If it is remembered that the plate A S vibrates under the influence of electro-magnetic attractions more readily in proportion to their approximation in number to the vibrations corresponding to the fundamental sound it can emit, it becomes clear that if this plate is tuned to the same note as that of the corresponding instrument, it will be rendered peculiarly sensitive to the vibrations transmitted by the sender, and the other vibrations which may affect it will only act faintly. Moreover, a resonator placed above the plate will greatly increase this predisposition; so that if several systems of this kind, tuned to different notes, produce simultaneous transmissions, the sounds corresponding to the different vibrations will be in a certain sense selected and distributed, in spite of their combination, into the receivers for which they are specially adapted, and each of them may retain the traces of the sounds emitted by adding the register, which may be so arranged as to act as an ordinary Morse receiver. Mr. Gray states that the number of sending instruments and independent local circuits may be equal to that of the tones and semitones of two or more octaves, provided that each vibrating plate be tuned to a different note of the scale. The instruments may be placed side by side, and their respective local keys, arranged like the keys of a piano, will make it easy to play an air combining notes and chords; there may also be an interval between the instruments, which may be sufficiently far from each other to allow the employés to work without being distracted by sounds not intended for them. In a new arrangement, exhibited at the Paris Exhibition, 1878, Mr. Gray considerably modified the way of working the various electro-magnetic organs which we have just described. In this case, the plates consist of tuning-forks with one branch kept in continual vibration at both stations, and the signals only become perceptible by intensifying the sounds produced. This arrangement follows from the necessity of keeping the line circuit always closed for multiple transmissions of this nature, so as to react with pulsatory currents, which are alone able, as we have already seen, to retain the individual character of several sounds simultaneously transmitted. [Illustration: FIG. 63.]