If we were to imagine a set of intelligent creatures, who were able to command only a range of temperatures from 50° to 200° F., engaged upon an investigation of the properties of water, we shall easily understand how unsafe it may be to extend generalisations far beyond the limits covered by actual experiment. Such beings, from their observation of the regular changes of volume of water at all the temperatures they could command, might infer that at still higher and lower temperatures the same rates of expansion and contraction would be maintained. Yet, as we well know, such an inference would be quite wide of the truth; for a little above 200° F. water suddenly expands to 1,700 times its volume, and not far below 50° F. the contraction is suddenly changed for expansion. It has been argued by the late Mr. David Forbes and others that, inasmuch as experiment has shown that--though the fusing points of solids are raised by pressure, yet that this rise of the fusing points goes on in a diminishing ratio as compared with the pressures applied--a limit will probably be reached at which the most intense pressure will not be sufficient to retain substances at a high temperature in their solid state. The fact that gases cannot be retained in a liquid condition by the most intense pressure at a temperature above their critical point, may seem by analogy to favour the same conclusion. Hence, David Forbes, Dana, and other authors, have argued in favour of the existence of a great liquid nucleus in our globe covered by a comparatively thin, solid crust. And if we accept the supposed proofs of a constant increase of temperature from the surface to the centre of the globe, such a conclusion appears to be at least as well founded as that which regards the central masses of the earth as maintained in a solid condition by intense pressure. A little consideration will, however, convince us that the facts which have been relied upon as proving the intensely heated condition of the central masses of our globe, are by no means so conclusive as has been supposed. The earth's form, which mathematicians have shown to be exactly that which would be acquired by a globe composed of yielding materials rotating on its axis at the rate which our planet does, has often been adduced as proving that the latter was not always in a rigid and unyielding condition. In the same way, all the remarkable facts and relations of the bodies of the solar system, which have been shown by astronomers to lend such support to the nebular hypothesis, have been thought, at the same time, to favour the view that our earth is still in a condition of uncompleted solidification.