Escher.gif (426 bytes)

History of the Atlantic Cable & Undersea Communications
from the first submarine cable of 1850 to the worldwide fiber optic network

Mr. Sabine on Telegraphy at the
Paris Universal Exhibition, 1867

Introduction: The 1867 Paris Universal Exhibition had a number of telegraphy exhibits, which telegraph engineer and author Robert Sabine described and reviewed in an 1868 British Government publication containing reports on all aspects of the Exhibition. The section on Submarine Telegraphs is reproduced here.

--Bill Burns

Submarine Telegraphs

The submarine telegraph is not so well represented in the Paris Exhibition as the brilliant success which this branch of art has achieved within the past year would have led us to hope. It is sincerely to be regretted that the public, in this Exhibition, have been denied the gratification and advantages of learning the methods and seeing the apparatus employed in making, laying, and testing the Atlantic cables. The engineers engaged in this scheme have not only laid the Atlantic cables, but have at the same time established the future fortunes of submarine telegraphy, which is, perhaps, of still greater importance. In exhibiting their remarkably beautiful appliances they might not have found, it is true, any direct pecuniary profit to themselves, but they would have had the satisfaction of knowing that they were enlightening thousands of inquiring minds on a subject of universal interest.

Regarding the Atlantic cables as the last and great standards for submarine work, although not forming a conspicuous part of the Paris Exhibition, it will not be out of place here to give some particulars in connexion with them.

The proportions between conductor and insulator, as well as exterior protection and other particulars connected with these cables, were referred to a scientific committee appointed by the Atlantic Telegraph Company. This committee, taking as the basis of their operations the evidence of several competent persons given before them whilst acting conjointly with a committee appointed by the Lords of the Committee of Privy Council for trade to inquire into the construction of submarine telegraphs, decided upon the details according to which the Atlantic cables should be constructed. The brilliant success which has followed proves the wisdom of this policy above the method previously adopted of referring the questions of construction, &c. to some private engineering firm, whose opinions, although given in good faith, might nevertheless be faulty from imperfect knowledge or warped, perhaps unintentionally, by business interests.

A specimen of these cables is shown in the English section of the Paris Exhibition by Messrs. Webster and Horsfall, of Birmingham, the manufacturers of the iron wires used to give longitudinal strength to the cables. These cables are constructed after the type of that which was laid in 1860 between Toulon and Algiers. The core which was made by the Wharf Road Gutta-percha Company, consists of seven No. 18 gauge copper wires, twisted into a spiral strand weighing 300 lbs. per knot, covered with four coats of gutta-percha, with intermediate compound, weighing 400 lbs. per knot. This core was manufactured in separate lengths of one knot, which were all tested before they left the factory, both under atmospheric pressure and under a pressure of 600 lbs. per square inch, at a temperature of 24° Cels. The sheathing was done at the Telegraph Construction and Maintenance Company's works at East Greenwich. It consisted, first, in a covering of hemp saturated in salt-water, immediately upon the gutta-percha; secondly, in an outer covering for the deep sea portion, formed by ten (No. 13, B.W.G.) wires drawn from Messrs. Webster and Horsfall's homogeneous iron, each surrounded by five yarns of manilla hemp laid on spirally with a preservative compound. The total diameter of the cable is 1.127 inch; its weight in air 36 cwt. per knot, and its breaking strain eight tons.

The particulars connected with the attempted submersion of the first cable in 1865, the successful laying of the second, and memorable recovery, reparation, and completion of the first in 1866, are too well known to need more than alluding to here. But the recovery of the 1865 cable was the crowning victory of submarine telegraphy, and from that date henceforth deep-sea cables may be regarded as certain of success. The confidence of capitalists, shaken by the earlier failures of the old Atlantic, Red Sea, and Indian lines, is re-established, and the only question which can arise regarding the undertaking of future deep-sea lines will be their probable employment, and therefore commercial value.

The subsequent faults which have occurred and been repaired sufficiently illustrate the value of laying duplicate cables, as was done between Valentia and Newfoundland. When one cable breaks down the other is used until it is repaired, and thus the service is in no danger of being interrupted.

That which has also very materially enhanced the commercial value of the Atlantic cables is the high speed attained in telegraphing through them, with the improved methods employed, resulting from the increased familiarity which telegraph engineers are beginning to have with the behaviour, under varied conditions, of the strange force we call electricity. Calculated from the data afforded by the measurements of the Red Sea and Malta-Alexandria cables, the speed of working through the new Atlantic did not promise to be higher than 1.15 words per minute [The Electric Telegraph, 1867, p. 422]. On their establishment, however, a new system of transmission was adopted, the joint invention of Professor William Thomson and Mr. Varley. The transmission was effected by means of a key of peculiar construction, which for each signal sent a series of five separate waves of different lengths, and alternately positive and negative, into the cable. The first one, which was positive, was the only one intended to reach the further end of the cable and move the receiving instrument. The second wave was negative, of shorter duration, and was wholly absorbed before it reached the further end. The third wave, of still shorter duration, positive, reached only a little beyond half the length; the fourth, negative, still shorter; and the fifth, positive, the shortest of all, attained but a short distance along the conductor. Thus, instead of leaving the cable charged, according to the old method, with one kind of electricity throughout its entire length after every signal, by this method the unabsorbed charge is distributed in five alternately positive and negative sections from one end to the other the instant after the reception of a signal. But this is a condition the most favourable for a speedy equation and neutralization of the whole charge; so that after each signal the cable may be regarded, in a practical sense, as electrically without charge. With this system, and the employment of a modification of Professor Thomson's reflecting galvanometer as receiving instrument, the rate of signalling was increased to five words per minute, or four times as high as that which would have been attainable with the method and with the instruments used on the Red Sea and Malta-Alexandria lines.

Subsequently this system has been replaced by a still more efficient one, suggested also by Mr. Varley [Du Moncel's report; “Etudes sur l'Exposition, etc." vol. p. 389]. The conductor at each end of the cable is connected with one of the sides of a condenser of about 40,000 square feet surface, and also, through a great resistance, with the earth. At one end the opposite plates of the condenser are connected with a manipulating key and battery of 10 galvanic elements, by which it may at pleasure be charged. At the other end the opposite plates of the condenser are connected through a Thomson's galvanometer with earth. Suppose a positive signal is to be given at one end, therefore, the condenser plates which are next to the apparatus are charged by the copper pole of the battery being put into connexion with them through the key. The plates to which the cable end is connected receive thereupon, by induction, a corresponding negative charge from the interior of the cable, the neutral electricities of which are decomposed, and a positive charge equal to nearly the fiftieth part of that communicated to the transmitting condenser flows on to the first plates of the condenser at the further end. The opposite plates receive thereupon, by induction, a negative charge from the earth through the galvanometer, the needle of which is deflected. When the signal is given the manipulating key, which remains depressed about one-fifth of a second at the sending station, is let go; and the first plate of the condenser being thus put again to earth, the charge of the second plate is released, falls back into the cable, and neutralizes the polarization produced there and in the distant condenser. In this neutralization the earth connexions of the cable ends, although having a very considerable resistance, play an important part. Suitable commutators at each of the stations enable the receiving and transmitting portions of the apparatus to be easily exchanged for one another, according to the direction in which the signals are to be sent

With this apparatus a speed of transmission of eight words per minute is easily attained through the whole length of the cable between Valencia and St. John's.

Since 1862 there has been no new insulator suggested for submarine cables. Gutta-percha still holds the foremost place, and since its manufacture has been brought to such a high degree of perfection, it has become, in this position, unexceptionable. A more reduced price would be desirable, but not if this is to be attained at the expense of the purity of the material. Gutta-percha is applied to insulated wires in a series of two or more coatings with intermediate layers of Chatterton's compound—a mixture of tar, resin, and gutta-percha. The application of this compound has the twofold object of filling up any cavities or air holes which might be left in the gutta-percha, and of forming a cement between the separate coatings. It is questionable, however, whether this compound is beneficial to the constancy of the gutta-percha, which has a tendency to absorb the tar out of it, and to leave the resin in a dry state.

The principal exhibitors of gutta-percha wires are Messrs. Rattier and Co., of Bezons, near Paris. The gutta-percha core made by them is of excellent quality. That employed by the French administration of telegraphs contains, per knot, 22 kilogrammes of copper and 33 kilogrammes of gutta-percha, in two coats, with intermediate compound. The writer was allowed to test the electrical conditions of some lengths of this core which were awaiting approval by the Government electrician. The resistances were extremely uniform, and in no case was the insulation less than 700 millions-units per knot, in water at 14 deg. cent. The wire resistances were also such as to show that the French manufacturers are in a position to procure copper of high conducting power.

In the department of the French colonies some wire is shown covered with a gum called “balata,” brought from the French African settlement of Senegal. This gum, which resembles in general appearance the ordinary gutta-percha, is less applicable for the insulation of telegraph wires, as its melting point is considerably lower, rendering it more easily liable to mechanical injury when exposed to an augmentation of temperature. In such positions cables insulated with vulcanized india-rubber have been found to behave well.

William Hooper, London, exhibits a case containing various specimens of his vulcanised india-rubber cables. Mr. Hooper's process consists in the consolidation of a coating of pure underneath a coating of vulcanised india-rubber. He attains this by covering the conductor first with two coatings of pure rubber, then with a coating of some substance called a separator, and lastly with vulcanised india-rubber. The duty of the separator is to prevent the sulphur of the outer coating penetrating into the interior. The whole is then submitted to a temperature of from 135 deg. to 140 deg. cent., by which the rubber is reduced into a compact coating.

The lengths of cable which have been made by this process have behaved, so far, very well. Their insulation resistances are great; a result, however, ascribable not so much to Mr. Hooper's peculiar process as to the repeated coatings, laid one upon the other, which compose his cables, and to the specific resistance of the material itself; any other manufacturer using the same thickness of rubber would obtain the same insulation. But the advance made by the method in question is in the favourable conditions under which the dielectric is applied to ensure its durability.

To the practical telegraphist so high a degree of insulation is not necessary. To the electrician, however, it is a great boon, enabling him, when the insulation of the whole wire is uniform, to determine the place of a very minute fault with great accuracy and ease, as he can neglect, in his calculations, the shunt resistances due to the leakage between the fault and the ends.

The physical properties of india-rubber are such that it may be heated to a higher degree than is possible with gutta-percha without materially altering either its molecular or electrical conditions. This in in favour of its employment in cables liable to be exposed to heat in transport or otherwise. The self-heating by oxydation of the iron covering, for instance, which was found in 1860 to have injured the core of a length of the Malta-Alexandria cable during the time it was stowed in the tank at Greenwich, would have been without any detrimental effect had the cable been insulated with india-rubber. In stowage in dry tanks during transport cables frequently become spoiled, as was the case with four of the five experimental lengths sent by the Government for submersion in the Persian Gulf in 1863. In such a position the cable proposed by Mr. Hooper would, no doubt, be found to be of great practical value.

Until Mr. Hooper introduced his peculiar process of insulation by pure and protection by vulcanised india-rubber the difficulty of applying the material in such a way as to avoid a chemical action setting in with the copper, had been the greatest drawback to its employment. This difficulty has now been got over, and it remains for time to show if india-rubber, which in air is not more durable than gutta-percha, behaves equally well under water. If so, this material may have an important future in submarine work.

In gathering data for the establishment of the merits of his system, Mr. Hooper has made a series of quantitative measurements of the electrical conditions of a compound india-rubber covering under different degrees of heat. Thus, a coil of 3,394 yards was tested at different temperatures, between 14 deg. and 100 deg. cent., at intervals of 5 to 6 deg. The result of the series is the establishment of a constant for calculating the effects of temperature on the insulation of india-rubber. This constant is found to be equal to an increase of the conducting power of the 0.1034 part for every 1.66 deg. cent. (= 3 Fahr.)

Mr. Hooper does not exhibit the apparatus which he proposes to use for making joints, nor does he state whether it would be applicable at sea, where it is sometimes necessary to make a joint between deep-sea and shallow-sea ends in a small boat.

Three different types of sheathing have been employed for the protection of deep-sea cables. The first of these is that of the Dover-Calais cable, in which the insulated core was first spun round with hemp, and over this with a close armature of ten No. 1 iron wires laid on spirally. This cable, laid in 1851 and still at work, was the first iron-covered cable made. It has been taken as the type for many deep-sea lines, and for almost all the shallow-sea cables which have been laid down subsequently. The second type is that of the Toulon-Algiers cable, laid in 1860, in which the specific gravity was first diminished by spinning hemp round each of the wires used in the sheathing. The core was first covered with hemp and then with ten No. 14 steel wires, each separately served. This is the type adopted with success by the scientific committee for the Atlantic cables. The third type is that in which the hemp serving of the core is covered by a tape, plaiting of hemp, or laminous sheathing of copper or zinc. This form was tried for the Oran-Carthagena in 1863, and the Bona-Bizerta in 1864. Could sufficient strength be given to a cable constructed upon this type, a material advantage would be found in the freedom it would have from rapid oxydation of the outer covering. To remedy this drawback in iron-covered cables, Messrs. Bright and Clark proposed a method of external protection by means of a bituminous compound, which they adopted in the Persian Gulf cable, laid in 1863. The cable, strengthened in the usual way with an armature of stout iron wires, was served, on the exterior of this, with two coatings of Russia hemp, and then passed through a melted mixture of pitch, tar, and silica, forming upon it a coating impervious to water. Messrs. Siemens have attempted to preserve the iron wires by converting them into the negative pole of a galvanic element, the positive pole of which is gradually destroyed. They have done this by covering up the iron armature in hemp, and this in an outer laminous sheathing of zinc. An electrical action is set up between the two metals, the zinc being dissolved and the iron left unaffected as long as any of the zinc pole lasts. The idea is very ingenious, it was first suggested about 30 years ago by Dr. Davy for the preservation of iron buoys in sea-water, and has very recently been advocated. as a method of sheathing iron ships, but has not yet been employed to any extent.

W. T. Henley, North Woolwich, shows a case of cable samples with sections. In addition to the cables shown in the London Exhibition of 1862, we have the following:—

Cables Laid Conductors Length in Miles Weight in Tons
Denmark 1863 4 12 134
Norway 1863-6 1 50 177
Persian Gulf  1863-6 1 1,615 690
Ramsgate 1864 6 23 207
Italy and Turkey 1864 1 61 186
England and Ireland 1865 7 26.5 307
Prussia and Sweden 1865 3 55 405
Wrexford 1865 4 17.5 321
River Plate 1866 3 30 473
Cooke's Strait 1866 3 53 440
Behring Sea 1866 1 601 784
Norway 1866 1 7 10
England and Hanover 1866 4 240 2,465

Siemens Brothers, London, also show a case of cable samples, amongst which are lengths of heavy as well as of light cables. A fine shallow-sea cable is shown by these manufacturers, made by them for the Egyptian Government. It consists of six separate conductors, each of one copper wire, 1-6 mm. diameter, covered to 5.5 mm. with gutta-percha, the whole being protected with tarred hemp and 16 galvanised iron wires, each 5 mm. diameter. For war purposes Messrs. Siemens show a cable of which considerable lengths were made for the Austrian Government during the late war with Prussia. It consists of a single conductor, a strand of three steel wires, 0.75 mm. diameter, covered with gutta-percha to 3.5 mm., and protected with hemp and copper sheathing. As a further protection, sometimes the exterior of this cable is taped, and painted with white lead.

Rattier and Co., Bezons, show various specimens of submarine cables insulated with gutta-percha and protected by hemp and iron wires. The first cable made by Messrs. Rattier for submarine purposes was that laid in 1859 for the semaphore telegraph on the coast of Brittany, and which is still at work. Messrs. Rattier likewise show specimens of the cables, with one and two conductors, made for the colony of Senegal in 1861, and for Cochin-China in 1860 and 1864.

S.E. Morse, New York, exhibits a model to illustrate a new method of laying and picking up submarine cables. Mr. Morse is a brother of the celebrated inventor of the recording apparatus which bears his name. In order to overcome the inconvenience of not being able to raise a deep-sea cable to the surface at any point, Mr. Morse proposes to provide special points of slack, which can be raised to the surface and underrun. The way in which he proposes to do this is by laying the cable to a certain distance, where a second ship supports it, whilst the cable ship continues the paying out. When the ship is clear away, and the cable considered to be well down upon the ground on each side, the second ship is to proceed in a straight line, at right angles to the direction of the line, easing it down gradually at the end of a rope or chain, to which a buoy is afterwards to be attached. The buoy is intended to facilitate the refinding of the cable if it should be wanted. The writer thinks that Mr. Morse has scarcely been well advised in selecting a buoy for the purpose in question, because it would, in all probability, be hauled up on board some passing ship, and thus not only the trace lost, but possibly also the cable, if it were a light one. The buoy might, it is true, be held by the chain at a distance underneath the surface, and be dragged for by a line between two ships; but the writer doubts if Mr. Morse will succeed in finding a chain sufficiently durable to anchor a buoy in a depth of a couple of miles or so in mid-ocean. To lay a cable in this way would cost a great deal of extra line; but it would at least leave slack—a most essential point for the successful laying—and would increase the chances of being able to haul the cable on board without breaking it.

Bibliography:
[Sabine, Robert]: Reports on the Paris Universal Exhibition, 1867, Vol IV, pages 521-530.
London, HMSO.

Copyright © 2007 FTL Design

Last revised: 30 November, 2008

Return to Atlantic Cable main page

Research Material Needed

The Atlantic Cable website is non-commercial, and its mission is to make available on line as much information as possible.

You can help - if you have cable material, old or new, please contact me. Cable samples, instruments, documents, brochures, souvenir books, photographs, family stories, all are valuable to researchers and historians.

If you have any cable-related items that you could photograph, copy, scan, loan, or sell, please email me: billb@ftldesign.com

—Bill Burns, publisher and webmaster: Atlantic-Cable.com