1864. | Crooxess on the Atlantic Cable and its Teachings. 45 
Unfortunately, the first-laid submarine cables were attended with 
complete success; these precedents were used as arguments aguinst 
any further investigation, and hence the hasty enterprise of the Atlan- 
tic cable, involving an expenditure of three-quarters of a million, was 
rushed into in the most reckless manner, and with so utter a dis- 
regard of precautions, as to seem from the first actually to invite 
failure. 
The perfection of a cable depends upon the perfection of each 
individual inch of it; in this respect it is similar to a chain, which is 
valueless if a single link be faulty. The insulating covering of the 
conductor is composed of substances so delicate in texture, and laid on 
in such a manner as to render it extremely difficult to avoid faults. 
‘These are generally noticed as soon as they appear, and by taking the 
precaution to test the cable in definite lengths under water, they can 
be readily detected at any time, and their position ascertained. What 
is generally known as a fault, is a communication between the 
conducting wire and the water; this may be either very slight, in 
which case, the insulation is more or less injured, or it may be 
sufficient for the whole of the electricity to leak through. A small 
fault, which would not be of serious consequence in a short line, 
cannot be tolerated when the cable is of considerable length, as the 
powerful currents necessary to force a signal through, find out all the 
weak points, and eat them into fatal holes. There is another reason 
why faults or even weak places must not be admitted in submarine 
lines; it is that they are so liable to injury through lightning. In 
the Channel Islands’ telegraph, the lightning struck the cable in 
Jersey, and passing under the sea along the wire for sixteen miles in 
the direction of Guernsey, met with a weak place, where it burnt itself 
through into the water, destroying the insulation. 
The material of the outer covering of the cable, and the manner 
in which it is laid on, are matters of great importance. There must 
be no strain on the core, and the finished cable must have as little 
elasticity as possible. Many cables have been injured from a neglect 
of this precaution : an elastic rope will stretch four or five per cent. 
during deposition, and will contract when the tension is removed and 
the temperature is lowered by the surrounding water. The copper 
wire is however permanently stretched, and where the gutta-percha 
contracts over it, the wire occasionally knuckles through and produces 
a serious leakage. The outer coat of mail is almost invariably of a 
spiral form, which perhaps is the only kind that could be adopted, 
having regard to the frequent coilings and uncoilings which the rope 
has to go through, but such a form is very lable to kink whenever the 
rope is.not kept in a state of tension. 
The copper of which the conducting wire is now invariably made, 
should be selected with the greatest care. When pure it is one of 
the best solid conductors known ; but very slight impurities, such as 
are almost always met with in the commercial metal, are sufficient to 
greatly diminish its value. ‘Taking the conducting power of pure 
copper as 100, Dr. Matthiessen found that of samples of American, 
Australian, Russian, and Spanish copper to be respectively 92, 88, 
