lower extremity proportionate to
their advance. But in considering the past history of glaciers, we must
think of their changes as retrograde, not progressive movements; since,
if the glacial theory be true, a great mass of ice, of which the present
glaciers are but the remnants, formerly spread over the whole Northern
hemisphere, and has gradually disappeared, until now no traces of it are
to be found, except in the Arctic regions and in lofty mountain-ranges.
Every terminal moraine, such as I described in the last article, is the
retreating footprint of some glacier, as it slowly yielded its
possession of the plain, and betook itself to the mountains; wherever we
find one of these ancient semicircular walls of unusual size, there we
may be sure the glacier resolutely set its icy foot, disputing the
ground inch by inch, while heat and cold strove for the mastery. There
may have been a succession of cold summers, or, if now and then a warmer
summer intervened, a colder one followed, so that the glacier regained
the next year the ground it had lost during the preceding one, thus
continuing to oscillate for a number of years along the same line, and
adding constantly to the _debris_ collected at its extremity. Wherever
such oscillations and pauses in the retreat of the glacier occurred, all
the materials annually brought down to its terminus were collected; and
when it finally disappeared from that point, it left a wall to mark its
temporary resting-place.

By these semicircular concentric walls we can trace the retreat of the
ice as it withdrew from the plain of Switzerland to the fastnesses of
the Alps. It paused at Berne, and laid the foundation of the present
city, which is built on an ancient moraine; it made a stand again at the
Lake of Thun, and barred its northern outlet by a wall which holds its
waters back to this day. Other moraines, though less distinct, are
visible nearer the base of the Bernese Alps, and, above Meyringen, the
valley is spanned by one of very large dimensions. Again, on the other
side of the first chain of high peaks, the glacier of the Rhone,
descending the valley toward the Lake of Geneva, has everywhere left
traces of its ancient extension. We find the valley crossed at various
distances by concentric moraines, until we reach the lake. There are no
less than thirteen concentric moraines immediately below the present
termination of the glacier of the Rhone, the one nearest to the ice, and
t