It may seem strange that, if this is so, they should become electrified when rubbed together. Yet this is the case, and must be taken as a fact. It seems to depend partly on the circumstance that glass and silk are not what is called conductors. In a conductor, if one part receives electricity, this electricity at once runs over the whole of the conductor, whether it be an inch long or many feet. And if any part of the conductor be touched by another conductor which is in contact with the earth, every trace of electricity leaves the conductor, flowing, as it were, freely out of it.

Now both glass and silk do not let the electricity run from them so easily. To discharge a glass rod it has practically to be touched in every part. Thus, when by the rubbing with silk electricity is produced on it, it is conceivable that this electricity should be kept to a certain extent, and not combined immediately with the electricity on the silk.

Besides, the same cause—the friction which produced the electricity on the glass, and the other kind of electricity on the silk—would probably prevent their combination as long as it was applied.

Now let us suppose that the electrical charge which the glass has consists in this.

Let us suppose that the particles of the glass on the surface of it are twisted, strained out of their natural position, and twisted.

Let us also suppose that the particles of the silk are twisted too, but let them have the image twist.

Now these two twists, the glass twist and the silk twist, its image, when brought together, will run down. In unwinding each other they will give off a certain amount of energy, which will manifest itself as a spark, make a crackling sound, and so on. But when they unwind each other there is no more tension of the particles.

This does not explain in the least why the glass particles should receive a twist in one direction, the silk particles a twist of the image kind.

But instead of inquiring into this, it is best to see if this supposition is in accordance with other known facts of electrical action; because, if it is not, we may dismiss it more easily than we could if we had to test it with regard to the very inaccessible question of why some bodies, when rubbed, get electrified in one way, others in another.

When the glass and silk are near together the twist on the glass and that on the silk are related to each other as twist and image twist, and there is no action on surrounding bodies from either of them, as they, so to speak, satisfy each other.