If the glass rod is brought near a mass of metal, which we will call a conductor, the following effect is observed:—The part of the conductor near the glass rod becomes charged with electricity of the silk kind; the part of the conductor away from the rod becomes charged with electricity of the glass kind. Now let us bring into play the supposition which we made before. Let us suppose that there is some process in nature which, when there is a twist, makes a real image of that twist come into being. If we assume this process, we see that, opposite to the silk, on whatever objects are near it, will be a twist of the glass kind, and opposite the rod will be a twist of the silk kind. That is to say, that on the conductor there will be a twist of the silk kind when the glass rod is brought near it. But there is nothing to make the particles of the conductor twist as a whole. The glass rod is not supposed to touch the conductor—it is simply brought near it, and no actual communication takes place between them. No force is actually applied to it, nor electricity communicated to it. Hence, on the whole, the particles of the conductor will not be twisted. That is to say, since there is a twist of the glass kind on one end, there will be at the other end a twist of the image kind—that is, of the silk kind. And these two twists are like the two twists on the pencil—if allowed to run together they will run each other out. So if the conductor were removed from the neighbourhood of the glass rod it would be found to be no different from what it was at first. It would not be “charged.” Now this is what actually happens. Thus we have, firstly, a glass rod with its twist; secondly, a mass of metal with two twists on it—one near the glass rod, and of the image kind; the other at the other end of the mass of metal, and related to the original twist in the following way. It is the image of its image. Now it will be found by using a mirror that the image of the image of a twist is the twist itself. Hence on the other end of the conductor there is a twist of the same kind as that on the glass rod. And it is obvious that the rod, with its twist, is connected with the twist on the conductor nearest to it, and the twist on the other end of the conductor will, by the same arbitrary process which we have assumed as real, call up a twist, the image of itself, on any object near it. If it is touched by the object with this contrary twist the two will run together, and the conductor will, if it is left free, have only one twist—the silk kind, which has come up opposite the rod. If now the rod be moved away, the conductor will be twisted as a whole; that is, all the particles of which it is composed will be slightly twisted with a twist of the silk kind. In this state it is said to be charged.