The telegraph manual: a complete history and description of the semaphoric, electric and magnetic telegraphs of Europe, Asia, Africa, and America, ancient and modern.

646 ATLANTIC OCEAN TELEGRAPHY.
two electrified media are so far removed, as in a line of 2,000
miles, if connected with the earth, a very considerable time is
occupied both in charging and discharging, causing much retardation of the current, so that I think four words per minute
will be the maximum rate of transmission through any Atlantic cable with the present dot and dash system. If other
plans can be worked by which a letter would be indicated by
one or two signals, the rate would be increased in proportion.
As I have made use of the terms resistance and retardation, and as they are words having different meanings, I will
explain what constitutes the difference. The " resistance " of
a wire has the effect of keeping part of an electric current
back, or diminishing its quantity, without affecting its velocity,
the remainder passing as quickly as it would through a wire
of the same length with less than a hundredth part of the
resistance. The effect of "retardation," on the contrary, is to
diminish both the quantity and velocity of the current. For
example, in an overground well-insulated wire, 2,000 miles
long, an electric current or impulse would traverse the entire
length in one tenth of a second; through the same extent of
submarine line, owing to the effect of the charge, the time
occupied would be nearly a second and a half.
Respecting the question of injury to the line from the use
of powerful currents-if a small hole leading to the wire
exists in the gutta-percha covering near either end, there is no
doubt that a current of great quantity and intensity, whether
produced by battery or coils, would have the effect of enlarging
the breach by burning; but this can only take place to a
limited extent. Heat can only be developed by an electric
current when the latter meets with great resistance; consequently, as soon as that is diminished by a slight enlargement
of the hole,, all burning ceases. I tried the experiment alternately with the large induction coils, with the battery now
here (400 cells of Daniell's) and with my large magneto-electric
machine. They were each connected in turns with the line
and the earth, and at the same time with a piece of guttapercha-covered wire, in which the copper was bared to one
thirty-second of an inch diameter, and a piece of copper in a
basin of sea-water, thus dividing the current between the two
routes. The coil current enlarged the fault to one twentieth
of an inch in diameter; the batteries to a sixteenth-both very
slowly. That from the magneto-electric machine made no
change in the fault it was applied to until it was disconnected
with the line and earth, and allowed the one road only; when
burning took place, as might have been expected. The