following those divisional 

 planes and form a basin. If an existing lake-basin, for example, 

 were filled with brick-shaped masses, and a glacier moved down 

 upon it, one cannot doubt that it would have sufficient power, gradu- 

 ally to push these masses before it, and raise them up the opposite 

 slope. Unquestionably this action is in accordance with what we 

 see ice doing, and does not make those demands on our belief of the 

 eroding power of ice which are required by the ordinary erosion 

 theory. Moreover, several small lake-basins exist in Norway, of 

 which the form appears to be determined by the divisional planes. 

 It will not be difficult to put this hypothesis to the test of facts, and 

 thus to ascertain whether the condition for a lake-basin is the requi- 

 site curvature in divisional planes lying in the path of a glacier. I 

 may add one or two other remarks on points of geological interest. 



The glacier near Odde on the Hardanger Fiord, called the 

 Buerbrae, is worth special attention. It is easily visited in a day 

 from Odde, which is accessible by steamer in a day from Bergen. 

 It is an entirely new glacier, there having been no glacier in 

 the valley fifty years ago, as I was assured. It has advanced 



