nes with angular
edges have left their sharp scratches, where sand and gravel have rubbed
and smoothed the rocky surface, and left it bright and polished as if it
came from the hand of the marble-worker. These marks are not to be
mistaken by any one who has carefully observed them; the scratches,
furrows, grooves, are always rectilinear, trending in the direction in
which the glacier is moving, and most distinct on that side of the
surface-inequalities facing the direction of the moving mass, while the
lee-side remains mostly untouched.

It may be asked, how it is known that the glacier carries this powerful
apparatus on its sides and bottom, when they are hidden from sight. I
answer, that we might determine the fact theoretically from certain
known conditions respecting the conformation of the glacier; to which I
shall allude presently; but we need not resort to this kind of evidence,
since we have ocular demonstration of the truth. Here and there on the
sides of the glacier it is possible to penetrate between the walls and
the ice to a great depth, and even to follow such a gap to the very
bottom of the valley, and everywhere do we find the surface of the ice
fretted as I have described it, with stones of every size, from the
pebble to the boulder, and also with sand and gravel of all sorts, from
the coarsest grain to the finest, and these materials, more or less
firmly set in the ice, form the grating surface with which, in its
onward movement down the Alpine valleys, it leaves everywhere
unmistakable, traces of its passage.

We come now to the moraines, those walls of loose materials built by the
glaciers themselves along their road. They have been divided into three
classes, namely, lateral, medial, and terminal moraines. Let us look
first at the lateral ones; and to understand them we must examine the
conformation of the glacier below the _neve_, where it assumes the
character of pure compact ice. We have seen that the fields of snow,
where the glaciers have their origin, are level, and that lower down,
where these masses of snow begin to descend toward the narrower valley,
they follow its trough-like shape, sinking toward the centre and sloping
upward against the sides, so that the surface of the glacier, about the
region of the _neve_, is slightly concave. But lower down in the glacier
proper, where it is completely transformed into ice, its surface becomes
convex, for the following reason: The rocky walls of t