directions. All these molecules are constantly changing the direction of each other's motion; they are flying about with very different velocities, although, as I have said, their mean velocity is about seventeen miles a minute. If the velocities were all marked off on a scale, they would be found distributed about the mean velocity just as shots are distributed about a mark. If a great many shots are fired at a target, the hits will be found thickest at the bull's-eye, and they will gradually diminish as we go away from that, according to a certain law, which is called the law of error. It was first stated clearly by Laplace; and it is one of the most remarkable consequences of this theory that the molecules of a gas have their velocities distributed among them precisely according to this law of error. In the case of a liquid, it is believed that the state of things is quite different. We said that in the gas these molecules are moved in straight lines, and that it is only during a small portion of their motion that they are deflected by other molecules; but in a liquid we may say that the molecules go about as if they were dancing the grand chain in the Lancers. Every molecule after parting company with one finds another, and so is constantly going about in a curved path, and never gets quite clear away from the sphere of action of the surrounding molecules. But, notwithstanding that, all molecules in a liquid are constantly changing their places, and it is for that reason that diffusion takes place in the liquid. Take a large tank of water and drop a little iodine into it, and you will find after a certain time all the water turned slightly blue. That is because all the iodine-molecules have changed like the others and spread themselves over the whole of the tank. Because, however, you cannot see that, except where you use different colors, you must not suppose that it does not take place where the colors are the same. In every liquid all the molecules are running about and continually changing and mixing themselves up in fresh forms. In the case of a solid quite a different thing takes place. In a solid every molecule has a place which it keeps; that is to say, it is not at rest any more than a molecule of a liquid or a gas, but it has a certain mean position which it is always vibrating about and keeping fairly near to, and it is kept from losing that position by the action of the surrounding molecules. These are the main points of the theory of the constitution of matter as at present believed. It differs from the theory of Democritus in this way. There is no doubt that in the first origin of it, when it was suggested as a whole, it was a guess of his.
In order to make out that your supposition is true, it is necessary to show, not merely that that particular supposition will explain the facts, but also that no other one will. Now, by the efforts of Clarges and Prof. Clerk Maxwell, the molecular theory of matter has been put in that other position, namely, instead now of saying, "Let us suppose that such and such things are true," and then deducing
