e frost, with the fuze-hole uppermost. A portion of the
water expanded in freezing, so as to protrude a cylinder of ice from the
fuze-hole; and this cylinder continued to grow inch by inch in
proportion as the central nucleus of water froze. As we cannot doubt
that an outer shell of ice is first formed, and then another within, the
continued rise of the column through the fuze-hole must proceed from the
squeezing of successive shells of ice concentrically formed, through the
narrow orifice; and yet the protruded cylinder consisted of entire, and
not fragmentary ice.[290]

The agency of glaciers in producing permanent geological changes
consists partly in their power of transporting gravel, sand, and huge
stones to great distances, and partly in the smoothing, polishing, and
scoring of their rocky channels, and the boundary walls of the valleys
through which they pass. At the foot of every steep cliff or precipice
in high Alpine regions, a talus is seen of rocky fragments detached by
the alternate action of frost and thaw. If these loose masses, instead
of accumulating on a stationary base, happen to fall upon a glacier,
they will move along with it, and, in place of a single heap, they will
form in the course of years a long stream of blocks. If a glacier be 20
miles long, and its annual progression about 500 feet, it will require
about two centuries for a block thus lodged upon its surface to travel
down from the higher to the lower regions, or to the extremity of the
icy mass. This terminal point remains usually unchanged from year to
year, although every part of the ice is in motion, because the
liquefaction by heat is just sufficient to balance the onward movement
of the glacier, which may be compared to an endless file of soldiers,
pouring into a breach, and shot down as fast as they advance.

The stones carried along on the ice are called in Switzerland the
"moraines" of the glacier. There is always one line of blocks on each
side or edge of the icy stream, and often several in the middle, where
they are arranged in long ridges or mounds, often several yards high.
(See fig. 18, p. 223.) The cause of these "medial moraines" was first
explained by Agassiz, who referred them to the confluence of tributary
glaciers.[291] Upon the union of two streams of ice, the right lateral
moraine of one of the streams comes in contact with the left lateral
moraine of the other, and they afterwards move on together, in the
centre,