If user and manufacturer would concentrate upon obtaining purity, if they would try the effects of damping upon loud speakers, which are easily obtained; if they would realise that the horn of the loud speaker should be without resonance, that it should be also damped and pocketed, that its goose-necked shape is not adopted without an object, and if they would aim at the delivery of true music instead of noise--then, we should make a great advance. The average loud speaker can often be greatly improved by padding the horn with some kind of tape, and as an example of the great difficulties of proper transmission let it be made quite clear that with most cases of wireless communication the sending is nearly perfect. Reception is greatly at fault: it is the reception that mangles the sound and makes it too often almost unbearable to anybody of reasonably sensitive hearing. At some large transmission stations it was at one time quite common to use three separate microphones for the modulation. One received notes of high pitch, one of low, and a third attempted to obtain the “S” sound with the result that, when this “S” microphone was adjusted for a man who did not say his “S” very loudly and someone appeared who did, it sounded exactly as if the speaker had dropped his false teeth. All this is now avoided. The ordinary diaphragm is no longer in use, but a very small coil of aluminium wire is suspended between the poles of an electro magnet, allowed to rest against an ordinary pad of cotton wool, and that is all! The infinitesimal movements of this aluminium coil will reproduce speech up to about 40,000 periods per second in oscillatory speed, yet speech is well recognisable if all frequencies over 4,000 per second are gridded out. How difficult it is going to be to make a large, heavy, and rapidly moving diaphragm reproduce accurately when we have had to take all these precautions to obtain accuracy of transmission! It is not impossible; it will come one day. Now let us see what is the result of our sound troubles. We are told that before long it will be quite easy to hear birds singing in trees and the waves beating against the seashore. Quite right, quite easy to do it now, but if a bird singing in a tree sounds like a man moving his condenser or walking about with a pair of squeaky boots, is it progress? [Illustration: Fig. 2. If light falls on a very thin diaphragm carrying a mirror, the sound waves being directed into it by a trumpet can be photographed upon a revolving sensitised film.] Perhaps it is. The whole point of wireless is that it brings a man into your room, but it must sound like the man himself if it is to be really effective; it is this pitiful quality of reproduced sound that has wrecked the talking cinema.