We may here remark that any arrangement of two metallic plates with a sheet of insulator or non-conductor between them is called a _condenser_. Thus, a condenser can be built up by coating a sheeting of glass on its two sides with tinfoil, or in place of glass we may use mica, paraffin paper, or any other good non-conductor. We may even use air at ordinary pressures; and thus, if two metal plates are placed near to one another in air, the plates being both insulated—that is, supported on non-conductors,—this arrangement constitutes what is called an air condenser. An air condenser, therefore, is virtually only a kind of Leyden jar in which the glass is replaced by air, and the tinfoil by two stout metal plates. [Illustration: FIG. 69.—Hertz oscillator.] I must now proceed to describe and show you a particular kind of air condenser which was invented by the late Professor Hertz, and, in consequence, is called a Hertz oscillator (see Fig. 69). It consists of two square or round metal plates which are carried on glass or ebonite legs, and these plates have short, stout wires attached to them, ending in brass knobs. If these plates are placed in line with one another, they constitute an air condenser of a very peculiar kind, the two brass plates correspond with the tinfoil surfaces of a Leyden jar, and the air all round them corresponds with the glass of the jar. Supposing the plates are so arranged that the brass knobs are about ¹⁄₄ inch apart, or rather less, if then we connect these two brass plates to the secondary terminals of an induction coil or electrical machine capable of giving long sparks in air, we shall find, when the electrical machine or induction coil is set in action, that a very bright crackling spark passes between these little knobs, and with proper experience it is easy to adjust the distance from the knobs so that this spark is an oscillatory spark. Under these circumstances, what is taking place is as follows: In the first place, an electromotive force is acting between the two plates, and creating an electric strain in the air all round them along certain lines, and also between the two knobs. The air, and all other gases like it, possess this peculiar property, that whilst at ordinary pressures they are nearly perfect non-conductors, yet, nevertheless, if they are subjected to more than a certain electric pressure, they pass instantly into a condition in which they become very good conductors. Accordingly, if we progressively increase the electromotive force acting between the plates, up to a certain point the whole arrangement acts like a Leyden jar; but there comes a moment when the air between the knobs breaks down and passes from a non-conductive to a conductive condition. The two plates then resemble at that moment the surfaces of a charged Leyden jar which are connected together by a good conductor, and, as we have already seen, under those circumstances the discharge