If you recall the remarks made just now in connection with the special electrical properties of air and other gases, you will notice that there is a remarkable similarity between the electrical behaviour of air at ordinary pressures to electromotive force, and that of a loose mass of metallic filings. Both the air and the metallic filings are non-conductors as long as the electromotive force acting on them does not exceed a certain value, but if it exceeds this critical value, they pass at once into a conductive condition. The fact that pieces of metal in loose contact with one another behave in a similar manner was discovered more than twenty years ago by the late Professor D. E. Hughes, who, as you may perhaps know, was the inventor of a printing telegraph, the microphone, and many other most important electrical instruments. Professor Hughes was a great genius, and in many respects in advance of his age. He it was who undoubtedly discovered that an electric spark has the power of affecting at a distance the electric conductivity of a metallic junction consisting of two metals in loose contact. The peculiar behaviour of metallic filings under electromotive force, and under the influence of electric sparks at a distance, was subsequently rediscovered by Professor Branly; and the effect of an electric oscillatory spark in changing the conductivity of a light metal contact was also rediscovered by Sir Oliver Lodge, and the phenomena investigated by many other observers. I can show you the experiment on a large scale in the following manner:— [Illustration: FIG. 70.—A metal disc-coherer.] I have here a number of aluminium discs, the size of sixpences, stamped out of thin metal, and these are arranged in a sort of semi-cylindrical trough between two terminal screws, so that the discs are very lightly pressed together. Under these circumstances the pile of metal discs is not a conductor, and it will not pass the electric current from a battery which is joined up in series with an electric bell and the pile of discs (see Fig. 70). Supposing, however, that I make an oscillatory spark in proximity to this pile of metal discs, as I can do by taking the discharge from a large Leyden jar near it; the pile of discs at once becomes a conductor; the electric current from the battery can then pass through it, and the bell rings. Such an arrangement has been named by Sir Oliver Lodge a _coherer_, because, under the action of the oscillatory spark, the discs cohere or stick together. We can separate the discs by giving them a sharp rap, and then the operation can be again repeated.