683 
of Edinburgh, Session 1879-80. 
5 Cm. long, was soldered to a tolerably thick copper wire, which 
served as a terminal ; the other end was fixed securely to the copper 
spring attached to the mica-diaphragm of the ear-piece. The whole 
was then put in line with four Bunsen’s cells, and a microphone 
attached to a violin. The experiment succeeded at once. The 
music was perfectly audible close to the diaphragm, and a tune was 
reproduced quite distinctly. 
For convenience in experimenting with different wires, I con- 
structed the apparatus which I now exhibit to the Society. It 
consists of a fine palladium-silver wire (4 Ag. 1 Pd. Ees. *52 Ohm 
per Cm.), 8 Cm. long, soldered to two copper terminals, which are 
well amalgamated, and lie in the mercury of two cups forming the 
line terminals. One terminal is hooked to the membrane of a toy 
drum,* the other end of which is removed ; and the other terminal 
is attached to a string, to which is hung a scale pan, with small 
weights for producing the requisite tension. 
With this apparatus, I can reproduce the music of the violin in 
the far room, so that all present can hear it. The roughness which 
mars the effect, is simply due to vibrations of the microphone, which 
happen to be in unison, now and then, with the note of the violin. 
I have satisfied myself that the action of this instrument is not 
due to loose contacts, or to the earth’s magnetism. I believe it to 
be due to the variations in the heating of the wire, which follow 
the variations of the current strength caused by the microphone. 
The tension of the wire does not seem to be material, farther than 
that there must be a certain tension before the effect is produced ; 
for a wire absolutely loose, gives little or no effect. A thick wire of 
platinum, with the four cells, did not act until it was made very 
short ; and a wire of copper, 4J Cm. long and about *01 Cm. in 
diameter, would scarcely act at all. The only apparent exception 
\ that I found was iron. I found that I could get a tolerable result 
with an iron wire 4 Cm. long, and thicker than the copper wire last 
mentioned. A fine steel wire hairspring acted very well, but not so 
well as the long palladium silver wire. I also tried other metals, 
but none surpassed the platinum and palladium-silver wires. 
So far as I have been able to go with a very fine wire, the effect 
* For reproducing articulate speech, a small mica -diaphragm like those 
used by Edison, Blyth, and others, is best. 
