Construcithn of the 47 ( jil. 57) exhibits the construction of the deep-sea cable of 1865, that of 1866 being substantially the same. A cable formed of seven plaited strands of fine copper wire constitutes the conductor. This is protected in the following manner: It receives first a coating of an insulating compound (called Chatterton's compound) com posed of a mixture of gutta-percha, wood-tar, and resin. Over this collies an accurately concentric primary layer of gutta-percha. Upon this come, alternately, four thin layers of Chatterton's compound and gutta-percha, and over all these a covering of jute. The outer protective covering con sists, filially, of iron wires one-tenth of an inch in thickness, enclosed in prepared manilla hemp and wound around the central core in long spirals. The deep-sea portions of this cable had a diameter of inches, while the thickness of the shore-sections, where it would be exposed to the action of waves, injury from icebergs, from anchors, etc., was increased to 2.;< inches by a heavier protective armor.
It was found, however, that the iron wires of the protecting sheathing soon became corroded. In certain waters, also, the cable is exposed to the attacks of worms, which penetrate between the protecting wires Gov. 57, fig. 48) and eat into the core. To avoid this danger, C. W. Siemens adopted a plan of constructing submarine cables which Prescott describes as follows: The copper conductor is first covered with a layer of Chatter ton's compound and then with two spiral layers of India-rubber. The second layer is so put on that its joints are at an angle of about go° to those of the first one. The insulated core is then covered with another layer of Chatterton's compound, and ag-ain with gutta-percha. The outer envelope of the cable consists of a double layer of tarred hempen bands wound spirally in opposite directions, and lastly of an outer metallic en velope composed of two copper strips wound on spirally, so that the turns lap over one another. The covering of the later Atlantic cables is sub stantially like that of the 1865 cable, above described; the principal differ ence consists in dispensing with the tarred-hemp covering- of the iron sheathing and in galvanizing the iron wires composing the latter.
Cable the operation of submarine cables, their great length and perfect insulation render it necessary to make provision against the so-called " return currents," which manifest themselves in the slug gishness with which they part 1.vitli their electrical excitation, and so exert a distiirbin,t4- influence upon the transmission of signals. Again, the sub marine cables with their insulating coverings may be compared, when in an electrified condition, to a condenser or elongated Leyden jar, the gutta percha covering corresponding to the glass of the jar, the copper con ductor within the gutta-percha to the inside coating of the jar, aud the iron sheathing to the outer coating. When a current floi.vs in the cable,
it induces in the outer sheathing an electrified condition of the kind oppo site from that flowing in the line wire, and this by reaction causes induced currents in the latter. This phenomenon is termed "electro-static induc tion," and seriously interferes with the discharge of the line. To over come these difficulties, telegraphing through submarine cables is done with comparatively feeble and alternating, currents.
To Locate Faults or breaks in a telegraph-line several methods are available. The one most commonly' employed involves the use of the so called "differential galvanometer" and a rheostat, or resistance coil. The principle on which this method is based is that of passing a current of 1-znown strength through a series of coils of German-silver wire, the resist ance of each of which is known, until the needle of the galvanometer indicates that the resistance obtained is the same as that exhibited by the defective line. The normal resistance of the line is, of course, supposed to be known. The indicated resistance, when found by this mode of meas urement, therefore, corresponds to a given length of the line wire, which (-,-ives the location of the fault.
The consideration of the subjects of duplex, quadruplex, and multiplex telegraph systems, by' which two or many signals may be sent over a line wire simultaneously and in opposite directions, and by which a notable increase in the speed of transmission has been accomplished, would un duly extend this section. For the explanation of these modern improve ments the reader is referred to special treatises.
Train Telegraphy by of the most interesting recent advances in telegraphy, that promises shortly to come into gYeneral use be cause of its great utility and convenience, is the maintenance, by several ingenious methods, of communication between moving railway-trains and fixed stations along- the line. The practical application of telegraphy to this service is due to several American electricians—notably, to Edison, Gilliland, Phelps, and Smith. The principle involved is that known to electricians as induclion, in virtue of which an electric current traversing a conductor will cause a current to be generated in a contiguous con ductor not in contact with it, and this inductive effect will be appre ciable in conductors at some distance from one another. There is no con tact or line conductor, the telegraphic communication being obtained simply by the spontaneous reproduction in separate conductors of electrical impulses similar in kind to those originally produced. This system of train telegraphy, which has been introduced on the Lehigh Valley .Rail road, in Pennsylvania, has proved a valuable acquisition both in respect of greater safety in the operation of the trains and in the convenience it affords the travelling public.