314 



PROF. W. MORRIS DAVIS ON 



[Aug. 1909, 



are, as a rule, steeper than the uncovered side-slopes of the valley. 

 Assuming that the land-surface exposed to erosion is above grade, 

 it may be believed that the early stages of a cycle of erosion would 

 be marked by a relatively rapid incision of the (water or ice) 

 stream - channel beneath its initial profile, accompanied by a 

 steepening of the valley-sides above the stream-level; that in- 

 equalities in rock-resistance would for a time result in inequalities 

 of channel and stream-slope, causing cascades or seracs ; but that 

 the inequalities of stream-slope would, after reaching maximum 

 values, gradually be lessened, and at the same time the slope of 

 the valley -sides above the channel -banks would be decreased 

 by super-stream weathering. If glacial conditions continued in- 

 definitely, an ice-stream would, like a water-stream, eventually 

 reduce its surface-profile to an even and gentle slope : that is, the 

 stream would become graded ; and after the graded slope was 

 reached, further deepening of the valley would be extremely slow. 

 In the early stages of down-cutting, a main stream of water or 

 of ice might deepen its valley so rapidly that its surface would be 

 below the surface of its tributaries, which would therefore fall down 

 from their side-valleys to the main stream, as in figs. 4 and 15 ; 



Fig. 15. — Diagram of a young 

 main glacial trough, with 

 hanging lateral glaciers. 



Fig. 16. — Diagram of a mature 

 main glacial trough, with 

 accordant lateral glaciers. 



but as the main stream approaches grade, it will be overtaken by 

 its tributaries, which will then establish accordant junctions with 

 the surface of the main stream, as in figs. 5 and 16. It is, however, 

 of importance to note that, as Gannett pointed out ten years ago 

 (1898), while the junctions of the stream-surfaces may thus come 

 to be perfectly accordant, the bed of a larger, deeper main stream 

 and the bed of a shallower tributary must always be discordant 

 at their junction ; and, if the streams were removed, this dis- 

 cordance would be easily seen, and the bed of the tributary would 

 be described as hanging above the bed of the main stream. 

 Hollows or basins in the channel-bed might be expected in both 

 cases to result from variation in stream -velocity and in rock- 

 resistance : the size of the basins would be roughly proportional 

 to the size of the (water or ice) stream-channels. In the later 



