+The Spark Discharge.+--In practice the wire _w_ is not actually used, but the air itself suddenly becomes a conductor and makes the connection. When the electric field at a point in the air exceeds a certain limiting strength, the air seems to break down and {85} suddenly become a conductor and remains one for a short time. This breaking down is accompanied by light and heat, and is known as the spark discharge or electric spark. +Experiments of Hertz.+--In the brilliant experiments carried out by Hertz at Karlsruhe between 1886 and 1891, he not only demonstrated the existence of the waves produced in this way, but he showed that they are reflected and refracted like ordinary light, he measured their wave-length and roughly measured their speed, this latter being equal to the speed of light within the errors of experiment. [Illustration: FIG. 30.] One arrangement used by Hertz is shown in plan in Fig. 30. A Ruhmkorff coil R serves to charge the two conductors A and B until the air breaks down at the gap G, and a spark passes. Before the spark is {86} produced, the lines of force on the lower side of AB will in form be something like the dotted lines in the figure, but as soon as the air becomes a conductor, the positive ends of the lines will surge from A towards B and on to B, and the negative ends will surge on to A. These to and fro surgings will continue for a little while, but will gradually die out. As the surgings are all up and down AB, the electric vibrations in the electromagnetic waves sent out {87} will all be parallel to AB, and therefore they will be polarised. [Illustration: FIG. 31.] This is characteristic of all electric waves, as no single sparking apparatus will produce anything but waves parallel to the spark gap. The electric vibrations coming up to a conductor placed in the position of the wire rectangle, M, will cause surging of the lines along it, and, if these surgings are powerful enough, will cause a spark to pass across the small gap S. Such a rectangle was therefore used by Hertz as a detector of the waves, but since that time many detectors of very much greater sensitiveness have been devised. +Reflection.+--In order to show that these waves are reflected in the same way as light waves, Hertz placed the sparking knobs, G, at the focus of a large parabolic metallic reflector, and his detector, D, at the focus of a similar reflector placed as in Fig. 31, but much farther away (cf. Fig. 1). In this position sparking at G produced strong sparking in the detector, although the distance was such that no sparking was produced without the reflectors. +Refraction.+--The refraction of the waves was {88} shown by means of a large prism made of pitch. This had an angle of 30° and was about 1.5 metres high and 1.2 metres broad. [Illustration: FIG. 32.]