549 
of Edinburgh, Session 1879 - 80 . 
apart on the thermo-electric scale, — the greatest current, as far as I 
have yet observed, being got from antimony and bismuth. 
It is very difficult to decide as to the cause or causes of such 
currents. They may be (1) purely thermo-electric ; (2) the cur- 
rents, which are the supposed cause of friction ; (3) currents pro- 
duced by contact force between adhering films of air, moisture, or 
other substances with which the surfaces rubbed are tarnished ; or 
(4) they may arise from all these causes combined. The following 
experiments were made in hopes of getting some information on these 
points. 
My first experiment was to obtain the exaot difference, as far as 
the production of a momentary current is concerned, between 
rubbing two pieces of metal together, and knocking the one against 
the other. For this purpose I repeated, with greater care, an ex- 
periment which I formerly described. It consisted in attaching a 
wire firmly to an ordinary hammer, and leading it to one of the 
erminals of a telephone circuit, while the wire from the other 
terminal was rigidly attached to a stiff bar of copper held vertically 
in a small table vice. When the face of the hammer was rubbed 
against the end of the copper bar, a very distinct grating noise was 
always heard in the receiving telephone ; but the sound was almost 
inaudible when the bar was knocked by the hammer, if proper care 
were taken not to combine rubbing with knocking. This is, how- 
ever, so difficult practically, that it is just possible that the sounds 
which I heard are due to faint rubs accompanying the knocking. 
Should this not be the case, however, this difference of effect 
seems to show that the currents are not wholly, although they may 
be mainly, thermo-electric, as it is hard to believe that the heat pro- 
duced at the junction of the surfaces by a smart blow can be less 
than that produced by a faint rub. Granting that the knocking is 
actually heard, it seems not unlikely that this effect may be due to 
the currents associated with rapid changes of form in matter. As 
has been remarked by Professor Tait (Proc. Eoy. Soc. Edin. vol. 
ix. p. 552), these currents are such as would be capable of detection 
by the telephone. 
In order to detect what effect, if any, the presence of the air had 
upon these friction currents, I employed the apparatus commonly 
called the electric egg. Having unscrewed the interior balls, I 
