Page:Whymper - Scrambles amongst the Alps.djvu/191

Rh agency), renders it certain that the entire surface of the bed-rock has been lowered, at the most, but a few yards.

Weathered rocks, upon a small scale, do not take shapes such as are figured in this diagram, but rather such as those which are shown in Fig. 1, We do not find deep pits or troughs produced in rocks (whatever may be their nature or composition) by weathering or through any of the ordinary operations of nature. Still less do we find a large number of such pits or troughs close to one another. Therefore, when we see lee-sides as at D and F, Fig. 3, (separated, perhaps, from each other by a distance of less than a dozen feet; and representing, as it has been already stated, the remains of hollows or fractures which existed before the glacier began to work), it is certain that the eminences B C, between them, have been lowered only a few feet; and probable that the depth of the rock which has been removed does not exceed the length of a line drawn from D to F. The unworn lee-sides to glaciated rocks have, therefore, a special value, as they afford indications (although imperfect ones), of the amount of excavation that has been performed by the glaciers which worked above and around them.

5. In § 6 it was stated that the amount, in depth, of the matter which is removed constantly diminishes, if the power that is employed continues to be the same. That is to say, if a glacier 1000 feet thick, moving down a valley at the rate of 300 feet per annum, is able to remove a depth of one inch from the whole of those portions of the surfaces that it touches, in the course of one year, the amount that it will remove in the course of the next (assuming that the depth of 1000 feet is maintained, and the rate of motion is the same) will not be one inch, but will be something less; because the power employed will be distributed over a greater area. It does not, however, follow that the bulk of the matter which is removed will be less and less from the very beginning.