it, and must return for the afternoon milking; and just then, as good
luck would have it, we stumbled upon an immense clump of nettles which
had been one of our landmarks two days before, so that he was no longer
necessary, and we said affectionate adieux.

The glacière was in a state of ruin. Only the right-hand column, not
speaking heraldically, was standing, the others lying in blocks frozen
hard together on the ground. The column which still stood was much
shrunken, and seemed too small for its fissure, the sides of which it
scarcely touched. The wind blew down the entrance slope so
determinedly, that a candle found it difficult to live at the bottom
of the first cave; and a portion of the current blew into the
glacière, and in its sweep exactly struck the fallen columns, the
edges of which were already rounded by thaw. Much of this must be
attributed to the recent opening of the second shaft (p. 5), which
admits a thorough draught through the first cave, and so exposes the
glacière to currents of warmer air; and I should expect to find that
in future the ice will disappear from that part of the cave every
summer, [7] whereas in 1861 we found it thick and dry (excepting a few
small basins containing water) and evidently permanent, in the middle
of a very hot August. The low part of the cave was so completely
protected from the current, that the candle burned there quite
steadily for an hour and a half: still, like the others, the column at
that end of the glacière was broken down, and it therefore became
necessary to attribute its fall to some other agency than the current
of external air. There had been a very large amount of rain, and the
surface of the rock in the fissures was evidently wet; so I have no
doubt that the filtering through of the warm rain-water had thawed the
upper supports of the ice-cascades, and then, owing to their slightly
inclined position, the pedestal had not provided sufficient support,