The fragment of compact ice whose elements assume such beautiful crystalline forms is itself a crystal. This was shown by Sir David Brewster, who employed for the purpose of analysis that modified form of light which we call _polarised light_. It is singularly well adapted to bring out the peculiarities of the main structure of substances, owing to the coloured figures which it outlines on a screen after passing through them. All crystals with an axis--such, for instance, as Iceland spar--yield a series of brilliantly-tinted rings, traversed by a regularly-formed cross entirely black. As ice produces the same figures, we are justified in attributing to it the same kind of crystallization. We must note, however, that we are referring now to the thick ice formed on our canals and lakes. If we examined the first film formed on the surface of the water, we should discover in it a completely irregular crystallization, the ray of polarised light producing only a mosaic of varied tints, distributed without any order. But it is easy to explain the way in which this primary crust or film is produced. Those portions of the fluid mass in contact with the air are the first to freeze, but each molecule of ice abandons its heat to the contiguous water, which thereby is slightly raised in temperature, and the result is a partial congelation. The surface we are examining then presents a network of fine needles intercrossed in every direction, and forming a kind of delicate lace, the meshes or intervals of which are gradually filled up. When the network is transformed into a continuous sheet, the loss of heat is diminished more and more as this external crust grows thicker and thicker; but the development of the ice invariably takes place by means of long interlaced needles, as the reader may see for himself by breaking off a portion from the nearest pond (in winter), and examining the sectional surface. Having said thus much in reference to the crystallization of ice and snow, we proceed to explain the _regelation_ and _moulding of ice_. Some years ago, Faraday astonished the scientific world by a very curious experiment. Splitting into two parts a piece of ice, he brought together the parts at the moment that fusion took place on their surfaces, and they united immediately. How are we to account for this effect, which can be produced even in hot water?