704 
Proceedings of the Royal Society 
the sound is very much enhanced, arising no doubt from the wire 
coming in direct contact with the disc. To see if the dipping 
copper wires had any influence on the sound, I first held them fixed 
in the hand and found no change. I then took out the mercury and 
put in a solution of common salt using copper electrodes ; the disc 
sounded for a beat or two and then ceased, but when the dipping 
parts were of platinum the beats were quite regular. In this latter 
case the sound was accompanied by a slight hissing arising from the 
action of the disengaged gases, but this was quite removed on 
holding the wires. The sound got when brine or acidulated water 
is put in the cups is not so loud as with mercury, in consequence of 
the diminution of current strength. 
We do not get the same loudness with this arrangement of the 
disc in the middle as when only one wire is tightened and the 
other left loose, for the simple reason that the tightness and conse- 
quent elasticity which is favourable to the action of the telephone 
cannot be so easily got at with a disc balanced between two equal 
pulls. However, as the common disc is stretched more perfectly 
the sound rises, so as to leave no doubt, that if the exact tightness 
were got, the two-pull telephone would be as good as the one with 
the single pull. Much may be said of the necessary imperfection 
attending the two-pull telephone. The threads on each side may 
not be precisely in a straight line, the friction on the rods may not 
be exactly alike, and that coupled with the comparative fixity of the 
disc may make a slight difference of pull on opposite sides. But 
taking all these into account, one would expect that as the condi- 
tions approach perfection there would be a corresponding silence, but 
such is not the case. You may take the telephone into your hand, 
move it gently about in all directions to secure a position where the 
loudness is less, but such is not to be found. So far, then, as I 
can interpret this two-pull telephone, it is something else than the 
pull of the wire that emits the sound, and the single pull experi- 
ment by no means proves the existence of sudden extensions and 
contractions in the wires. I do not, on the other hand, say that 
my two-pull experiment disproves them. The conduct of a wire 
expanding instantaneously in every part may produce an effect like 
a molecular click, but if so its action must be something very 
different from the simple pull supposed. A stretched disc with a 
