ble is
essential. It has been found that the electrostatic capacity of one mile of
submarine cable is equal to the capacity of 20 miles of overhead line, and
as the effect of capacity is to retard the current and reduce the speed of
working, it is evident that where there is any great length of cable in the
circuit the distance of possible transmission is enormously reduced.

If we take for an example the London-Paris telephone line with a length of
311 miles and a capacity of 10.62 microfarads, we find that about half this
capacity, or 5.9 microfarads,[1] is contributed by the 23 miles of cable
connecting England with France.

In practice the reduction of speed due to capacity has, to a great extent,
been overcome by means of apparatus known as a line-balancer, which hastens
the slow discharge of the line and {4} allows each current sent out from
the transmitter--the current in several systems being intermittent--to be
recorded separately on the receiver. Photographs suitable for press work
can now be sent over a line which includes only a short length of cable for
a distance of quite 400 miles in about ten minutes, the time, of course,
depending upon the size of the photograph. In extending the working to
other countries where there is need for a great length of cable, as between
England and Ireland, or America, the retardation due to capacity is very
great. On a cable joining this country with America the current is retarded
four-tenths of a second. In submarine telegraphy use is made of only one
cable with an earth return, but special means have had to be adopted to
overcome interference from earth currents, as the enormous cost prohibits
the laying of a second cable to provide a complete metallic circuit. The
current available at the cable ends for receiving is very small, being only
1/200000th part of an ampere, and this necessitates the use of apparatus of
a very sensitive character. One system of photo-telegraphy in use at the
present time, employs what is known as an electrolytic receiver (see
Chapter III.) which can record signals over a length of line in which the
capacity effects are very slight, with the marvellous speed of 12,000 a
minute, but this speed rapidly decreases with an increase of distance
between the {5} [Illustration] two stations. The effect of capacity upon an
intermittent current is clearly shown in Fig. 1. If we were to send twenty
brief currents in rapid succession over a line of moderate