552 
Proceedings of the Royal Society 
swing. The result is, of course, only a very rough approximation. 
It is that a single Grove’s cell would produce, in a circuit of some- 
where about a billion B.A. units resistance, a current sufficient, if 
reversed 500 times per second, to produce an audible sound in the 
telephone I employed. 
3. Several attempts at explanation of the action of the telephone 
have been given here and elsewhere, and others are promised for 
to-night. For my own part,' I think there are at least three separate 
causes at work in the telephones I have used. 
There can he no doubt that the inventor’s own explanation is, at 
least to a certain extent, correct. For we can easily dispense with 
the magnet in the receiving telephone, using merely a thin iron disc 
in front of a coil.’ And Mr Blyth has, I believe, found that we may 
make the disc, even in this case, of copper, and yet have transmission 
(though very feeble) of intelligible sounds. 
But this cannot be the full explanation. For it does not attempt 
to account for the peculiar nasality of the transmitted speech. 
Without going more closely into the matter, the difference of quality 
between an open and a closed pipe suggests a certain amount of 
constraint as the cause. And we know that the sounds in the 
original telephone of Reiss were produced by molecular motions due 
to magnetism in soft iron. Mr Blyth has shown conclusively that 
molecular motion in the magnet itself has a large share in the results, 
because he has successfully substituted other metals than iron, and 
even non-conductors, for the disc, and in certain cases finds that he 
can dispense with the disc altogether. 
Besides this, however, it seems to me that there is a third cause, 
which in certain cases is more effective than either of the others. 
This is suggested by the fact that (at least with the instruments I 
have tried) high notes, even of comparatively small intensity, are 
much more clearly transmitted than low notes, — indicating that the 
rapidity of the molecular change has a great deal to do with the 
result. In fact, in this respect, the telephone is really a variety of 
the so-called curb-key, giving very sudden reversals. 
These considerations have led me to fancy that rapid change of 
form in matter, whether paramagnetic or not, may probably be 
capable of detection by the telephone, for the associated electric 
currents may be in certain cases powerful enough to produce audible 
