The electrical capacity of even a small Leyden jar is surprisingly great, and this is due to the mutual attraction between opposite kinds of electricity. If we stick a piece of tinfoil on the centre of each face of a pane of glass, and charge one positively and the other negatively, the two charges attract each other through the glass; and in fact they hold on to each other so strongly that we can get very little electricity by touching either piece of foil. This mutual attraction enables us to charge the two pieces of foil much more strongly than if they were each on a separate pane, and this is the secret of the working of the Leyden jar. If the knob of the jar is held to the positive ball of a Wimshurst, the inside coating receives a positive charge, which acts inductively on the outside coating, attracting a negative charge to the inner face of the latter, and repelling a positive charge to its outer face, and thence away through the hand. The electricity is entirely confined to the sides of the jar, the interior having no charge whatever. Leyden jars are very often fitted to a Wimshurst machine as shown at A, A, Fig. 4, and arranged so that they can be connected or disconnected to the collecting combs as desired. When the jars are disconnected the machine gives a rapid succession of thin sparks, but when the jars are connected to the combs they accumulate a number of charges before the discharge takes place, with the result that the sparks are thicker, but occur at less frequent intervals. It will have been noticed that the rod of a Leyden jar and the discharging rods of a Wimshurst machine are made to terminate not in points, but in rounded knobs or balls. The reason of this is that electricity rapidly leaks away from points. If we electrify a conductor shaped like a cone with a sharp point, the density of the electricity is greatest at that point, and when it becomes sufficiently great the particles of air near the point become electrified and repelled. Other particles take their place, and are electrified and repelled in the same way, and so a constant loss of electricity takes place. This may be shown in an interesting way by fastening with wax a needle to the knob of a Wimshurst. If a lighted taper is held to the point of the needle while the machine is in action, the flame is blown aside by the streams of repelled air, which form a sort of electric wind. CHAPTER III ELECTRICITY IN THE ATMOSPHERE