Very nice photographic experiments do, however, show that vibrations are produced on the diaphragm of the receiving telephone; they are indeed excessively slight, but Mr. Blake asserts that they are enough to cause a very light index, resting on the diaphragm, to make slight inflections on a line which it describes on a register. Yet this small vibration of the diaphragm does not show that it is due to the effect of attraction, for it may result from the act of magnetisation itself in the centre of the diaphragm.[11] An interesting experiment by Mr. Hughes, repeated under different conditions by Mr. Millar, confirms this opinion. If the magnet of a receiving telephone consists of two magnetised bars, perfectly equal, separated from each other by a magnetic insulator, and they are so placed in the coil as to bring alternately the poles of the same and of contrary signs opposite to the diaphragm, it is known that the telephone will reproduce speech better in the latter case than in the former. Now, if the effects were due to attraction, this would not be the case; for the actions are in disagreement when the poles of contrary signs are subjected to the same electric influence, while they are in agreement when these poles are of like signs. On the other hand, it is known that if several iron plates are put together in order to form the diaphragm of the receiver, the transmission of sounds is much stronger than with a simple diaphragm; and yet the attraction, if it has anything to do with it, could only be exerted on one of the diaphragms. It further appears that it is not merely the magnetic core which emits sounds, but that they are also produced with some distinctness by the helices. Signor Rossetti had already ascertained this fact, and had even remarked that they could be animated by a slight oscillatory movement along the bar magnet, when they were not fixed upon it. Several observers, among others M. Paul Roy, Herr Wiesendanger, and Signor Canestrelli, have since mentioned similar facts, which are really interesting. ‘If,’ writes M. Paul Roy, ‘a coil of fine wire, which is at the extremity of the bar magnet of a Bell telephone, receives the pulsatory currents transmitted by a carbon telephone, it is only necessary to bring the coil close to the ear in order to hear the sounds. ‘The sounds received in this way are very faint, but become much stronger if a piece of iron is introduced into the circuit coil. A magnet acts with still greater force, even when it consists of a simple magnetised needle. Finally, the sound assumes its maximum intensity when an iron disk is inserted between the ear and the coil. ‘By placing the end of the coil to the ear, and sending a current through it from the bar magnet, it is ascertained that the sound is at its minimum when the neutral line of the magnet is enclosed by the coil, and that it increases until attaining its maximum, when the magnet is moved until one of its poles corresponds to the coil.