on the glacier, but
the most numerous and dangerous are the transverse and lateral ones. The
transverse ones were readily accounted for after the motion of the
glacier was admitted; they must take place, whenever, the glacier
advancing over inequalities or steeper parts of its bed, the tension of
the mass was so great that the cohesion of the particles was overcome,
and the ice consequently rent apart. This would be especially the case
wherever some steep angle in the bottom over which it moved presented an
obstacle to the even advance of the mass. But the position of the
lateral ones was not so easily understood. They are especially apt to
occur wherever a promontory of rock juts out into the glacier; and when
fresh, they usually slant obliquely upward, trending from the prominent
wall toward the head of the glacier, while, when old, on the contrary,
they turn downward, so that the crevasses around such a promontory are
often arranged in the shape of a spread fan, diverging from it in
different directions. When the movement of the glacier was fully
understood, however, it became evident, that, in its effort to force
itself around the promontory, the ice was violently torn apart, and that
the rent must take place in a direction at right angles with that in
which the mass was moving. If the mass be moving inward and downward,
the direction of the rent must be obliquely upward. As now the mass
continues to advance, the crevasses must advance with it; and as it
moves more rapidly toward the middle than on the margins, that end of
the crevasse which is farthest removed from the projecting rock must
move more rapidly also; the consequence is, that all the older lateral
crevasses, after a certain time, point downward, while the fresh ones
point upward.

Not only does the glacier collect a variety of foreign materials on its
upper surface, but its sides as well as its lower surface are studded
with boulders, stones, pebbles, sand, coarse and fine gravel, so that it
forms in reality a gigantic rasp, with sides hundreds of feet deep, and
a surface thousands of feet wide and many miles in length, grinding over
the bottom and along the walls between which it moves, polishing,
grooving, and scratching them as it passes onward. One who is familiar
with the track of this mighty engine will recognize at once where the
large boulders have hollowed out their deeper furrows, where small
pebbles have drawn their finer marks, where the sto