of fracture. Acting on the hint, he placed
the two broken ends of the wire together again, and kept them so by
the application of a definite pressure. To his joy he found that he had
discovered what he had been in search of. The imperfect contact between
the broken ends of the wire proved itself to be a means of transmitting
sounds, and in addition it was found to possess a faculty which he had
not anticipated--it proved to be sensitive to very minute sounds, and
was in fact a rude microphone. Continuing his researches, he soon found
that he had discovered a principle of wide application, and that it was
not necessary to confine his experiments to wires, since any substance
which conducted an electric current would answer the purpose. All that
was necessary was that the materials employed should be in contact
with each other under a slight but definite pressure, and, for the
continuance of the effects, that the materials should not oxidise in air
so as to foul the contact. For different materials a different degree
of pressure gives the best results, and for different sounds to be
transmitted a different degree of pressure is required. Any loose,
crazy unstable structure, of conducting bodies, inserted in a telephone
circuit, will act as a microphone. Such, for example, as a glass tube
filled with lead-shot or black oxide of iron, or 'white bronze' powder
under pressure; a metal watch-chain piled in a heap. Surfaces of
platinum, gold, or even iron, pressed lightly together give excellent
results. Three French nails, two parallel beneath and one laid across
them, or better still a log-hut of French nails, make a perfect
transmitter of audible sounds, and a good microphone. Because of its
cheapness, its conducting power, and its non-oxidisability, carbon is
the most select material. A piece of charcoal no bigger than a pin's
head is quite sufficient to produce articulate speech. Gas-carbon
operates admirably, but the best carbon is that known as
willow-charcoal, used by artists in sketching, and when this is
impregnated with minute globules of mercury by heating it white-hot and
quenching it in liquid mercury, it is in a highly sensitive microphonic
condition. The same kind of charcoal permeated by platinum, tin,
zinc, or other unoxidisable metal is also very suitable; and it is a
significant fact that the most resonant woods, such as pine, poplar, and
willow, yield the charcoals best adapted for the microphone. Professor