of their appearance which I hold to be quite untenable.
Indeed, to consider these successive lines of dirt on the glacier as
limited only to its surface, and to explain them from that point of
view, is much as if a geologist were to consider the lines presented by
the strata on a cut through a sedimentary mass of rock as representing
their whole extent, and to explain them as a superficial deposit due to
external causes.

A few more details may help to make this statement clearer to my
readers. Let us imagine that a fresh layer of snow has fallen in these
mountain-regions, and that a deposit of dirt has been scattered over its
surface, which, if any moisture arises from the melting of the snow or
from the falling of rain or mist, will become more closely compacted
with it. The next snow-storm deposits a fresh bed of snow, separated
from the one below it by the sheet of dust just described, and this bed
may, in its turn, receive a like deposit. For greater ease and
simplicity of explanation, I speak here as if each successive snow-layer
were thus indicated; of course this is not literally true, because
snow-storms in the winter may follow each other so fast that there is no
time for such a collection of foreign materials upon each newly formed
surface. But whenever such a fresh snow-bed, or accumulation of beds,
remains with its surface exposed for some time, such a deposit of dirt
will inevitably be found upon it. This process may go on till we have a
number of successive snow-layers divided from each other by thin sheets
of dust. Of course, such seams, marking the stratification of snow, are
as permanent and indelible as the seams of coarser materials alternating
with the finest mud in a sedimentary rock.

The gradual progress of a glacier, which, though more rapid in summer
than in winter, is never intermitted, must, of course, change the
relation of these beds to each other. Their lower edge is annually cut
off at a certain level, because the snow deposited every winter melts
with the coming summer, up to a certain line, determined by the local
climate of the place. But although the snow does not melt above this
line, we have seen, in the preceding article, that it is prevented from
accumulating indefinitely in the higher regions by its own tendency to
move down to the lower valleys, and crowding itself between their walls,
thus to force its way toward the outlet below. Now, as this movement is
very gradual, it is