326. You have not forgotten our conversation regarding "atomic poles" (§ 10), and how the notion of polar force came to be applied to crystals. With this fresh in your memory, you will have no great difficulty in understanding how expansion of volume may accompany the act of crystallisation. 327. I place a number of magnets before you. They, as matter, are affected by gravity, and, if perfectly free, they would move towards each other in obedience to the attraction of gravity. 328. But they are not only matter, but _magnetic_ matter. They not only act upon each other by the simple force of gravity, but by the polar force of magnetism. Imagine them placed at a distance from each other, and perfectly free to move. Gravity first makes itself felt and draws them together. For a time the magnetic force issuing from the poles is insensible; but when a certain nearness is attained, the polar force comes into play. The mutually attracting points close up, the mutually repellent points retreat, and it is easy to see that this action may produce an arrangement of the magnets which requires more room. Suppose them surrounded by a box which exactly encloses them at the moment the polar force first comes into play. It is easy to see that in arranging themselves subsequently the repelled corners and ends of the magnets may be caused to press against the sides of the box, and even to burst it, if the forces be sufficiently strong. 329. Here then we have a conception which may be applied to the molecules of water. They, like the magnets, are acted upon by two distinct forces. For a time while the liquid is being cooled they approach each other, in obedience to their general attraction for each other. But at a certain point new forces, some attractive, some repulsive, _emanating from special points_ of the molecules, come into play. The attracted points close up, the repelled points retreat. Thus the molecules turn and rearrange themselves, demanding, as they do so, more space, and overcoming all ordinary resistance by the energy of their demand. This, in general terms, is an explanation of the expansion of water in solidifying: it would be easy to construct an apparatus for its illustration. § 48. _The Dirt Bands of the Mer de Glace._ 330. Pass from bright sunshine into a moderately lighted room; for a time all appears so dark that the objects in the room are not to be clearly distinguished. Hit violently by the waves of light (§ 3) the optic nerve is numbed, and requires time to recover its sensitiveness. 331. It is for this reason that I choose the present hour for a special observation on the Mer de Glace. The sun has sunk behind the ridge of Charmoz, and the surface of the glacier is in sober shade. The main portion of our day's work is finished, but we have still sufficient energy to climb the slopes adjacent to the Montanvert to a height of a thousand feet or thereabouts above the ice. [Illustration]