GLACIAL EROSION IN THE AAR VALLEY 591 



part of the valleys which they occupy. Water streams are nimble ; the cross-sec- 

 tion of their beds is but a small part of their valleys. Moreover, river beds are 

 seldom visible, being usually occupied by water; but today the beds of former 

 glaciers are evacuated by ice, and are so plainly laid bare that they are generally 

 regarded as a part of the valleys in which they have been excavated. The water 

 surface of rivers and the floor of their valleys are graded to accordant junction, 

 but if the water were dried away the bed of a small branch stream would be found 

 distinctly shallower than that of a large main trunk ; a distinct break might occur 

 between the level of the two. So with valleys that are now occupied by glaciers. 

 The surface of the confluent ice-streams and the slopes of the valleys above ice- 

 level are normally related to each other, but the beds of the smaller ice-streams 

 are not worn so deep as that of the larger ones, and when the ice disappears the 

 discordance of the larger and smaller glacier beds may become a striking feature 

 of the landscape. The water streams which then perform the drainage of the region 

 may for a considerable post-glacial period possess strong cascades where the side 

 valleys open on the main valley walls. 



G. K. Gilbert said 



That glaciation was not the only process causing discordance between main valleys 

 and their tributaries, and cited the canyon of the Virgen river through Juratrias 

 sandstones and the canyon of the Columbia river through the Cascade range as 

 instances of trunk streams corrading so rapidly as to outstrip their feeble affluents 

 and leave the truncated gorges of the latter stranded high in the canyon walls ; but 

 such a process does not produce the broad-floored hanging valleys of glaciated 

 regions, and the elevation of these valleys he believed to record the difference 

 between the powerful action of trunk glaciers and the feebler action of side glaciers. 

 He had been greatly impressed, years ago, by the magnitude of the glacial excava- 

 tion indicated by such phenomena in the high Sierra, and last summer had found 

 the coast of Alaska replete with similar evidence. After sailing for weeks through 

 Alaskan fiords and observing scores of hanging valleys, he had come to regard their 

 occurrence as diagnostic of the former extent of glaciation, and had used them with 

 confidence as criteria for the discrimination of glaciated districts. 



H. W. Turner spoke as follows : 



The hanging valleys of the Sierra Nevada are considered by some writers as evi- 

 dence that the canyons were formed by glacial erosion, but in some portions of the 

 range we will find these hanging valleys bordering canyons which never contained 

 glacial ice. Such an example is on the Bidwell Bar quadrangle in the northern 

 Sierra Nevada. The canyons of the north and middle forks of the Feather river, 

 in this quadrangle, are largely in granite. They are deep and rugged, one portion 

 of the North Feather canyon being nearly a mile deep, as measured from the highest 

 part of the plateau to the east. The side streams enter these canyons from hang- 

 ing valleys over precipitous slopes in cascades and falls, one of which has a height 

 of 450 feet or more. While small glaciers existed during the Glacial period on the 

 plateaus about these canyons, there is no evidence that glacial ice ever occupied 

 any portions of the main canyons. As is well known, the whole range during the 

 Tertiary had been greatly worn down, so that broad shallow valleys were charac- 



LXXXIII— Bull. Geol. Soc. Am., Vol. 11, 1899 



