In saline combinations, and also in many organic forms, we must regard the water as condensed to the solid form; that is, to exist as ice. We well know that, by the abstraction of heat, this condition is produced; but, in chemical combinations, this change must be the result of the mechanical force exerted by the power of the agency directing affinity. In the case of water passing from a liquid to a solid state, we have a most beautiful exemplification of the perfection of natural operations. Water conducts heat downwards but very slowly; a mass of ice will remain undissolved but a few inches under water, on the surface of which, ether, or any other inflammable body, is burning. If ice (solid water) swam beneath the surface, the summer sun would scarcely have power to thaw it; and thus our lakes and seas would be gradually converted into solid masses at our ordinary winter temperatures. All similar bodies contract equally during the process of cooling, from the highest to the lowest points to which the experiments have been carried. It has been thought that if this applied to water, the result would be the sudden consolidation of the whole mass. A modification of the law has been supposed to take place to suit the peculiar circumstances of water. Nature never modifies a law for a particular purpose; we must, therefore, seek to explain the action of the formation of ice, as we know it, by some more rational view. Water expands by heat, and contracts by cold; consequently, the coldest portions of this body occupy the lower portions of the fluid; but it must be remembered that these parts are warmed by the earth. Ross, however, states that at the depth of 1,000 fathoms the sea has a constant temperature of 39°. Water is said to be at its point of greatest density at 40° of Fahrenheit’s thermometer; in cooling further, this fluid appears to expand, in the same way as if heated: and, consequently, water colder than this point, instead of being heavier, is lighter, and floats on the surface of the warmer fluid. It does not seem that any modification of the law is required to account for this phenomenon. Water cooled to 40° still retains its peculiar corpuscular arrangement; but immediately it passes below that temperature, it begins to dispose itself in such a manner that visible crystals may form the moment it reaches 32°. Now, if we conceive the particles of water, at 39°, to arrange themselves in the manner necessary for the assumption of the solid form, by the particular grouping of molecules in an angular instead of a spheroidal shape, it will be clear, from what we know of the arrangement of crystals of water--ice--that they must occupy a larger space than when the particles are disposed, side by side, in minute spheres. Even the escape of air from the water in which it is dissolved is sufficient to give an apparent lightness to the colder water. This expansion still goes on increasing, from the same cause, during the formation of ice,