Again. Insulate, in a horizontal position, a metallic rod with blunt terminations, and about two feet long. We shall designate the ends of this rod by A and B. Let a cork ball electrometer be affixed to the extremity A; then bring an excited glass tube within eight or ten inches of the other end B—the balls will immediately diverge with positive electricity. If the tube be removed the balls will immediately collapse, and no electricity will remain in them, or in the rod.—But if, while the tube is near one end B of the rod, and the balls diverge with positive electricity, the other end A be touched with a finger or other uninsulated conductor, the cork balls will immediately come together, as if the rod were in its natural state: but if, in this state of things, the excited tube be removed, the balls will again diverge, but with negative electricity, shewing that the whole rod AB is now under-charged. This last experiment is thus explained.—When the rod is in its natural state, the electric fluid proper to it is equably distributed throughout the rod; but when the excited glass tube is brought near one of its ends as B, the fluid belonging to that end will be driven towards A; which extremity becomes over-charged, and the other extremity B under-charged; yet the rod has no more electricity now than it had before, and when the tube is removed beyond the sphere of its action, the redundant fluid of A returns to its former place B, and the equilibrium is restored. But if the extremity A be touched, while it is over-charged, by a conductor, this will carry off its superfluous fluid, and leave the extremity A in its natural state, the extremity B being at the same time negatively electrified: and when the tube is removed, part of the fluid naturally belonging to A goes towards B, and the whole rod remains under-charged. CHAP. XII. _Of electrometers._ We have already seen that it is a general law of electricity, that similar electricities repel, and that dissimilar electricities attract each other.—On this law all electrometers are constructed. In fact the cork balls, which have been mentioned are electrometers, and exhibit at once the most important phenomena for the explanation or ascertaining of which the instruments which bear this name are constructed. Still it is of use to see the application which may be made of this general principle. It is applied to ascertain the quantity of the electric fluid collected either in a prime conductor or a coated jar; and also the state of the atmosphere in regard to electricity, and the character of that electricity at any particular time and place. The instruments by which these purposes are effected we shall now shortly describe.