Page:Heroes of the telegraph (IA cu31924031222494).djvu/104

 the instrument is dumb, while in Bain's the solution requires a certain power of current to decompose it and leave the stain.

In overland lines the current traverses the wire suddenly, like a bullet, and at its full strength, so that if the current be sufficiently strong these instruments will be worked at once, and no time will be lost. But it is quite different on submarine cables. There the current is slow and varying. It travels along the copper wire in the form of a wave or undulation, and is received feebly at first, then gradually rising to its maximum strength, and finally dying away again as slowly as it rose. In the French Atlantic cable no current can be detected by the most delicate galvanoscope at America for the first tenth of a second after it has been put on at Brest; and it takes about half a second for the received current to reach its maximum value. This is owing to the phenomenon of induction, very important in submarine cables, but almost entirely absent in land lines. In submarine cables, as is well known, the copper wire which conveys the current is insulated from the sea-water by an envelope, usually of gutta-percha. Now the electricity sent into this wire induces electricity of an opposite kind to itself in the sea-water outside, and the attraction set up between these two kinds 'holds back' the current in the wire, and retards its passage to the receiving station.

It follows, that with a receiving instrument set to indicate a particular strength of current, the rate of signalling would be very slow on long cables compared to land lines; and that a different form of instrument is required for cable work. This fact stood greatly in the way of early cable enterprise. Sir William (then Professor) Thomson first solved the difficulty by his invention of the 'mirror galvanometer,' and rendered at the same time the first Atlantic cable company a