ice or water.  We observed, by the way, that the slope of ice to which the candle descended in the deeper pit, and the shelf on which it rested, were quite dry, or at any rate free from all apparent signs of the abundant water we should have seen, had that been the outlet for the streams which poured into the moulin.  The maire said that the columns and cascades of ice in the cave had been much more beautiful in the previous summer.

The whole cavern would thus appear to be something of the shape of an egg, with the longer axis vertical, and the entrance about half-way up the side.  The lower end of this egg-shaped cavity in the rock is filled with ice, which in some parts shrinks from the rock below the surface, though, as far as outward appearance goes, it fills the cavern to its farthest corners.  The depth of this ice at one side is 60 feet, and how much more it may be in the middle it is impossible to say.  As we have seen, there is a second ice-cave opening out of the principal one, at a depth of 190 feet below the surface; and with respect to this second cave imagination may run riot.  Rosset told me that he had noticed, the year before, a strong source of water springing out of the side of a rock, at some little distance from the glaciere; but he could not reach it then, and could not find it now.  This may possibly be the drainage of the glaciere in its summer state.

The thermometer stood at 34 deg. in the middle of the cave; and though the others felt the cold very much, I was myself surprised to find so low a register, for the atmosphere seemed to be comparatively warm, judging from what I had experienced in other glacieres.  The only current of air we could detect was exceedingly slight, and came from the deeper of the two pits in the ice.  It was so slight, that the flame of the candle burned apparently quite steadily when we were engaged in determining the depth and shape of the pit.

The sun had by this time produced such an effect upon the slope of snow outside the glaciere, that we found the ascent sufficiently difficult, especially as our hands were full of various instruments.  The schoolmaster was not content to choose the straight line up, and in attempting to perform a zigzag, he came to a part of the slope where the snow lay about 2 inches thick on solid ice, and the result was an unscholastic descent in inverted order of precedence.  He got on better over the rolling stones after the snow was accomplished, but the clumsy style of his climbing dislodged an unpleasant amount and weight of missiles; and though he was amiable enough to cry ‘Garde!’ with every step he took, it will be found by experiment that it is not much use to the lower man to have ‘Garde!’ shouted in his ears, when his footing is insecure to begin with, and a large stone comes full at his head, at the precise moment when two others are taking him in the pit of the stomach.