551 
of Edinburgh, Session 1879 - 80 . 
rubbing substances are arranged like a pair of mill-stones, the lower 
stone being a disc of iron laid horizontally, and the upper a disc of 
copper mounted on a vertical axis on which it can revolve. The 
surfaces are kept pressing against each other by a strong spring. 
When the upper disc is made to revolve rapidly, a very decided 
current is produced ; and this I found to be markedly increased, as 
indicated by the telephone, by feeding in between the discs powdered 
antimony and bismuth combined. Of course we have here a series 
of rapid reversals of the current, as the direction of the current will 
depend upon whether particles of antimony or particles of bismuth 
are in contact with the lower plate. This clearly indicates a thermo- 
electric effect ; and I have no doubt that the effect will be increased 
by applying a means whereby the upper surface of the copper plate 
and the lower surface of the iron one can be kept cold by a freezing 
mixture. As yet, however, I have not had time to try that. In 
another form I took two cylinders, the one of antimony and the 
other of bismuth, and placed them together end-wise, the pressure 
between them being regulated by a screw. The antimony cylinder 
was kept stationary, and the bismuth made to revolve very rapidly 
against it, so much so that both cylinders rapidly became hot. This 
also gave a pretty strong current. 
Seeing that the friction between metals does certainly produce an 
electric current, it seemed natural to inquire whether an electric 
current sent from a battery across the surface between two metals 
would not modify the friction of the one against the other. I have 
tried to test this in a variety of ways, and the results leave me in 
doubt whether to attribute the indications which I have received to 
actual changes in the friction or to incipient fusion of portions of 
the surfaces together by the heat produced by the current, or to an 
effect similar to the Trevelyan rocker. In one experiment I made 
an inclined plane which carried a pair of parallel rails of copper 
about three quarters of an inch apart. The rails were hinged at the 
lower end, so that the plane could be set at any angle with the 
horizon. It was so arranged that the current from the battery could 
be sent up the one rail, through any conductor laid across the two, 
and down the other rail. The surfaces of the rails were made quite 
smooth. When a heavy piece of metal was laid across the rails, the 
angle of repose was the same both when there was and was not a 
