On being connected, every two conductors assume at once the same potential. With this the means is given of determining the potential of a conductor through the agency of a second conductor expressly adapted to the purpose called an electrometer, just as we determine the temperature of a body with a thermometer. The values of the potentials of bodies obtained in this way simplify vastly our analysis of their electrical behavior, as will be evident from what has been said. Think of a positively charged conductor. Double all the electrical forces exerted by this conductor on a point charged with unit quantity, that is, double the quantity at each point, or what is the same thing, double the total charge. Plainly, equilibrium still subsists. But carry, now, the positive electrostatic unit towards the conductor. Everywhere we shall have to overcome double the force of repulsion we did before, everywhere we shall have to expend double the work. By doubling the charge of the conductor a double potential has been produced. Charge and potential go hand in hand, are proportional. Consequently, calling the total quantity of electricity of a conductor _Q_ and its potential _V_, we can write: _Q = CV_, where _C_ stands for a constant, the import of which will be understood simply from noting that _C = Q/V_.[32] But the division of a number representing the units of quantity of a conductor by the number representing its units of potential tells us the quantity which falls to the share of the unit of potential. Now the number _C_ here we call the capacity of a conductor, and have substituted, thus, in the place of the old relative determination of capacity, an absolute determination.[33] In simple cases the connexion between charge, potential, and capacity is easily ascertained. Our conductor, let us say, is a sphere of radius _r_, suspended free in a large body of air. There being no other conductors in the vicinity, the charge _q_ will then distribute itself uniformly upon the surface of the sphere, and simple geometrical considerations yield for its potential the expression _V = q/r_. Hence, _q/V = r_; that is, the capacity of a sphere is measured by its radius, and in the C. G. S. system in centimetres.[34] It is clear also, since a potential is a quantity divided by a length, that a quantity divided by a potential must be a length. Imagine (Fig. 36) a jar composed of two concentric conductive spherical shells of the radii _r_ and _r₁_, having only air between them. Connecting the outside sphere with the earth, and charging the inside sphere by means of a thin, insulated wire passing through the first, with the quantity _Q_, we shall have _V = (r₁-r)/(r₁r)Q_, and for the capacity in this case _(r₁r)/(r₁-r)_, or, to take a specific example, if _r = 16_ and _r₁ = 19_, a capacity of about 100 centimetres. [Illustration: Fig. 36.]