620 
Proceedings of the Eoyal Society 
used by electricians. On connecting one of the coils with a 
telephone and making the fork vibrate, I was astonished to hear 
the sound in the telephone astonishingly loud. I found, how- 
ever, that this arose from the connecting wires, which though 
loose were able mechanically to transmit the vibrations of the fork. 
This was shown by leaving only one wire in the connection, when 
the sound of the fork was still heard. It is well known that the 
thread of a mechanical telephone delivers its message to the first 
fixed obstacle it meets and no further, and in this case, when the 
lines were held in the fingers, a mere residuum of sound was heard, 
which could only be properly estimated in a distant room away from 
the direct sound of the fork. In this, as in all subsequent experi- 
ments, I took the utmost care that no mechanical transmission was 
possible, and the telephone used at the operating station was put out 
of circuit each time a sound was made for investigation. Loud 
tapping was made in every conceivable place to ascertain if such 
could be transmitted in a mechanical way. The circuit was about 
150 ft. of wire with a gas pipe return, and the two stations were in 
different buildings. The results obtained from this fork were not 
of any value other than illustrating the difficulty of insulating for 
transmission what may be looked upon as an internal and not an 
external vibration. They were these. The sound of the fork was 
heard much less loud than if the prong had been vibrating in front 
of the coils, but was louder than what was got when it vibrated at 
a distance equal to that of the prong on which the coil was not fixed. 
When the two coils were joined up consecutively, one way of joining 
gave a maximum, the other a minimum sound. In another arrange- 
ment the coil was attached to a hollow prism of thin brass § inch 
inside, which in its turn was attached to the prong. The sound 
given by this coil was less than when it was attached immediately 
to the prong, but was louder than when the other prong vibrated 
freely at the same distance. When I mounted the fork in this 
way, one prong was feebly magnetised, and the other scarcely if 
at all. 
A perfect maze of reasons may turn up to account for the currents 
dispatched by these fixed coils. One might be their motion in the 
magnetic field of the earth, which is not likely, for earth-induced 
currents proceed from the rotation of a coil, and the faint approach 
