[Illustration: Fig. 224. A, B, C, D, E, F. Various tubes of different kinds of glass, and containing gases and vapours. Each tube has a platinum wire inserted at both ends, with which the contact is made with the coil. The tube A contains mercury, which has been boiled in it, and the air expelled. By moving the conducting wire to G or H, the light which otherwise passes through the whole of the tubes stops at these points.] _Third Experiment._ When two very thin iron wires are arranged in the upright pillars (Fig. 223), and held sufficiently close to each other, as in Fig. 225, light passes from one to the other. The wire from which the light passes remains _cold_, the other becomes so _hot_ that it melts into a little globule of liquid iron, and if paper is held between the wires it rapidly takes fire. (Fig. 225.) [Illustration: Fig. 225. Melting of the iron wire.] [Page 235] _Fourth Experiment._ Remove the break. Attach two wires to X X (Fig. 226). Hold them so as at pleasure to complete and interrupt the galvanic circle. Two other wires are attached at P P, their ends being about three-quarters of an inch asunder. When the current is closed or broken at A A, a spark passes between B B. (Fig. 226.) [Illustration: Fig. 226. The making and breaking of the circuit.] _Fifth Experiment._ A Leyden jar may be charged and discharged with singular rapidity when connected with the coil, and the snapping noise is so rapid, that it produces a continuous sharp sound. (Fig. 227.) If a piece of paper is held between the ball of the Leyden jar and the wire, it is instantly perforated, but not set on fire. [Illustration: Fig. 227. A B. Leyden jar coated with tinfoil, and standing on any non-conductor, such as gutta percha or the resinous or glass plate, C.] [Page 236] _Sixth Experiment._ When the Leyden jar is coated with spangles of tinfoil, a spark appears at each break, and the whole jar is lit up with hundreds of brilliant sparks each time it is charged and discharged, and as this occurs with amazing rapidity, the light is almost continuous. (No. 1. Fig. 228.) The larger the Leyden jar, the shorter the spark, and _vice versâ_. By the employment of a nicely-made screw and inch-scale, the distance between the discharging points connected with a Leyden jar can be accurately determined; and Mr. Hearder states that supposing a Leyden jar has one square foot of charging surface, it will give a spark of one inch in length, but if a smaller jar is used, with only half a square foot of charging surface, the spark would be about one inch and a quarter in length. (Fig. 228.) [Illustration: Fig. 228.--No. 1. Spangled Leyden jar. No. 2. Hoarder's apparatus for measuring the length of spark for Leyden jar and coil. P P. Glass pillars. No. 3. Two best forms of spangles to paste on a Leyden jar.] _Seventh Experiment._