The literature of glaciation has not escaped the blemish of too free generalization. It was early asserted of the Sierra Nevada, for example, that Pleistocene glaciers of the alpine type, descending from an icecap in the summit region of the range, had reached to the range foot, and that the abnormally large canyons, particularly of the western flank, were the products, from head to foot, of glacial erosion. Such a statement made of southeastern Alaska, of the Scandinavian peninsula, or of the Patagonian Andes would not on the face of it be absurd. Nor was it absurd of the Sierra Nevada. It was possible, despite the low latitude, that ice-streams should have descended to the range foot, and it was theoretically not impossible that they should have excavated deep canyons. The matter, especially of glacier efficiency in erosion — a vexed question — is one mainly of the evidences. It cannot safely be handled deductively; and it need not be, since in glaciated mountains the evidences crowd the field. But the announcement was unscientific, because unsupported by facts of observation. Its author had no right to make it. On the other hand, it was no less unwarrantable and dogmatic to assert the contrary, which also was freely done.