sent Produced by its action an electric current which, after traversing
the line, passed through a coil and deflected a suspended magnet to the
right or left, according to the direction of the current. A mirror
attached to the suspension magnified the movement of the needle, and
indicated the signals after the manner of the Thomson mirror
galvanometer. This telegraph, which was large and clumsy, was
nevertheless used not only for scientific, but for general
correspondence. Steinheil, of Munich, simplified it, and added an alarm
in the form of a bell.

In 1836, Steinheil also devised a recording telegraph, in which the
movable needles indicated the message by marking dots and dashes with
printer's ink on a ribbon of travelling paper, according to an
artificial code in which the fewest signs were given to the commonest
letters in the German language. With this apparatus the message was
registered at the rate of six words a minute. The early experimenters,
as we have seen, especially Salva, had utilised the ground as the return
part of the circuit; and Salva had proposed to use it on his telegraph,
but Steinheil was the first to demonstrate its practical value. In
trying, on the suggestion of Gauss, to employ the rails of the Nurenberg
to Furth railway as the conducting line for a telegraph in the year
1838, he found they would not serve; but the failure led him to employ
the earth as the return half of the circuit.

In 1837, Professor Stratingh, of Groninque, Holland, devised a telegraph
in which the signals were made by electro-magnets actuating the hammers
of two gongs or bells of different tone; and M. Amyot invented an
automatic sending key in the nature of a musical box. From 1837-8,
Edward Davy, a Devonshire surgeon, exhibited a needle telegraph in
London, and proposed one based on the discovery of Arago, that a piece