556 
Proceedings of the Royal Society 
This theory gives a vera causa for the action of the instrument in 
its normal form, and probably explains in part the true action 
which takes place, but the author believes that this is not the whole 
of the action. The fact that the inventor was working upon this 
theory when he designed the instrument, does not prove the theory 
to be correct, as is shown by his having at first employed a powerful 
battery with the instrument, believing it to be necessary, and it was 
only after diminishing the number of cells without impairing the 
action of the Telephone that it occurred to him to try it without 
a battery at all, when he found the action to be equally good. 
The author’s mistrust of the theory commenced when he learnt 
that a thick iron plate might be used, whose vibrations must be 
extremely minute. And when he learnt that the instrument might be 
used with plates of glass, wood, tinfoil, or vulcanite, or even with- 
out any plate at all, he was convinced that the theory was imperfect, 
and resolved to propose another theory which seemed more consis- 
tent with facts. 
This theory is divided into two parts — the action at the sending 
end, and that at the receiving end. It is very simple, and is founded 
upon two well-known experiments. 
The theory of the action at the sending end is founded upon an 
experiment by Sir William Thomson, who finds that iron subject 
to magnetic induction has its magnetism increased when slightly 
stretched in the direction of its magnetization, and we may assume 
that it is diminished when compressed. How, what happens in the 
Telephone ? Take the simplest case which has been tried, i.e. } where 
there is no vibrating plate. S ound waves (which are capable of 
passing through solid bodies as they do through the air) strike the 
end of the bar magnet, and are propagated through it. Waves of 
alternate compression and extension pass through the length of the 
magnet. Consequently, the magnet is alternately increased and 
diminished in intensity. Hence, currents are generated in the coil 
of wire, and transmitted through the telegraph wire alternately, in 
opposite directions, as in the theory hitherto adopted. The presence 
of a vibrating plate of any kind may help this action, by increasing 
the intensity of the waves transmitted through the bar magnet ; 
and if the plate be of iron it will undoubtedly help to propagate 
these currents. 
