oration and melting, amounts to about nine or ten
feet in a year. At this rate of diminution, a glacier, even one thousand
feet in thickness, could not advance during a single century without
being exhausted. The water supplied by infiltration no doubt repairs the
loss to a great degree. Indeed, the lower part of the glacier must be
chiefly maintained from this source, since the annual increase from the
fresh accumulations of snow is felt only above the snow-line, below
which the yearly snow melts away and disappears. In a complete theory
of the glaciers, the effect of so great an accession of plastic material
cannot be overlooked.

I now come to some points in the structure of the glacier, the
consideration of which is likely to have a decided influence in settling
the conflicting views respecting their motion. The experiments of
Faraday concerning regelation, and the application of the facts made
known by the great English physicist to the theory of the glaciers, as
first presented by Dr. Tyndall in his admirable work, show that
fragments of ice with most surfaces are readily reunited under pressure
into a solid mass. It follows from these experiments, that glacier-ice,
at a temperature of 32 deg. Fahrenheit, may change its form and preserve its
continuity during its motion, in virtue of the pressure to which it is
subjected. The statement is, that, when two pieces of ice with moistened
surfaces are placed in contact, they become cemented together by the
freezing of a film of water between them, while, when the ice is below
32 deg. Fahrenheit, and therefore _dry_, no effect of the kind can be
produced. The freezing was also found to take place under water; and the
result was the same, even when the water into which the ice was plunged
was as hot as the hand can bear.

The fact that ice becomes cemented under these circumstances is fully
established, and my own experiments have confirmed it to the fullest
extent. I question, however, the statement, that regelation takes place
_by the freezing of a film of water between the fragments_. I never have
been able to detect any indication of the presence of such a film, and
am, therefore, inclined to consider this result as akin to what takes
place when fragments of moist clay or marl are pressed together and thus
reunited. When examining beds of clay and marl, or even of compact
limestone, especially in large mountain-masses, I have frequently
observed that the rock present