Thus, everybody invents, and, indeed, is compelled to invent, hypotheses in order to account for phenomena of the cause of which he has no direct evidence; and they are just as legitimate and necessary in science as in common life. Only the scientific reasoner must be careful to remember that which is sometimes forgotten in daily life, that a hypothesis must be regarded as a means and not as an end; that we may cherish it so long as it helps us to explain the order of nature; but that we are bound to throw it away without hesitation as soon as it is shown to be inconsistent with any part of that order. 48. =The Hypothesis that Water is composed of Separate Particles (Molecules).= It has been pointed out that we cannot see, and indeed that there is not much hope of our ever being able to see, the separate particles of water, even if water is composed of such particles. But it is perfectly legitimate to suppose that water is made up of such particles, if that hypothesis will enable us to explain the properties of water. Let us suppose then that any portion of fluid water is really composed of a prodigious number of particles less (and probably much less) than a millionth of an inch in diameter. We may call these particles =molecules=.[4] Footnote 4: Diminutive of _moles_, a mass. We are justified, in accordance with the general properties of matter (§ 18), in supposing that these molecules tend to approach one another. But the fact that water is slightly compressible justifies the supposition that its molecules are not in actual contact, but that they are separated by interspaces, just as the motes in the air of a dusty room are so separated. What is it that keeps the molecules apart? We have seen that great mechanical pressure brings them but slightly nearer to one another; hence there is an equivalent resistance of some kind which keeps them apart. This resistance must have the same origin as the sensation which we know as heat, for it has been seen that diminution of heat diminishes the bulk of water; that is, allows the molecules to come closer together; that is, diminishes their tendency to keep asunder. Increase of heat, on the other hand, increases the volume of water; that is to say, drives the molecules further apart, or increases their tendency to keep asunder. Suppose we call the cause of the tendency of the molecules of water to come together an =attractive force=; and the cause of their keeping apart, which manifests itself to us as the sensation of heat and is, as we have seen, in all probability, a rapid vibratory or whirling motion of the molecules, a =repulsive force=; then, in the liquid state, these forces are so adjusted that the molecules are quite free to move, and yet hold together.