and a high pyramid of snow reaching up towards the uncovered hole
already spoken of. The atmosphere of the cave is damp, and this causes
the ladders to fall speedily to decay, so that they are by no means to
be trusted: indeed, an early round gave way under one of my sisters,
when they visited the cave with me in 1861, and suggested a clear fall
of 60 feet on to a cascade of ice.[16] There are three ladders, one
below the other, and a hasty measurement gave their lengths as 20, 16,
and 28 feet. The rock-roof is only a few feet thick in the neighbourhood
of the hole of entrance.

[Illustration: ENTRANCE TO THE GLACIÈRE OF S. GEORGES.]

The total length of the cave is 110 feet, lying NE. and SW., in the line
of the main chain of the Jura. The lowest part of the floor is a sea of
ice of unknown depth, 45 feet long by 15 broad; and Renaud tried my
powers of belief by asserting that in 1834 the level of this floor was
higher by half the height of the cave than now; a statement, however,
which is fully borne out by Professor Pictet's measurements in 1822,
when the depth of the glacière was less than 30 feet. Indeed, the floor
had sunk considerably since my previous visit, when it was all at the
same level down to the further end of the cave; whereas now, as will be
seen in the section, there was a platform of stones resting on ice at
that end. There are two large fissures passing into the rock, one only
of which can be represented in the section, and these were full of white
ice, not owing its whiteness apparently to the admixture of air in
bubbles, but firm and compact, and very hard, almost like porcelain.
Small stalactites hung from round fissures in the roof, formed of the
same sort of ice, and broken off short, much as the end of a leaden pipe
is sometimes seen to project from a wall. With this exception, there was
no ice hanging from the roof, though there were abundant signs of very