Escher.gif (426 bytes)

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

1915: How submarine cables are made,
laid, operated and repaired
Commercial Cable Company


Although it is more than sixty-four years since oceanic telegraph communication was established, the first submarine telegraph line being laid between Dover, England, and Calais, France, in 1850, and although there are now (1914) upwards of 322,000 nautical miles of submarine cables in operation throughout the entire world, the great majority of the public seems to know very little about the making, laying and operation of cables. The general belief is that a submarine cable is very large in diameter and contains numerous electrical conductors; some people even suppose that a submarine cable is a sort of pneumatic tube through which the original message is passed. Submarine telegraphy has often been described in scientific books and publications, but the technicality of such descriptions leaves nothing but a vague and confusing comprehension to any but students of the science. The purpose of this paper is therefore to impart a general knowledge of "How Cables are made; How they are laid; How they are operated; and How they are repaired."


It was not until the value of gutta percha as an insulating material became known that submarine telegraphic communication was seriously entertained and it will no doubt be interesting to devote a little space here to a description of this valuable product without which submarine telegraphy could probably not have developed so extensively as it has during the past three decades.

Gutta percha is the gum product of certain wild growing trees which are found in the Malay and Sunda Archipelagos. The gutta percha used in the manufacture of submarine cables is of a refined character, all impurities being removed therefrom. By the purification process the weight of gutta percha in its natural state is decreased from 25% to 50%. It is considered when used in this purified condition, to be practically indestructible so long as it is kept submerged in water of a cool temperature. In fact this theory has been practically proved recently when a submarine cable which had been lying at the bottom of the ocean for over thirty years was brought to the surface during a repair and the gutta percha found in as good condition as the day the cable was first laid. No deterioration had taken place.

For years after gutta percha trees were first discovered to have commercial value there was reckless destruction of them. It was then the practice to cut down the trees to get the sap from them and it estimated that not less then 26,000,000 gutta trees were destroyed annually in Borneo for that purpose. When it is considered that takes from 25 to 30 years before a gutta tree will yield any valuable sap, it can easily be understood what effect this wholesale destruction of the gutta tree has had upon future market supplies and prices of submarine cables.

Tapping gutta percha tree

Gutta percha trees are evergreen and reach a height of between 60 and 70 feet. There are a number of species of these trees. The quality of gutta percha yielded by each tree varies according to species. The data of the quantity of gutta percha yielded by an adult tree is very conflicting. 2 lbs. 5 oz. have been taken from a felled tree at least 100 years old while other adult trees only yielded 11 oz. The method of bleeding trees is now different from the method in vogue years ago. Formerly it was the custom to fell a tree and make a series of incisions along its entire trunk from which the sap would run into small receptacles. This custom was followed by making incisions in trees without cutting them down, as shown in the photograph above.

The Governments of the countries in which the gutta percha tree is found now prohibit the felling of the tree. The present method of extracting the sap is very much the same as the method used in North America for extracting the fluid from maple trees. The bark of the tree is pierced at various parts and small drain tubes are inserted, through
which the sap flows into receptacles. These drains are allowed to remain in the trees throughout the entire year. The trees are said to yield more sap after the rainy season. The color of the sap when taken from the tree is milky white but soon changes into a dark brown. The gum collected from the trees is made up into bricks and taken to the coast by the natives, where they trade it, its value being based upon its weight. It was not long before the natives acquired some of the business instincts of modern civilization and bricks of gutta were found to contain stones, iron implements, etc., concealed in them to give them artificial weight.

The conductor, through which the electric impulses are transmitted, is composed of the purest quality of copper. It varies in size according to the length of the cable and speed required by the operating company. From all practical standpoints copper is superior as a conductor of electricity to any other metal. It is a well known fact that some metals are better conductors of electricity than others. Copper offers less resistance to the transmission of the current than any other metal except one or two like platinum, which are too expensive to be practicable. This resistance varies in proportion to the size and weight of the conductor. In other words the electric impulses will not pass through a copper wire weighing 100 pounds per nautical mile with the same freedom and speed as it would pass through a copper wire weighing 200 pounds per nautical mile and so forth. In some of the modern Atlantic cables over 700 pounds of copper per mile and 360 pounds of gutta percha are used.

It should be borne in mind that cables are lying at the bottom of the ocean in depths ranging from two to three thousand fathoms, or between two and three miles deep, and as the lifting of a cable from such a depth involves a great strain upon it and all the materials used in its construction, it will be understood that flexibility must be seriously reckoned with. While there is a good deal of flexibility in copper yet a single copper wire would not have the same flexibility or tensile strength as a number of smaller copper wires with an aggregate weight no greater than the single wire. Consequently in order to provide this flexibility without affecting the conductivity it is customary to use a number of smaller copper wires making up the desired weight of copper per mile. This strand of copper wires is used as one conductor just as a number of fine wires are used in flexible corded electric light conductors instead of a single wire of the same conductivity. If this flexible cord contained just one wire instead of a number of small wires, it would not stand handling or twisting without soon breaking.

The copper conductor covered with the gutta percha insulating material is called the "core" of the cable. In the early experiments made with gutta percha some difficulty was experienced in making it adhere to the surface of the copper conductor. In the present construction of cores a material called Chatterton's Compound is used to overcome this difficulty. The wires are first heated to cleanse the surface and they are then steeped in Chatterton's Compound, which is a chemical mixture of tarry materials. This then furnishes a base upon which the gutta percha will adhere. The gutta percha is applied hot and is squeezed on to the copper conductor by a machine through which the copper strands of wire pass, continuously enveloping the wire throughout. Some of the cable manufacturers apply the gutta percha in two or three coatings. It is claimed by some that the latter method insures safer insulation, inasmuch as an imperfection in the first application of the gutta percha is covered by the second or by the third.

The conductor in present-day cables consists of a large central copper wire surrounded by a number of small copper wires or copper ribbons, the total making up the required weight of copper and conductivity per mile. Another advantage of the stranded type of conductor over the solid wire is that if the solid wire should contain a mechanical imperfection or should break for other reasons inside of the gutta percha insulation, electrical continuity would cease, but in the case of a stranded conductor even if one wire should break the remaining wires would still conduct the current, as it is not likely that all of the copper wires of a stranded conductor would break in the same place.

The quantity of gutta percha used for insulating submarine cables also varies according to the length of the cable and size of the copper conductor in order to obtain the proper electrical results. The core is manufactured in lengths of approximately three miles and is coiled temporarily on drums. These lengths of core are later joined together during the process of completing the manufacture of the cable. The jointing of these lengths of core is of the greatest importance. It is done by hand and requires skillful workmanship. If for example any dust or gases are allowed to remain or to form in the gutta percha whilst making such a joint it may mean the loss of thousands of dollars to the company owning the cable because it is possible that this weakness may not become apparent until the cable is submerged in the sea and thus placed under great pressure when the most minute impurity or gas bubble in the joint would manifest itself and cause faulty electrical continuity. Such faults are most difficult to localize and sometimes remain in the cable for years before they can be found and removed, in the meantime causing great trouble in the operation of the cable and loss of money. Furthermore the deep-sea cable jointer must be a man of temperate habits and in good health. While it may seem almost inconceivable, yet numerous joints made by skilled but intemperate or unhealthy jointers have proved faulty through what was believed to be the injurious exudations from the pores of the fingers - this will give some idea of the extreme delicacy and importance of perfect jointing.

One of the greatest enemies of submarine cables is the teredo, an aqueous worm. These worms in some parts of the sea particularly near land are very abundant and if they can work their way through the outer covering of the cable on to the gutta percha insulation they not infrequently interrupt communication by boring minute holes through the gutta percha to the copper conductor, allowing an escape of electricity and causing faulty transmission of the current. To guard against this the core of present day cables is covered with brass tape. This brass tape is impervious to the assault of the teredo. The completed core with the brass taping is covered with jute yarn, a coarse hemp, steeped in a tarry preservative. This jute yarn is wound round the core to serve as a bedding for the outer protecting wires. After several servings of the jute yarn have been applied to the core, the whole is then covered with galvanized iron wires which vary in number and thickness according to the depth of water the cable is to lay in. A cable laid in deep water must necessarily be of lighter weight than one laid in shoaler water, because if it were too heavy the strain of raising it in very deep water would he so great as to render it impossible to recover the cable for repair purposes. Furthermore in the deep water there is very little to injure submarine cables. The cable rests in a soft bed of decomposed shell matter known technically as Globigerina Ooze. Cable recovered from places where it has been lying in deposits of this kind has been found to be in almost a perfect condition after thirty years' submersion.

Where the cables approach the shore and are subject to the effect of motion in the water it is necessary to guard against deterioration and corrosion by protecting the core with a heavier armouring. Thus the cables vary in thickness from one inch in diameter, deep-sea cable, weighing about 2.5 tons per mile, to four and a half inches in diameter, shore end ice cable, weighing approximately 60 tons per mile. Nearly all the shore end types of cable have two sheathings of galvanized iron wire as an additional protection against chafing and corrosion. The inner sheathing is made of small iron wires and the outer sheathing of heavy wires. The iodine contained in the sea weed found in shore deposits, has a corrosive effect on the sheathing wires.

(Left) Shore end type of cable with various layers stripped

(Top right) Shore end cable

(Bottom right) Deep-sea cable

All long submarine cables have only one core, it being electrically impossible to work submarine cables with two conducting wires for great distances because the current which passes through one wire induces a current into the second wire, thereby setting up interference with and mutilation of the signals. The longest cable with two cores in it is the cable of the Commercial Cable Company between Canso, Nova Scotia, and Rockport, Mass. This cable is about 528 nautical miles long.

The outer protecting wires of submarine cables are steeped in tar and are covered with tarred tape or yarn. This is to protect them as far as possible against the corrosive effect of any chemical deposit over which they may lie. The number of outside protecting wires also varies according to the depth of water in which the cable is laid. This number must be decided upon scientific principles. It must always be borne in mind that a cable has to be so constructed that it can be recovered again from the bed of the ocean. It mint therefore not be made either too stiff or too flexible, and the size and number of wires and the shape of their windings arc engineering problems which have to be carefully calculated in the manufacture of the cables.

During the process of manufacture daily tests are made by experts of the electrical and mechanical fitness of the manufactured cable and the completed cable is then coiled in large tanks, where it is kept under water until the time arrives for trans-shipment to the cable steamer which is to lay the cable.

At the present time nearly all the oceanic telegraph cables are manufactured in Europe. The principal submarine cable manufacturing plants are the Telegraph Construction and Maintenance Company at Greenwich, England; the India Rubber, Gutta Percha and Telegraph Works Co., at Silvertown, London; Siemens Brothers & Company, at Woolwich, London, and the Norddeutsche Seekabelwerken of Nordenham, Germany.

The time occupied in the manufacture of cables depends upon the number of machines employed by the manufacturer, but a single set of machinery is capable of turning out about 5 or 6 miles of deep sea cable per day.

The works of the cable manufacturers are always located on the banks of a river or close to docks where the cable can easily be trans-shipped into a cableship. When the cable is nearing completion the cableship is brought alongside of the wharf to receive the cable. The largest cableship in the world at the present time is the "Colonia," which is owned by the Telegraph Construction & Maintenance Company, London, England, and was used in laying the trans-Pacific cable of the Commercial Pacific Cable Company. The cableship "Colonia" is 500 feet long, 56 feet wide, and has a displacement of 7,976 tons, and a carrying capacity of 4,000 nautical miles of cable. She is equipped with four cable tanks. The cable is coiled into these tanks with almost as much exactness and precision as thread is coiled around a spool. Before the cable is placed into the tanks of the ship all the various types to be used are joined together and the cable is stored into the tanks with regard to the manner in which it is to be laid out; that is to say, the shore end will appear on the top of the tank, as that is to be laid first. Then come the various intermediate types down to the deep-sea type; and again, as the ship is approaching the shore of destination, the cable becomes heavier as the water becomes shoaler, until the end is landed.


The laying of long submarine cables is not an easy matter, nor is it an inexpensive operation. To safeguard the millions of dollars required to manufacture and lay a submarine cable every possible precaution must be taken to prolong its life and at the same time to assure the company of being able to recover the cable for repair purposes. It is well known that the contour of the bottom of the ocean varies similarly to that of dry land, that is, it has hills and valleys and plateaus. It is therefore of the utmost importance before a cable is laid to know the contour of the bed of the ocean, so as to avoid suspending the cable between two hills, where it would hang in a festoon and soon become chafed through by its own weight. The question has often been asked, "Does a submarine cable sink to the bottom of the ocean?" The answer is that unless it rested on the bottom all the way across its span, it would only be a matter of a very short time before it would be chafed through and communication interrupted. In order to find a resting place for the cable which will enable it to lie without strain throughout its entire length, a survey of the bottom of the ocean is first made. This survey consists of a series of deep-sea soundings which not only furnish the depth of water, but also produce specimens of the bottom of the ocean and the temperature of the water, both of which are important factors in the laying of submarine cables, because if by chemical analysis it should transpire that there is any mineral deposit on the bottom of the ocean which would injuriously affect the cable, or if the temperature should show that there is volcanic action at certain places, these places must be avoided.

Specimens of deep-sea sounders

The present-day type of deep-sea sounding machine was invented by the late Lord Kelvin, formerly Sir William Thomson. Lord Kelvin contributed more towards the science of cable telegraphy than any other person. This sounding machine is composed of a large drum containing thousands of feet of pianoforte steel wire. The machine is driven by electricity and has a gauge registering the number of fathoms paid out. The sounders (weights) vary in pattern. The form of sounder invented by Captain (now Rear Admiral) Sigsbee, U.S.N., probably possesses all the essential features of the others and may be taken as a typical one. It consists of a central tube fitted with valves at top and bottom, and three smaller tubes fixed beneath the central one. As the sounder descends the valves of the central tube open upwards, and the water rushes through. On reaching the bottom, both valves close, and a sample of the bottom water is retained. The three smaller tubes sink into the ooze, and bring specimens of it to the surface for chemical analysis. In order to increase the speed with which the sounder sinks, shot weighing from 30 to 60 pounds are slipped over it, becoming automatically detached on reaching the bottom and remaining there.

The temperature of the bottom water plays an important part in submarine telegraphy. A low temperature increases, at the same time, the conductivity of the copper wire and the insulation of the gutta percha covering, and is therefore the most suitable for cables. Faults are usually located by the electrical resistance of the conductor. It is known that a conductor when in normal condition gives a certain resistance per nautical mile at a given temperature. If the cable is broken so as to expose the conductor, one has only to divide the resistance obtained from the tests by the resistance per nautical mile in terms of the same temperature in order to arrive at the approximate distance of the fault from the testing station. A knowledge of the bottom temperature is therefore indispensable for accurate results, and in taking soundings a thermometer is always attached to the wire a short distance above the sounder.

Sea water follows a different law from fresh water in the relation between its density and its temperature. The density of fresh water increases with decreasing temperature down to 39° F., so that until this temperature is reached the coldest water is always sinking to the bottom. After 39° F., fresh water begins to expand again, the coldest water remains at the top, and ice forms there instead of at the bottom. This law plays a very important part in the economy of nature. Our fresh water lakes and ponds would otherwise be frozen into a solid mass in winter, and all the fish would be destroyed. As it is, the ice on the surface prevents any further loss of temperature from evaporation, and the process of freezing is greatly retarded. In the case of sea water, contraction continues down to its freezing point, 25° F., and the coldest water is always at the bottom.

For this reason, when taking a sounding, a maximum and minimum thermometer give the correct top and bottom temperatures. In great depths, an ordinary thermometer cannot be used, as the pressure causes an error 8° to 10° F., and has been known to cause implosion of the bulb. For deep soundings a special thermometer, the Miller-Casella, is used. The bulb of this instrument is enclosed in an outer bulb filled three-quarters full with alcohol. Before sealing this outer bulb, the alcohol is warmed, so as to expel some of the air. Between the two bulbs a cushion is thus formed, which takes up the pressure, and the inner bulb remains unaffected by it.

As low temperature as 27° F. has been taken in the South Atlantic in the neighborhood of icebergs. Another sounding of 2,900 fathoms further north in the same ocean gave a bottom temperature of 32° F., the last 1,000 fathoms being described as absolutely glacial. In the second 1,000 fathoms the temperature rose 36.5° F., and in the next 500 fathoms to 40° F. Four hundred fathoms seems to be the limit to which the heat of the sun can penetrate. The most marked difference is at the top, where the surface temperature was in one instance 83.5° F., while at only 20 fathoms depth the thermometer stood at 64° F.

The light rays of the sun penetrate only a short distance beneath the surface of the sea, and as far as extraneous illumination is concerned the ocean abysses remain in absolute darkness. Some deepsea fish, however, have two parallel rows of small circular phosphorescent organs running down the whole lengths of their bodies, so that they resemble ships at night with double rows of shining portholes. It is thought possible by certain naturalists that portions of the sea bottom mar be as brilliantly illuminated by this kind of light as the streets of a large city after sunset.

But the conditions which obtain at these great depths are very unfavorable to animal life. Apart from the glacial temperature, the pressure for every 1,000 fathoms is one ton to the square inch, or 160 times greater than the atmosphere in which we live. At 2,500 fathoms it is twenty times more powerful than that of the steam in an average locomotive boiler. As late as 1880 a leading zoologist explained the existence of deep-sea animals at such depths by assuming that their bodies were composed of solids and liquids of great density, and that they contained no air. This, however, is not the case with deep-sea fish, which are provided with air-inflated swimming bladders. If one of these fish in pursuit of its prey ascends beyond a certain level, its bladder becomes distended with the decreased pressure, and carries it towards the surface in spite of all its efforts. This kind of misadventure may be described as falling upwards, and victims to it no doubt meet a violent death soon after leaving their accustomed level, and long before their bodies reach the surface in a distorted and unnatural condition. Even ground sharks brought up from a depth of only 500 fathoms die before they gain the sea level.

Landing of cables

But to return to the laying of submarine cables. On reaching the place selected for the landing of the cable, the ship approaches as close to the shore as possible and, letting go anchor, prepares to land the shore end. By some companies this is done by means of rafts, in others a couple of spider-sheaves, or large VV-shaped wheels in light iron frames are sent ashore and fixed by sand anchors some 60 yards apart. Hauling lines are paid out from the ship, reeved through the sheaves and brought back on board again. One end of this continuous line is attached to the cable and the other to the picking-up gear. The engines are then set in motion, and the cable is dragged slowly out of the ship towards the shore. As it goes, large inflated India rubber buoys or wooden casks are lashed to it every 50 or 60 feet, to keep it afloat and prevent the damage which would result from it being dragged along the bottom.

When sufficient cable has been landed, the length on shore is laid in a trench which runs from low water mark to the cable hut, and the end is inserted through a hole in the floor. Testing and speaking instruments are set up in the hut, which is occupied day and night during the laying by the electrician in charge and his assistants. When a satisfactory test has been taken the ship gets slowly under way.

The scene on the deck of a cable ship during the commencement of paying out is full of interest. The cable, being made fast on shore, drags itself out as the ship moves forward. At first it rises slowly from the tank, passing along a series of guiding troughs to the paying-out machine, round the drum of which it runs several times. Between this drum and the sheave at the stern, by which the cable leaves the ship, stands the dynamometer. This machine shows the strain to which the cable is subjected, and plays an important part in cable laying. As the water deepens the weight of cable in suspension tends to make it run out more quickly. This tendency is counteracted by increasing the weights on the brake arms of the paying-out machine. If the water shoals, the weight of the cable outboard is considerably reduced, the strain lessens and the brakes have to be eased, in order to allow it to run out at the same speed as before. The amount of cable in suspension varies according to the depth and rate of paying out. In 2,900 fathoms, with the ship steaming at 8 knots per hour, no less than 23 miles of cable are in suspension in the water. Two and a half hours are occupied in this case by any particular point in the cable, from the time of leaving the ship to touching the bottom.

One of the most interesting spots on board of a cable ship during laying is the testing room. Here, in front of a table, glistening softly with the polished ebonite and bright brass terminals of various testing instruments, sits an electrician, watching a spot of light as it sways to and fro on a graduated scale. This spot is a reflection from the mirror of the galvanometer, and the swaying movement is caused by the induced currents set up in the coiled cable by the rolling of the ship. At the end of every fifth minute the spot gives a kick sideways on the scale, and the electrician duly notes its magnitude. This kick is caused by a signal from the shore, and proves that the continuity of the conductor is still preserved. As the cable leaves the ship and sinks into the almost freezing temperature of the ocean bed, the insulation improves and the spot of light gradually works down towards zero. When, on the other hand, the deflection grows larger and the spot shows an inclination to steal off the scale, something is wrong and a careful test must be taken. Should this prove that a fault exists, the ship is stopped, the cable cut and, if the fault is near, picking up is commenced. If the fault is some distance away the cable is buoyed, and the ship steams to the locality of the faulty portion.

Supposing, however, all goes well, the whole section is payed out and buoyed, and the ship steams to the second landing place. Here the shore end is landed in the same manner as before, and the cable is payed out up to the buoy. When the end of the first section has been hauled on board the splicing operation commences. This consists of, firstly, a joint between the cores, and, secondly, a splice between the outer sheathing. A skilled jointer and his assistants cleans and solders together the ends of the two conductors. Then, drawing down over this joint the gutta percha covering from either side, he applies two or three more coats of the same material. As heretofore stated no air holes must be left between these different coverings, as the enormous pressure at the sea bottom might burst them and render the cable faulty. The splice in the sheathing wires is performed by the cable hands, and is a much less delicate piece of workmanship.

Under the most favorable conditions cable laying is anxious work for those in charge of the operation. At any time during the paying out - which may last, with a long section, some ten or fourteen days - a storm may arise or some mishap may occur on board, which results in losing the cable in a depth of over 2,000 fathoms. In such a case the date of its recovery cannot be predicted. It may be in three or four days, it may be in as many weeks or months. Every precaution is therefore taken against such accidents. Buoys are slung in the rigging ready for slipping into the water, buoy ropes and grappling ropes are coiled where they can be paid out at a moment's notice, and means of signaling with a bell in the engine room are placed in convenient positions at the top of each cable tank.

Should the cable break outboard, and be lost in spite of these precautions, a mark buoy is immediately lowered to guide the ship in grappling operations. Dragging is then begun at right angles to the line in which the cable lies. Should the dynamometer, under which the grappling rope runs, show a steady rise in the strain, the cable is evidently hooked and heaving up commences. As soon as the grapnel reaches the bows with the bight of the cable on one of its prongs, the two sides are firmly secured by lines from the ship, and the bight is cut. After the two ends have been tested from the testing room the short length is abandoned or buoyed, and the other is spliced to the cable in the tanks, when paying out is once more resumed.

The rate at which submarine cables are laid varies according to the depth of water. In shallow water they are laid at a rate of approximately eight miles an hour, but in the deeper water the rate is necessarily slower, sometimes not exceeding five miles an hour. Sufficient amount of slack must be laid out so that every part of the cable will find a resting place. If the cable were laid too taut it would be impossible to lift it without severely straining and damaging it.

Contour of part of the Pacific Ocean bottom

Above is an illustration of the contour of the bottom of a portion of the Pacific Ocean. During the very exhaustive soundings taken in the Pacific Ocean when the route for a trans-Pacific cable was surveyed a depth of over 5,000 fathoms (over 5 miles) was discovered off the Island of Guam, Ladrone Islands. The pressure at this great depth would be about 5 tons per square inch, and it would have been practically impossible to maintain and repair a cable laid in such a depth. It was therefore necessary to make a wide detour of this "deep" and approach the island from another direction. This was done after further surveys were made and a path found with submarine slopes of gradual contour. The Commercial Pacific Cable Company's cable is lying in depths of over 3,000 fathoms (3 miles).

Before the carrying capacity of present-day cable ships was attained the operation of laying a long cable had to be done either by two vessels, each one starting from opposite ends, and meeting in the center, or it required two trips of one vessel. A cable-laying ship like the "Colonia," heretofore described, has sufficient carrying capacity to lay a cable approximately 4,000 miles in length.

During the laying of a cable careful nautical records are kept by the navigating officers of the vessel and the engineer in charge of the cable expedition, showing as near as possible the position in which the cable is being laid. When the entire cable has been finished a copy of these records and charts showing the route of the cable are furnished to the company owning the cable for future reference in locating the position of breaks in the cable.

After the laying of the cable is completed the electrical engineers at both ends of the line also take final tests, ascertaining the electrical measurements of the cable for future use, and the records of these tests and all data relating to the cable are handed over to the company owning the cable.


The cable having been brought into the office and final tests completed, it is turned over to the company's operating staff for use in the transmission of cable messages.

Photograph of an original message sent over the first Atlantic cable, August 17, 1858, the day after the cable was laid. The actual time occupied in sending the above message was 25 minutes. A similar message can be transmitted over the Commercial Cables in less than a minute.

One of the first theories advanced by electrical engineers for the operation of long submarine cables was that on account of their great unbroken length it required a battery of high voltage to force the currents through it. This theory, however, was opposed by the late Lord Kelvin (formerly Sir William Thomson), who claimed that a long cable could be operated by a current generated in a lady's thimble, which was later proved correct. The first trans-Atlantic cable was operated by means of large induction coils and batteries with a potential of 500 volts or more. It is believed by many that this high battery power broke down the insulation of the cable. It only worked from the 17th August to the 20th October, 1858, and carried 732 messages. It was impossible to send any signals through the cable after that date and it had to be abandoned.

In April, 1858, Sir William Thomson (Lord Kelvin) invented the Mirror Galvanometer, an instrument of high sensitiveness, which not only was the first instrument by means of which long submarine cables were successfully operated, but which made it possible to increase the speed of transmission five or six times greater than any other instrument.

The Reflecting Galvanometer, commonly known as the Mirror, was composed of a coil of very fine wire, in the center of which a very light mirror of about 0.25 inch diameter, with tiny magnets cemented on its back, was centrally suspended by a silk fiber. A beam of light thrown on this mirror was reflected from it onto a scale several feet from the instrument. The passage through this coil of the received positive and negative currents tended to move the magnets to right or left, and the reflections of the mirror's movements were read from the scale. When there is no current passing through the cable the beam of light thrown upon the scale remains at rest at a zero point on the scale. As soon as a current is passed through the cable, no matter how weak, it causes the mirror of the galvanometer to move to one side, carrying the reflected spot across the scale. A double-current key is employed for sending the signals - one of which sends negative currents and the other side sends positive currents to line. The positive currents represent the dots and the negative currents represent the dashes. Hence an operator sending the letter "A" would first press the left key representing a dot, and then the right key, representing a dash. These signals act upon the Mirror Galvanometer as explained above, viz.: the light moves from the zero point to the left side of the scale (representing a dot), and then immediately over to the right side of the scale (representing a dash), corresponding with the manipulation of the keys at the sending end.

Operators soon became skilled in reading these movements of the light. While this system of reception proved satisfactory, there was one element of weakness about it which required improvement, namely, that once the flash of light representing the signal had passed, there was no record left to prove what was received, and consequently letters, or parts of letters, were sometimes dropped out by mistake. This led to the invention of the Siphon Recorder by Lord Kelvin, in 1870. This was the first instrument used on long cables that recorded the received signals. This instrument is now used at nearly every cable office in the world except a few places in the West Indies, which still use the Mirror Galvanometer.

In the Siphon Recorder a light rectangular coil of very fine wire is suspended centrally between the poles of a powerful horseshoe magnet. A fine glass siphon, dipping into an ink-well, is suspended in front of this coil and attached to it by silk fibers. A narrow paper tape is drawn in front of the lower end of the siphon, and the movements of the coil, communicated to the siphon by the silk fibers, are permanently recorded on the paper tape. As the coil moves from the right to the left, or vice versa, actuated by the electric impulses sent into the line, it carries the siphon with it, and it will thus be seen that the siphon, through which specially prepared ink is constantly flowing, will leave a record of the signals; that is to say, if the letter "A," consisting of one negative and one positive impulse is sent into the line, the siphon recorder will reproduce these impulses on a paper tape used to record the signals, the siphon moving to one side of the tape when actuated by a positive current representing a dot, and to the other when actuated by a negative current representing a dash. So long as no current is passing through the line the siphon remains stationary in the center of the tape. The tape is kept constantly running under the siphon by a small motor, and to avoid excessive friction between the glass siphon and the tape a vibrator (invented by the late Charles Cuttriss, electrical engineer of the Commercial Cable Company) keeps the siphon constantly vibrating thus avoiding continuous contact with the paper. Mr. Cuttriss was also the inventor of an automatic transmitter to replace hand-sending. A paper tape is first perforated by hand, each perforation representing a dot or a dash. This perforated paper is then passed through the automatic transmitter, the perforations permitting contact to be made between two contact points, which send either positive or negative currents to the line just as in the case of hand-sending. The advantage of this form of sending is that it insures uniformity of signals and much higher sending speeds than is possible by hand. Mr. T.J. Wilmot, deceased, who was also an electrical engineer of the Commercial Cable Company, invented another automatic transmitter. The transmitters of both Mr. Cuttriss and Mr. Wilmot are still in use in all parts of the world.

Signals as they appear on
a slip perforated for use on the
Cuttriss automatic transmitter
Signals as they are received
at the distant end

Since the invention of the Siphon Recorder no one has ever invented a mode of reception of long cables signals varying in principle to that of the recorder until Mr. John Gott, chief engineer of the Commercial Cable Company, invented, in 1912, a system by which ordinary Morse can be applied to the cable.

One of the most remarkable and valuable devices used in long cable transmission is the invention by which it is possible to duplex, or send and receive two messages at the same time over one wire. Although the invention of duplex telegraphy, as applied to submarine cables, was made in 1873, the first transatlantic cable was not duplexed until 1878 (by Dr. Alex. Muirhead and H.A. Taylor). All long cables are now duplexed. The principle of duplex telegraphy is briefly as follows:

A signaling current sent from the battery A enters the circuit at the apex B and divides, half going to the line C and the other half going to an artificial line D. The artificial line is an arrangement constructed so as to represent an exact counterpart electrically of the actual cable. Thus the current which passes into the artificial line D meets the same resistance and other obstacles as it meets in the actual cable. E represents a Siphon Recorder on which the signals received from the other end are recorded. It will therefore be seen that signals sent from A divide at B and meet the same obstacles at C and D. Consequently no current passes between C and D, and the Recorder at E will not be affected by the current sent from battery A. Therefore the Recorder at E, unaffected by the signals sent out from its own station, is responsive to signals arriving from the distant end.

In the operation of long submarine cables a battery of low voltage is used. The Commercial Cable Company does not use more than 60 volts for operating its main Atlantic cables. There are a great many technicalities in the operation of these cables which would require too much space to describe here, but the foregoing is a general description of the manner in which cables are operated. The Commercial Cable Company has always led its competitors in enterprise and in the adoption of inventions of practical value for improving submarine cable working. Thus it was the first Atlantic Company to adopt the Brown & Dearlove cable relay, an instrument which made it possible to couple up two sections of cable and to send messages from Europe to America, or vice versa, without human translation between the sections. Prior to the adoption of this instrument messages had to be manually relayed at the intermediate stations.

The Commercial Cable Company has also adopted the recent invention of E.S. Heurtley, by means of which the quality of signals and speed of operation are considerably improved. This is an instrument of greater sensitiveness than the Siphon Recorder. One of the drawbacks to the employment of highly sensitive instruments is the fact that such instruments not only magnify the feeble signaling current but also magnify all the vagrant electric disturbances in the line. It is well known that the earth is a huge reservoir of electricity, and that great electric waves are constantly passing through the earth. These waves at times become so strong that currents are induced in the cable which seriously affect the signals.

The new cable station of the Commercial
Cable Company at Far Rockaway, L.I.
Seven Atlantic cables run into this office

The Commercial Cable Company's Atlantic cable stations at Far Rockaway, L.I., Canso, N.S., and Waterville, Ireland, are considered to be three of the finest and most perfectly appointed cable stations in the world. They are fitted with the latest and most improved cable apparatus and with everything conducive to the efficiency of the Commercial Cable Company's system. At Canso, N.S., and Waterville, Ireland, the company has provided living accommodations for its employees, the married men being furnished with very comfortable private cottages and the single men with bachelor quarters. There is also a mess building for single men and a club house, library and entertainment hall at each of these two places.

The station building and equipment at Far Rockaway is the latest and most up-to-date of the three.


The first indication that a cable is broken or faulty is the failure of the receiving apparatus to properly record incoming signals. Cables are interrupted either by complete severance of the conductors or by leakage in the insulation. When a conductor is completely broken there is an absolute loss of electrical continuity, but in the case of a leakage there is an escape of electric current in the cable without complete loss of continuity. These leakage faults are the most difficult to localize.

When the receiving instruments indicate that there is either a break or a fault in the line a test is immediately made from each end of the line. These tests are taken with very sensitive apparatus, constructed on the same principle as the Mirror Galvanometer heretofore described. A battery is applied through a galvanometer into the interrupted cable. An instrument containing a series of coils of very fine wire of various resistances is also employed. The instrument is so constructed that a resistance of one ohm to several thousand ohms can be attained. Several methods of testing are employed in the localization of complete or partial interruptions of cables, the most general being the Wheatstone Bridge balance. In this a galvanometer, or similar sensitive instrument, is joined up between the arms of the bridge. The circuit is arranged identically the same as the circuit described above for working duplex, except that the galvanometer takes the place of the recorder, and resistances of known and variable values take the place of the artificial line. It will be seen from the diagram illustrating the duplex circuit that a current passing into the apex marked A divides, part going to the line and the other part entering the resistance boxes. So long as the resistance in the line is greater than the resistance in the resistance boxes, or vice versa, a current applied to the apex will throw the spot of light of the galvanometer from the zero point. Therefore, the resistance in the cable, up to the point of the break, can be determined by altering the resistance of the resistance boxes until the beam of light remains stationary at the zero point, when the resistance in the line must equal the resistance in the boxes. After having made a series of confirmatory tests in this manner it is a simple matter to calculate the distance in miles from the shore. The unit of resistance is called an ohm, after the great German physicist who discovered and expounded the laws of the electric current. As heretofore explained the exact resistance per nautical mile of the conductor of the cable is known to the company. Supposing, therefore, that the known resistance per mile is two ohms, and the measuring apparatus indicates a total resistance in the broken cable of 800 ohms. The break would thus be ascertained by dividing the 800 ohms by 2, which would place it 400 miles from the shore. With this information the captain of the cable-repairing steamer is able to determine, by his charts showing the position of the cable, the latitude and longitude in which the break has occurred, and the ship proceeds to the repair. Having arrived at a point near which the cable is broken, a mooring with buoy attached is put overboard to mark the position.

Buoy put overboard by cableship to mark position of cable

The ship then steams to a convenient distance from the broken end of the cable; a heavy iron grapnel is lowered and is dragged along the bottom of the ocean at right angles to the line of the cable for the purpose of hooking it. There are different kinds of grapnels, but all have the same general features. The ordinary kind has a shank about 4 feet long with five prongs and weighs about 230 pounds. This style of grapnel is used for general work. For very deep water work a special type of grapnel is used, which cuts one side of the cable and clutches and holds the other. This is done to avoid excessive strain on the cable when lifting the cable from great depths.

Placing "Stoppers" on a picked up cable to prevent it slipping back into the sea

As the ship steams across the line of the cable the iron fingers of the grapnel rake the bottom of the sea. The hooking of the cable is indicated on the deck of the ship by an instrument called a dynamometer, which registers a steadily increasing strain on the grapnel rope. If rocks or other obstructions are encountered by the grapnel the fact is shown by the irregularity of the strain on the dynamometer. When the cable is hooked the ship is stopped and the picking-up machinery started. In due course the grapnel holding the cable appears.

Men are then lowered over the bow of the boat strapped in "bosun's chairs," and the cable is secured by chains on each side of the bight. The cable is then cut and hauled on board and connected with the ship's testing room. The end of the cable which is found intact to the shore is then chained to a mooring anchor with a buoy attached and dropped overboard temporarily. The short end to the point of fracture is picked up and stowed away. The steamer then proceeds to drag for the other end. Having secured it, and the tests indicating that it is in perfect condition, it is spliced onto a length of spare cable on board, and the ship commences to pay out toward the end which was buoyed, which in due course is reached and taken on board. The two ends now on board are tested by the ship's electrician. If these tests are satisfactory the cable is handed to the ship's jointer for the jointing of the core.

The two ends of the conductor are neatly and firmly soldered together. A small amount of Chatterton's Compound is evenly applied with a hot iron so as to leave no air spaces. The ends of the gutta percha insulation are heated with a spirit lamp and drawn down until they meet about the middle of the exposed conductor. A very thin strip of gutta percha is then softened and wrapped round and round upon itself at the joint, and is gradually worked down by the jointer with his fingers to the original diameter of the core. The joint is then placed in a trough of cold water, and allowed to remain there about 20 minutes so that it may become hard throughout. The jointer now gives way to the splicer, who places a serving of jute yarn around the core as a cushion for the armor wires.

Making a joint of the core

Replacing the armor wires

The armor wires are then put back as nearly as possible in their original lay, and a heavy tarred jute yarn is put on as tightly as possible over a distance of 15 or 20 feet, which completes the splice and the cable is dropped overboard.

Cableship passing icebergs at sea

Cables are broken from various causes. Where they approach the shore they suffer from corrosion and chafe and anchors of vessels and are occasionally broken by icebergs grounding on them and crushing them. During one repair made by the Commercial Cable Company's steamer "Mackay-Bennett" she counted as many as 100 icebergs, and to enable her to carry on her work had to tow an iceberg to sea to take it off the line of the cables.

Officers of Cable Steamer Mackay-Bennett

They have been also broken in the deeper waters of the Atlantic by submarine land slides, burying the cable for many miles. The Alaskan cable was broken by a whale, the decomposed carcass of which was found encircled by the cable when it was recovered during the repair. One of the cables in the Southern waters was interrupted by a shark's tooth, which was found imbedded in the gutta percha. The teredo or borer worm has also done considerable damage to submarine cables by boring through the gutta percha insulation. But the greatest menace to the safety of submarine cables is the steam trawler. These boats operate in great numbers on the European coasts, particularly off the Irish coast. They drag the bottom of the ocean for fish down to depths of 200 fathoms or more with great iron-shod beams trailing along the bottom. They have destroyed hundreds of thousands of dollars' worth of cable property. During the year 1908 four out of five of the Commercial Cable Company's main cables were interrupted by trawlers. The company took the matter up vigorously with the British and American Governments, which resulted in a conference being held in Lisbon, Portugal, to discuss the question of securing relief from these destructive operations. The British Government finally appointed a commission to look into the cause of each interruption off the British coast, and to take whatever steps were possible to prevent unnecessary damage to the cables.

Commercial Cable Company's
Cable Steamer Mackay-Bennett
Length, 270 feet; speed, 10 knots an hour

The operation of repairing deep-sea submarine cables is no child's play. It is work requiring sturdy, fearless manhood and skillful seamanship. Most of the breaks occur during seasons of the year when the weather conditions at sea are most severe. It is not uncommon for a cableship to spend a month or more at sea waiting for suitable weather conditions to carry on operations. The "Mackay-Bennett" been at sea as long as 90 days making one repair, and during severe gales has been blown hundreds of miles away from the ground of operations.

The above photo, showing the ice-covered deck of the cableship "Mackay-Bennett," was taken in the winter of 1912 while she was repairing a cable in the North Atlantic Ocean.

THE COMMERCIAL CABLE SYSTEM extends in unbroken continuity TWO-THIRDS AROUND THE WORLD; connecting the Continents: EUROPE, AMERICA and ASIA. The remaining one-third is covered by connecting lines.


18,588 Miles of Cable in the Atlantic.
10,010 Miles of Cable in the Pacific.
300,000 Miles of Telegraph Line reaching all important places in the United States.




GREAT BRITAIN AND IRELAND - Our own offices in London, Liverpool, Manchester, Bradford, Newcastle-on-Tyne, Bristol, Weston-super-Mare, Glasgow, Edinburgh, Dundee, Leith; all of which are in direct communication with the cable stations by special wires operated by our own staff.

FRANCE and CENTRAL EUROPE is served by our own cable landing at Havre and underground Wires to our own office in Paris.

GERMANY, HOLLAND and beyond are reached by the GERMAN ATLANTIC CABLES, which are exclusively connected with this system. The cables are in touch with the German Imperial Telegraph system and with the chief cities of Holland.

RUSSIA, NORWAY, SWEDEN, DENMARK, etc., are reached by direct wires from and to the Great Northern Telegraph Company, at London.

SPAIN, PORTUGAL, MEDITERRANEAN PORTS, AFRICA, INDIA. etc. At the Azores Islands our cables connect with a direct line to Portugal for these countries.

BRAZIL, ARGENTINE, EAST COAST OF AMERICA - At the Azores Islands our cables connect with the cable via the Cape Verde Islands to Brazil. This is the shortest route to South America.

WEST INDIES are reached by our connections, the Direct West India Cable Company, and the Halifax & Bermudas Cable Company.

CUBA IS IN DIRECT COMMUNICATION by our cable between New York and Havana. It is the ONLY ALL-CABLE ROUTE between these points.

NEWFOUNDLAND - Our cables from New York connect with the Newfoundland Government Cable & Telegraphs.

ALASKA - Direct connection is maintained with the United States Government cables and land lines to all important places in Alaska.

AUSTRALASIA - Direct communication with the Fiji Islands, Australia, New Zealand, Tasmania, etc., is maintained via the British Pacific Cable from Victoria. British Columbia.


CANADA - Our connection is the extensive system of the Canadian Pacific Railway Company's Telegraph.


MEXICO - The Postal Telegraph-Cable Co. connects with the lines to all of Mexico.



Text and images courtesy of New York Public Library, Myers Collection at SIBL
See also the main page on the Commercial Cable Company.

Last revised: 13 October, 2011

Return to Atlantic Cable main page

Search all pages on the Atlantic Cable site:

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:

—Bill Burns, publisher and webmaster: