MATURITY IN ALPINE GLACIAL EROSION 575 



stage, will be defeated by the rapidly increasing difficulty of waste 

 removal. The great arc-form crevasse at the glacier head, there- 

 fore, may be the indirect cause, not merely of recession of the canyon 

 head at a low grade into a high grade, but of recession at a grade 

 declining from the horizontal. Apparently there is a limit to such 

 extension — a limit the earlier reached, the sharper the acclivity of 

 the upland surface into which channeling is extended; and the head 

 wall, in consequence, breaks back into steps, successively shortening 

 in length of tread. The rearward steps may continue to be marked 

 by schrunds rising to the glacier surface; living glaciers, in fact, are 

 often characterized by *' cascades" in their upper courses; but sharply 

 defined cross-cliffs, in empty glacial canyons, are rarely to be found 

 far down-stream. Presumably they are so deeply buried there as to 

 be wholly left to the dulling influence of scour. The "rock-basin" 

 lakes of the bare floors toward the head have received more attention 

 than the great bowl of the amphitheater itself, the most phenomenal 

 of all constantly recurring mountain forms; but I suspect that they 

 are even more significant than has been supposed. 



The hypothesis as to the action at the bottom of the bergschrund, 

 in explanation of sapping, is slenderly supported. Physiographic 

 inquiry has been an avocation merely, and I have failed of oppor- 

 tunity to repeat and to extend my early observations. Those obser- 

 vations, furthermore, -were hastily and somewhat carelessly made, 

 and were not recorded at the time. But that deep basal sapping, in 

 massive rocks especially, as in the summit region of the High Sierra, 

 accounts for the anomalies of towering upland remnants, of canyons 

 gradeless for miles, and of the sharp scalloping in ground-plan 

 everywhere to be observed, is at once apparent, I think, upon full 

 recognition of the forms themselves. 



With these destructional effects assigned to glacial agency, a 

 novel possibility is at once suggested as to the part played in their 

 persistent development by glacial scour, or coarse abrasion. The 

 upright element in the profiles, it would seem, must be regarded as a 

 sapping effect in which scour plays no part at all. But the approxi- 

 mately horizontal element, considering its great extension, often, 

 and the relatively abrupt descent by which compensation is made, 

 constitutes a difficulty no less. The adjusted grade in river erosion is 



