552 Proceedings of the Royal Society 
current passing. It was different, however, when a light body, such 
as a sewing needle, was put on. Then, when the current from three 
Bunsen cells was passing, the plane could he elevated considerably 
past the angle of repose for no current, before the needle rolled 
down. On examination I found that the needle was actually stick- 
ing to the copper ; but that, in almost all cases, this sticking gave 
way without the angle being altered after the current had been taken 
off for some time, and the needle and copper allowed to come back 
to their normal temperature. In another experiment I employed a 
Bell telephone to enable me to detect any variation of friction when 
a current was passing between the rubbing surfaces. To the centre 
of the telephone disc was attached a long narrow strip of light wood; 
the object of making the strip so long being to remove the telephone 
as far as possible from the inductive action of the battery current 
which was to he used. To the other end of the strip was attached 
a flat piece of bismuth. This rested on the convex surface of a 
cylinder of antimony, which could he rapidly rotated. The battery 
current was sent through the antimony and bismuth by entering the 
antimony by the axis on which it revolved, and leaving the bismuth 
by a spring pressing tightly against it. In the battery circuit was 
included the violin with its microphone already mentioned, and 
the telephone with the rod attached was placed as the transmitting 
telephone in a telephone circuit. When the antimony cylinder was 
rapidly rotated, a listener in the receiving telephone watched atten- 
tively till his ear became accustomed to the sound produced hy the 
rubbing, and transmitted along the wooden rod to the telephone disc. 
The battery circuit was then joined, and the violin played, the anti- 
mony cylinder meanwhile rotating at the same rate as before. No 
alteration in the sound was audible, which indicated no alteration in 
the friction. I then substituted a sharp point for the flat piece of 
bismuth, and immediately the violin sounds were faintly but clearly 
heard. This led me to think that some sticking was produced hy 
the fusing of the sharp point, and more especially as the sound 
became a little clearer as the rotation became very slow. 
Acting on this hint, it immediately occurred to me that a receiv- 
ing telephone could be constructed depending upon this effect. I 
therefore took my bismuth cylinder and mounted it on a frame so 
that it could be made to rotate very truly on pivots. By wheels 
