But if in laud lines the insulation of the wires is attended with difficulty, the insulation is very much more difficult and uncertain in carrying a cable through great lengths of ocean. In transmitting a current along an insulated land line connected with the earth at one end, and with a battery communicating with the earth at the other, the strength of the current diminishes in an inverse ratio to the length of the wire. When, however, the conducting wire is wrapped up in an insulating material and immersed in water, we have the three parts of a Leyden jar, namely, the wire takes the place of the inner metallic coathig, the gutta percha that of the Insulating glass, and the iron wire covering or the water that of the outer coating, so that not only does the strength of the current diminish inversely as the length, but the arrangement must be discharged before a message can be sent, and the rate of signalling must depend ultimately on the rapidity with which the charge and discharge can be effected.
But there are mechanical difficulties arising from the materials employed. The copper conducting wire is deficient in strength, and a moderate strain put upon it may produces permanent elongation : while the materials which envelope it, being more elastic, return to their original bulk when the strain is removed, the elongated copper does not do so, but will thus cut its way through the gutta percha cover. The copper wire is usually made into a strand for the sake of strength, and it has even been proposed to cover the copper with fine steel wires before putting on the gutta pereha, so that the chief strength of the cable might be in its core. Copper wire is preferred on account of its superior conducting power for electricity, which is seven times greater than that of iron. But this superiority is greatly dependent on the purity of the copper. No substance added to copper increases its con ducting power ; but the presence of other metals may reduce its con ducting power below that of iron. The conducting power of copper also varies with the temperature.
Gutta perches is also subject to various changes which interfere with its insulating power ; although usually considered as a good insulator it is by no means perfect, especially when it Is exposed to a tolerably high temperature. At 32° there is a very small leakage of electricity through it ; at 32° the leakage is three times as great; at 72' it is nearly six times ; and at 92' ten times as great as at 32°. While laying the Red Sea cable, the temperature in the hold of the ship was 92°, and the insulation was ao bad that the engineers could not speak through the cable ; but when in the water at the depth of 300 fathoms and the temperature 73', the insulation became much better. Outta percha also varies greatly in its quality : it may contain foreign matter, or be porous, or air bubbles may become entangled with it during the laying on, tall of which may tend to destroy its insulating power.
There may be bad joints or small punctures, or a strong battery power may produce a chemical action injurious to its stability. Its durability is also seriously affected by exposure to light and air, so that in shallow water india rubber may be a preferable coating. Gutta percha is also liable to injury from friction or from pressure, and both it and the hempen packing are liable to the attacks of marine animals, a species of Teredo devouring the hemp, and another Teredo penetrat ing the gutta percha.
Hence it will be seen that the great problem of ocean telegraphy cannot by any means be regarded as solved. The best form of cable remains to be invented: it should be light and flexible for deep waters, and sufficiently strong in shallow ones to resist the rude grip of an anchor. Some cables weigh 3 or 4 tons per :mile ; others 8 or 9 or even more, while light cables weight about 1 or 2 tons per mile. The method of stowing them on board a ship, and the machinery for pay ing them out require to be reconsidered ; but, above all, the many costly blunders which have been perpetrated with various cables ought net to be again possible, such as the heating of the Rangoon cable, and the error iu reckoning in paying out the Cagliari and Algeria cable, so that there was not enough cable to reach the land : the ship held on during five days to the cable and then broke off' in the midst of a storm. It is not our business here to do more than allude to the failures of telegraphic cables consequent on the plan adopted by our government of granting premiums to telegraphic companies.
This part of our subject will not be complete without a few details as to the mode of preparing the Atlantic cable. A strand of seven wires of, pure copper of the No. 22 gauge, was first pre pared, it being the sixteenth of an inch in diameter when twisted. The strand of seven wires was adopted in preference to a single wire of the same practical capacity, because the probability of a destruction of continuity was in this way greatly diminished. In case of any accident occurring it was very unlikely that all the seven wires would be broken in exactly the same place, and so long as only one of them remained sound, the electrical transmission could be carried on. The strand itself was subject to a strain which stretched it twenty per cent., without any appreciable injury to its conducting power being discovered. To show that no amount of attenuation which could be produced by accident, could interfere to any important extent with its utility as a telegraphic conductor, one mile of wire eleveu times smaller than the strand, was introduced into a gap made in a 600-miles length of the cable, and the effect produced ou the trans witting power of the cable by the interpolation was tested. It proved that the transmitting capacity of the cable was only diminished by one thirty-seventh part.