IToUces of Memoirs — J. F. Blake — Mechanics of an Ice- Sheet. 511 



boulders. In the Harlech Mountains district areas showing these 

 opposite results lie side by side. Most of the glaciation is of the 

 negative kind, but the areas drained by the Crawewellt and the 

 Ysgethin are covered by glacial cones of dejection. This difference 

 is accounted for in the first instance by the local drainage being 

 opposite to the general drainage, and in the second by the small 

 size of the gathering ground for the ice. From these results it was 

 argued : 1. That drift deposits are, as a rule, left beyond the ai'ea of 

 ice-flow. 2. That no submergence could possibly have taken place 

 here since the Glacial period, or the features above noted would 

 have been obliterated. 



III. — On the Mechanics of an Ice-Sheet.— By the Kev. J. F. 

 Blake, M.A., F.G.S. 



THE author attempted to explain how an ice-sheet can carry 

 boulders up a slope, and leave them at a height of 1,000 feet or 

 more above sea-level. The sides of the channel are, in the first 

 instance, supposed to be parallel, so that the mass of ice may be 

 represented in a diagram by its longitudinal section. Taking, for 

 simplicity, the shape of the surface moved over to be represented by 

 two straight lines, one corresponding to the slope down from the 

 mountains, the other the slope up from the sea-bottom to the final 

 destination of the boulders, and, taking the surface of the ice as flat, 

 the ice-sheet is represented by a triangle. This is supposed to settle 

 down in such a way that, though the level of the end is higher, the 

 centre of gravity of the whole is lower. This fall of the centre of 

 gravity is the effective cause of the motion of the ice-sheet, the 

 resistance to be overcome being that of the ice to change its shape. 

 If the ice-sheet be supposed divided into strips parallel to the slope 

 from the mountains, these will be like a series of overlapping 

 glaciers, and under the influence of the pressure will swell out at 

 the bottom, and thus push the further end of the whole mass a little 

 way up the counter-slope. Continual additions of snow at the end 

 where the ice-sheet commences, or elsewhere on its surface, will be 

 cumulative in their effects, and thus the further end of the ice-sheet 

 will ultimately ascend as required. Again divide the triangle into 

 strips by lines parallel to the counter-slope. The lower of these 

 strips will be pressed together, and any point on the base will be 

 carried on in the direction of the whole motion at a greater rate than 

 the higher layers, and thus the stones, etc., on the sea-bottom will 

 be pushed up to their final resting-place, and anomalies of distribu- 

 tion might thus be accounted for by the previous dispersal of the 

 boulders. It was then shown that differences in shape of the ice- 

 sheet and its spreading out at the further end will make little 

 difference in the argument, and under certain conditions will aid the 

 motion. 



The author then discussed the question of the glacial erosion of 

 lakelets, and indicated the conditions under which this is possible, 

 particularly referring to the difference between an ice-sheet such as 

 that dealt with in the paper and an ordinary glacier. 



