THE LAKES OF SNOWDONIA AND EASTERN CARNARVONSHIRE. 457 



but also shatter and disrupt rock-masses. This rock-shattering has been observed by 

 Professor Simony on the bed of one of the Dachstein glaciers during the temporary 

 retreat of the ice. Similar phenomena have also been recorded by MM. Penck, 

 Bruchner, and Baltzer at the Uebergossenen Aim and other places. The ddbris formed 

 by the breaking up of the rock had been incorporated in the ground-moraine lower 

 down. 



But while such observations prove that moving ice is sometimes a great eroding 

 agent, capable of crushing and shattering the rocks over which it flows, instances have 

 also been recorded of glaciers moving over beds of gravel, sand, and other unconsolidated 

 material without scooping them away, or even disturbing the accumulations to any great 

 extent. This shows that a glacier varies in its action, being, under certain conditions, a 

 disturbing agent, and, under others, a preserving agent. 



Professor W. M. Davis, of Harvard, has recently made an important contribution 

 to glacial geology, by bringing forward fresh and striking evidence in favour of the 

 excavating powers of a glacier. This is all the more interesting in view of the fact that 

 Professor Davis had hitherto been a doubter of the erosive power of ice. In a paper 

 which appeared in Appalachia, March 1900, he gives an account of his observations on 

 " Glacial Erosion in the Valley of the Ticino," and he deals again with the same subject 

 in the Proc. of the Boston Soc. of Nat. Hist., July 1900, where the question of the 

 competence of glaciers to deepen and widen their valleys is more fully discussed. He 

 was struck by the fact that in the Ticino Valley the side valleys are strongly out of 

 accord with the main valley. " There is no nice adjustment of declivities, for the side 

 valleys mouth several hundred feet up on the wall of the main valley, and the side 

 streams cascade down in sharp-cut shallow clefts from their high perches to the main 

 valley floor." The walls of the main valley are steep, and run sub-parallel for miles 

 together ; no spurs enter the valley floor, and so it lacks the successively overlapping 

 profiles seen in normal valleys, and instead of being V-shaped, it is U-shaped on cross 

 section. At a height of 400 to 600 metres above the main valley floor, a bench of 

 gentler slope leans back from the top of the basal cliffs. " The benches seem to be 

 the remnants of the lower slopes of an ancient wide open valley, in whose floor the 

 present cliff-shaped deeper valley has been eroded, and this supposition is confirmed 

 when it is found that the high-standing lateral valleys are systematically related to the 

 ancient rather than to the modern valley floor." He shows conclusively that the cross- 

 ing spurs of the ancient river valley have been destroyed by glacial erosion, and that a 

 glacier has deepened and widened the main valley so as to give it the form of a trough, 

 with smooth and steep sides. The term " over-deepened " has been applied by Pence: to 

 valleys of this kind. Davis adds that " the deepening of a glaciated valley for a good 

 part of its length is thus seen to be a general result of glacial erosion ; it is accompanied 

 throughout by discordant or hanging lateral valleys, and the production of a lake in the 

 distal portions of such a valley is but a subordinate result of glacial erosion." At the 

 time when the valleys were filled with ice, the surface of the trunk and branch glaciers 



