Escher.gif (426 bytes)

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

On the Practical Operations connected with the Paying-Out
and Repairing of Submarine Telegraph Cables
by Frederick Charles Webb, Assoc. Inst. C.E.

Introduction: Following papers on submarine telegraphy presented to the Institution of Civil Engineers by F.R. Window (13 January 1857) and J.A. Longridge and C.H. Brooks (16 February 1858), this paper given by Frederick Charles Webb on 23 February 1858 examined the practical aspects of cable laying.

It was published in Minutes of Proceedings of the Institution of Civil Engineers, Volume XVII, Session 1856-1857, pages 262-297 (paper) and 298-366 (discussion).

Footnotes from the original have been consecutively re-numbered and arranged to follow the paragraph they reference. The figures from Plate 8, originally at the end of the paper and referred to extensively in the text, are inserted at the point where mentioned.

Thanks to the staff of the Library of the Institution of Civil Engineers for help in preparing this page.

-- Bill Burns

On the Practical Operations connected with the Paying-Out
and Repairing of Submarine Telegraph Cables

by Frederick Charles Webb, Assoc. Inst. C.E.

 

I N S T I T U T I O N

OF

C I V I L  E N G I N E E R S .

February 23, 1858.

JOSEPH LOCKE, M.P., President,

in the Chair.

No. 955.—“On the Practical Operations connected with Paying out
and Repairing Submarine Telegraph Cables.”[1]
By Frederick Charles Webb, Assoc. Inst. C.E.

[1]The discussion upon this and the preceding Paper occupied portions of four consecutive evenings, and was again resumed on April 27th, but an abstract of the whole is given consecutively.

Two Papers on submarine cables having already been read before the Members of the Institution, the production of a third requires some explanation and apology. It will be remembered that the first of these Papers, by Mr. F.R. Window,[2] treated principally of the effects of lateral induction, whilst the second, by Mr. Longridge and Mr. Brooks,[3] consisted of a mathematical investigation of the subject. The Author has, therefore, been led to believe, that a description of some of the practical operations connected with submarine cables, might prove of general interest. He regrets, however, that the difficulty of obtaining information and drawings, from those who have at times laid down cables, will prevent him from giving such detailed descriptions as he could desire, of any of those operations on which he has not been personally engaged. So long as the submerging of cables remains a matter of private contract, the details of the operations will, it is to be feared, be kept as much as possible from publication; for the experience and information gained at considerable expense, involves in its publication pecuniary interests, and it becomes therefore a matter of trade. Where, however, civil-engineers are employed by a Telegraph Company to submerge a cable, without the intervention of a contractor, no such interests can intervene; the Company simply requires their cable to be laid, and when that is accomplished, the publication of the operation can in no way interfere with their interests. It is to be hoped, therefore, that where engineers have charge of such operations, details of the machinery employed, and of the mode of carrying out the work, may, from time to time, be published, for the advancement of this branch of the profession. With this view the Author has attempted to contribute some information, by describing several of the works upon which he has been engaged; introducing also a few general remarks on the various operations connected with submarine lines.

[2] Vide Minutes of Proceedings Inst. C.E., vol. xvi., pp. 188-202.

[3] Vide Ibid., vol. xvii., pp. 3-43.

The first submarine cable, of any importance, was that laid on August 25th, 1850, from Dover to Cape Grisnez, by Mr. Wollaston, in the steam-tug ‘Goliath,’ at which expedition the Author was also present. It consisted simply of a copper wire, insulated with gutta-percha, and without any further protection. It was half an inch in diameter, and was wound on a light drum, made from a spar of wood. This drum was placed horizontally across the deck of the vessel, and the rope passed from it over a wooden roller on the taffrail. Flat-lead weights, with diagonal grooves, were riveted together, in pairs, upon the rope; an extra piece of gutta-percha being placed round the rope between it and the weights. The ship was eased every three, or four minutes to attach the weights. A gentle breaking friction was applied to the drum, to prevent it from obtaining momentum, and thus unwinding faster than the vessel was going over the ground. The ends near the shore were encased in lead pipes Signals were transmitted from the vessel to the shore during the operation, and subsequently from shore to shore.

This cable failed a day or two afterwards; but the experiment proved so completely, the practicability of telegraphing across the Channel, that the construction of the present Dover and Calais cable (manufactured by Messrs. R. S. Newall and Co.) was shortly afterwards commenced at the works of Messrs. Wetherly, at Millwall. The pattern of the cable was decided by Messrs. Crampton and Wollaston; but it is difficult to say, to whom is due the credit of the application of wire-rope as an outer covering, as it is attributed to several gentlemen, some claiming the general idea, and others certain portions of it. The cable consists of four No. 16 copper wires, insulated with two coatings of gutta-percha to gauge No. 2, laid up together, and served with hemp, and the whole covered with ten No. 1 galvanized-iron wires. The cable was coiled on board H.M.S. ‘Blazer,’ a vessel of about 500 tons; her engines and boilers having been previously removed. A simple compressor, or nipper, was fitted, to prevent the cable from running out too fast. This cable was payed out on September 25th, 1851, under the superintendence of Messrs. Crampton, Wollaston, and Reid.

The inconvenience of towing a vessel when paying out a cable, instead of using her own engines, was here first experienced. The tug broke her towing-hawsers; and whilst fresh ones were being laid out, the ‘Blazer’ drifted helplessly with the tide, tailing the cable round, which was payed out in the direction of the tideway, or nearly at right angles to the intended course. This, and the fact of not keeping the vessel sufficiently ‘headed’ to the tide, for the slow pace at which she was proceeding, caused 24 miles of cable to be payed out to reach only within half a mile of the shore, or a waste of nearly 25 per cent of cable. This cable was afterwards completed, by the addition of another short length, and, with the exception of one repair on the French coast, has remained perfect until last winter, when it was broken by a vessel at a distance of three-quarters of a mile from the South Foreland. It was, however, repaired by Messrs. Crampton and Andrews, in the ‘Monarch’ steamer.

The next cable laid was from Holyhead to Howth. It was successfully submerged by Mr. Newall, in the steamer ‘Britannia,’ on June 18th, 1852. This cable, consisting only of one No. 2 gutta-percha-covered wire, surrounded by twelve No. 13 iron wires, without any serving, appears to have been too slender, for it failed a few days after it was laid.

An attempt was then made by Mr. Newall, to lay a cable from Port Patrick to Donaghadee. The cable consisted of five No. 3 gutta-percha-covered wires, surrounded by ten No. 1 iron wires. The ‘Britannia’ was again employed, and a break, nearly similar to the one represented in Plate 8, Fig. 13, was fixed on the quarterdeck, round which the cable made several ‘turns.’ Sixteen miles of this cable were successfully payed out, when a heavy breeze sprung up, and as the breaking power necessary to prevent the cable from running out, in the depths then reached, was sufficient to swing her to the breeze, in such a strong tideway, the vessel, in spite of her helm, could not be kept in her course, but gradually swung round, stern to sea and wind. Thus, all attempts to steer her towards shore were fruitless, and the sea, beating against the counter of the vessel, rendered it necessary to cut the cable, and slip the end in a depth of water of 80 fathoms, at about seven miles from the Irish coast. This cable has since been recovered by Mr. Newall.

Fig. 13

On the 4th and 5th of May, 1853, a cable was successfully laid by Messrs. Newall and Co., from the South Foreland to Middlekirke, in Belgium. It consisted of five No. 2 and one No. 3 gutta-percha-covered wires, surrounded by twelve No. 2 iron wires. The cable was laid in the screw steamer ‘William Hutt.’

In the same month, Messrs. Newall laid a cable of a similar pattern from Donaghadee to Port Patrick.

Having thus briefly alluded to the different cables laid up to the period of the submerging of those belonging to the International Telegraph Company, on which work the Author was engaged, the description will proceed more in detail.

The place of departure of these cables on the English side is Orfordness, a low shingle point, on the coast of Suffolk; and that of arrival on the Dutch side is Scheveningen, a fishing-village, without any harbour, on a low, straight, sandy beach. The distance across, from shore to shore, is 114½ statute miles, and the greatest depth of water is 30 fathoms. The system decided on, and carried out, for this line of telegraph, is different to any of the others. Four single cables, each containing one conducting wire, are substituted for the compound cable containing four, or more wires. At the time these were projected, no other submarine cable had been successfully laid, excepting that from Dover to Calais. The great increase in distance, therefore, in an enterprise then only in its infancy, rendered caution and prudence necessary. The four single-cable system was, consequently, suggested by Mr. Edwin Clark, M. Inst. C.E., the engineer, as it prevented the possibility of the whole capital being risked, and perhaps lost, in one operation. It also greatly facilitated the laying of the cables, and, under certain circumstances, it allows of one cable being repaired, without disturbing the others, whilst it greatly lessens the probability of having the communication entirely stopped. On the other hand, though this is not an argument against the single-cable system, the actual pattern of the single International cables has proved too light for that particular locality, the amount of anchorage by the collier and fishing-traffic being enormous, and extending across the whole of the North Sea. Experience has also shown, that single cables should be laid apart, at a distance of several miles, so as to insure their not being overlayed by each other. Had the cables been thicker, and been eight, or ten miles apart, this system would, in the Author’s opinion, have been perfect in an engineering point of view. The cost, however, would, in that case, exceed that of a four-wire compound cable, in proportion as the single cables were made more weighty; for with the pattern now used, the two systems are very nearly of the same first cost. It is to be feared, that the saving in capital will always induce companies to lay down large compound cables, in preference to two, or more, heavy single-wire cables, though experience shows, that even heavy compound cables are liable to be broken by anchors, thus involving the entire interruption of communication.[4]

[4] The communication through the Ostend cable has now been entirely stopped, for several weeks, twice in a period of eighteen months.—F.C.W

On the English coast, a compound cable, weighing 13 tons to the mile, formed of six of the single cables twisted round a hempen core, containing a seventh conducting-wire, extends for 3 miles from the beach. The end of this cable terminates in seven tails, or tag-ends. Four of these are spliced on to the four main-cables, and the other three are attached by a chain to the mooring of a buoy, ready for use, should a fifth cable be required, or should it be necessary to shift a main-cable through another wire of the large one. On the Dutch coast, a similar large cable extends for 2 miles from the beach. The construction of the single, or main-cables, is as follows:—A No. 16 copper conducting-wire is doubly covered with gutta-percha, to gauge No. 1. This is then covered with strong tape, wound round it, and has a coating of tar and sand, forming a kind of asphalte. A serving of hemp is next added, which is again covered by ten No. 8 galvanized-iron wires; the whole forming a cable three-quarters of an inch in diameter, and weighing 2 tons to the statute mile.

The cables were manufactured, two at a time, at the works of Messrs. R.S. Newall and Co., at Sunderland. As the cables came from the machine, they were coiled in a dock 6 feet deep, constructed by the Author for the purpose, in which salt water was constantly kept, to the level of the top tier of the cable. During one day in each week this was filled to 2 feet above the coil, and the cable was tested for insulation with strong battery power, after several hours’ soaking. Thus a really true test for insulation was obtained.

The first cable was commenced to be coiled on board the ‘Monarch’ steamer, on the 20th of May, 1853. The first 15 miles were coiled into the fore-hold; then the ‘bight’ was led aft, and the next 36 miles were coiled into the main-hold; then 40 miles were placed in the fore-hold; then 36 miles into the main-hold, and, lastly, 9 miles into the fore-hold, making in all 136 miles of cable. By thus coiling the cable in sections into each hold alternately, the ship never became so far out of trim, as seriously to affect her steering, whilst paying out the cable. In the coiling, each tier was commenced from the outside, and coiled inwards. When the coil, which was oblong, was reduced to a minimum diameter of 6 feet, the cable was crossed over to the outside, and coiled towards the centre again, as before. Each ‘turn’ round the hold was stopped down, with rope-yarns, to the tier below, in six different parts of the coil. The ship was fitted with a break, almost similar to the one shown on Plate 8, Fig. 13. Two shears, supporting saddles, over which the cable passed, were erected over the hatchways, to give the ‘lead’ of the cable sufficient height to take out the ‘turns.’[5]

[5 ]Vide page 269.

Fig. 33

The length to be laid being so much greater than had ever before been attempted for a submarine telegraph, and there being a strong tide-way in the North Sea, it was deemed prudent to buoy off the line to be taken, so as to insure a straight course. This was effected by Lieutenant (now Captain) Burstall, R.N., in H.M.S. ‘Adder.’ Fourteen flag-buoys, somewhat similar to those represented in Plate 8, Fig. 33, were moored at intervals of about eight miles apart, on the proposed line.

On the 30th of May, at 8 A.M., the ‘Monarch’ was anchored close to the beach, and the end of the cable was landed at Orfordness. At 10 A.m. the vessel was got under weigh, and commenced paying out the cable; the ‘Goliath’ steam-tug being in attendance, and Lieutenant Burstall, in H.M.S. ‘Adder,’ piloting the way. Twelve men in the hold cut the hemp stops, and cleared the cable away, at the rate of 4½ miles an hour. These men were relieved every two hours. The ships proceeded for 15 miles, keeping some high trees (called Gedgrave Trees) in a line with Orfordness H.L. This gave an E.S.E. course over the ground, or the magnetic course for Scheveningen; and brought the ship to the first buoy. From this point the ship was so steered as to keep the buoy bearing, by an azimuth compass, exactly W.N.W., until lost sight of. By this time the ‘Adder,’ which was in advance, had sighted the next buoy, and immediately shaped her course, so as to bring it to bear E.S.E.; the ‘Monarch’ bringing the ‘Adder’ on the same bearing. When the second buoy was reached by the ‘Monarch,’ the same operation was again repeated, and so on until the whole work was completed. Thus the whole distance was, as it were, ranged out. When about 40 miles from the English coast, a breeze sprung up from the N.E., which freshened to a gale. The ‘Monarch’ (from the dead weight she carried) rolled very heavily, but in spite of this, the men steadily cleared away the cable; and although there were six hours of darkness, one buoy only was not sighted, so carefully were the tide and wind allowed for. Scheveningen was reached at 6 P.M. on June the 1st, the passage having occupied thirty-two hours. The end of the cable was not, however, landed until the 4th of June, on account of the heavy surf. The length of cable payed out was 119 statute miles, or only 4½ miles above the actual distance. For the first 40 miles, the cable was payed out at the rate of 4½ miles per hour. A slight strain was applied to the break, but, in shallow seas, the Author has found that the slacker the cable is allowed to lie, the more easily it can be hove up for repairs. In changing from the fore to the main-hold, the ship was stopped, and the ‘bight’ was hauled down into the main-hold, previous to paying it out. It was thus under command, and could have been cleared from any obstacle in its passage; which would not have been possible, had it been allowed to run itself aft. The cable was tested for earth (from the ship) every quarter of an hour, with a power of twenty-four 12-plate batteries, and gave 40° on a galvanometer, which stood at 70° short circuit, with one pair of plates. Signals were also exchanged every three minutes with Orfordness, to prove the continuity of the wire.

The second Hague cable was payed out in a similar manner to the first; but the third and fourth were payed out without any buoys being laid down.

Immediately after the two first cables were completed, the compound cables, or shore-ends, were laid. That from the English coast was payed out from the shore, towards a vessel moored 3 miles out, on the line of the other cables. Care was taken, by keeping the vessel well to the southward of the small cables, not to overlay them. When the cable was nearly all payed out, the vessel was brought up. The small cables were then grappled for, cut and spliced on to the large cable, and the whole was let go, with a buoy attached, as previously described. The Dutch shore end was then laid, in the same manner.

The two cables from Holyhead to Howth, belonging to the Electric Telegraph Company, which were laid in September, 1854, and June 1855, have each separate shore-ends. These are gradually tapered off, by decreasing the size of the wire, to the same size as the main-cables, so that the two can be conveniently spliced. This plan has since been adopted for the shore-ends of the Atlantic cable. The two Irish cables are of reverse lays, that is to say, the outside wires run in opposite spiral directions. Thus, when a cable is picked up, it can instantly be identified, a desideratum the want of which greatly increases the difficulty of repairing the Hague cables.

On Laying Submarine Cables.

In the arrangements for paying out a cable, the selection of a ship is evidently a matter of the greatest importance; and this embraces the question, whether a screw, or a paddle-wheel steamer is to be preferred? A screw-vessel, if she has her engines aft, affords the greatest facility for stowing the cable. In the first Atlantic expedition, the ‘Agamemnon,’ carried nearly the whole of her portion of the cable in a single coil, almost circular. In a screw-steamer, however, the head-way cannot be stopped so promptly as in a paddle-wheel vessel, which is a great disadvantage. The screw, being formed in the best shape for going ahead, has not, when reversed, the same effect in checking the speed of the vessel that the paddles have, which latter act equally well, whether going ahead, or astern. Further, the back-water of the screw, reacting against the vessel, prevents her gaining stern-way quickly, whilst the back-water from the paddles goes away clear of the vessel. When any sudden impediment to the egress of the cable takes place, the almost simultaneous stoppage of the vessel is necessary, to prevent the cable from parting, and the more promptly this can be effected, the less risk is incurred. When, therefore, the engines of a screw-vessel are not placed well aft, and consequently there are not greater advantages for stowage, a paddlewheel vessel is preferable.

The disposition of the cable on board, in the most convenient form for paying out freely, has also to be carefully considered. Before proceeding to describe the method at present adopted, it is necessary to make a few remarks upon the conditions of a coiled rope. When any rope is coiled, each length forming a convolution of the coil is twisted once round on its axis, Plate 8, Fig. 1.[6] When a rope so coiled is uncoiled, a twist in the reverse direction, for every twist given in coiling, is caused. If the rope is let down from a point exactly over the centre of the coil, and then laid in a circular coil, this twist is equally distributed over every part of each convolution (Fig. 2), and does not affect the flexibility of the rope unequally, or more in one place than in another; and if the coil is large, the amount of torsion so caused will be, in consequence, extremely small. When the rope so coiled is uncoiled, the reverse twist is also equally distributed, and consequently, the two neutralize each other, in the most perfect manner, over every inch of the rope, which is thus in exactly the same state as before coiling. If the rope is coiled in a long oblong coil, the twist will be unequally distributed, occurring mostly at the two ends, Fig. 3.

Figs. 1, 2 & 3

If the rope is led along to the coil, whether oblong, or circular, from a point only a few feet above the surface, and outside the plan of the coil, this twist is liable to be thrown in all at one part of the coil, by the manipulation of the men; as, for instance, if they carry the ‘bight’ successively into the positions A, B, C, D, Fig. 4. In this case the twist would take place at T, Fig. 4, and the effect of the torsion so caused would be visible, as a kind of hump, extending through the whole breadth of the coil, Fig. 5. When this rope is uncoiled, although a reverse turn must occur for every turn put in at T, Fig. 6, in coiling, yet this is, in certain cases, thrown into the rope at a different part of the coil to that of the first turn, as in Fig. 6, where the reverse turn is shown at R. The two turns will therefore exist, for a moment, in the rope several feet from each other, although they are in exactly reverse directions, as at T and R, Fig. 6. In this position, they both tend to twist round the intermediate rope, as at E, Fig. 6, to its natural state; when, if this action takes place thoroughly, the rope will be exactly as it was before coiling. If the ‘turns’ are far apart, the wire soft, or the rope exposed to friction, by trailing on the ground, or passing through fair leaders, this neutralizing action is liable to be impeded. The two twists may then partially remain in the rope, and these overtwisted and undertwisted parts will be liable to cause awkward bends and irregularities, which should always be avoided, as they sometimes lead to serious consequences.

Fig. 4

In oblong coils, or in a badly-coiled circular coil, it is advantageous to lead the rope from the coil over a saddle, or other fair leader, placed at a considerable height above the coil. When the rope is being payed out, this height allows that portion of the rope which is immediately leaving the coil to hang freely, and thus be unaffected by the friction which it would be subjected to, if it passed, directly after leaving the coil, over a saddle, or other fair leader. The two twists are then under the most favourable circumstances for completely neutralizing each other. At the instant this action takes place, a violent jerking and vibration becomes perceptible in the suspended rope, which the men employed have termed ‘the turn going out.’

[6] In this and the subsequent figures, the intermediate line is supposed to have been marked straight along the rope, when extended free from torsion, and the position this line assumes, under the different circumstances described, is represented in the figures.—F.C.W.

Figs. 5 & 6

If a rope is wound on a drum by the rotation of the latter, no ‘turn’ is put in, and when unwound, the rope is still free from ‘turns.’ If, however, a rope is wound on a drum, and is then slipped bodily off and uncoiled, one complete and permanent twist will be made for every convolution; and vice versa, if a rope is coiled and then unwound, a twist for every convolution will be the result. For this reason, gutta-percha-covered wire, and iron wire, which are always wound on a drum at the manufacturer’s, should never be uncoiled, but should be put on a drum and unwound.

Cables were always, during the first operations of this kind, and are even now in some cases, coiled, for convenience, in long oblong coils, to suit the hold of the ship. This plan, however, besides presenting an unequal distribution of the ‘turn,’ or twist, in each convolution, has a far more serious disadvantage, involving considerable risk in paying out. It may be thus described:—Suppose Fig. 7 to represent a plan and section of a coil of cable, which is led along under the beams of the ship, and up to the deck at one end at A. The cable is always coiled from the outside towards the inside, and when one tier is thus completed, it is crossed over to the outside, and a fresh tier commenced. The centre portion of each tier, therefore, leaves the ship first. Now, if it is assumed that the cable is being payed out from the last few turns of a tier, it will sweep in a ‘bight,’ K L M, over the surface of the tier below, and if two, or three of the inner convolutions of the lower tier B D C, should, from any spring, or twist, in them, lift themselves a little above the surface of the rest, the bight, K L M, will pass under them, and the result would be a fearful entanglement. This has been termed, by the sailors, a ‘foul f[l]ake.’ To prevent this arising, men are stationed in the hold, whose duty it is to carry the bight K L M from its position in the coil, and to drop it when it has passed the inside of B D C. The operation of paying out the cable depends, therefore, on manipulation, and the carelessness of a man, in throwing down the bight too soon, may cause the entanglement described. The Author, on one occasion, saw about two tons of cable thus dragged up the hatchway, carrying down the shears in its progress; the ship was rolling along before the wind, with a heavy sea, on a pitch-dark night, and the cable was buoyed and cut, only just in time to prevent its parting and the end being lost.

Figs. 7 & 8

In a circular coil, Fig. 8, the part of the cable which is in motion, sweeps round at right angles, or nearly so, to the convolutions of the tier below. This kind of accident cannot therefore occur. A further danger, however, has to be guarded against. Suppose, in a circular coil, the centre, or eye, is left hollow; then the inside convolution of any tier will be liable to slip down to the position shown in Fig. 9. When this convolution commences leaving the coil, its position being unfavourable to the gradual removal of the ‘turn in the cable, by the neutralizing ‘turn,’ the twist is gradually drawn into a shorter compass, Fig. 10, and frequently it is drawn so close as to result in a ‘kink,’ Fig. 11. The centre has therefore to be filled in solid. The plan introduced by Mr. Newall, Fig. 12, consisting of a cone in the centre, with rings suspended around it, to guide the rope in its course, presents the most perfect arrangement for uncoiling a cable. Under this system, few men are required in the hold, and these only to lower the rings as the coil decreases, and to attend to the ‘bight,’ when the cable changes from the outside to the inside convolutions. The cable thus sweeps regularly and smoothly round, and all danger of ‘foul f[l]akes’ and ‘kinks,’ is removed. The adoption of the circular form of coil, renders necessary several coils in a ship, where a single oval one would suffice, as the size is limited to a circle, whose diameter is equal to the beam of the ship; whilst an elliptical coil would have its breadth alone limited by the beam, and the length could be considerably increased. The disadvantage and danger of changing from one hold to another, although considerable, is, however, fully counterbalanced by the advantages and decrease in risk obtained by a circular coil.

Figs. 9 - 12

Great care should be taken in so disposing the cable on board the ship, that, during the paying out, the ship shall never be much ‘down by the head, or by the stern.’ If the vessel is disproportionately laden at the stern, with the wind on the ‘beam,’ her head being the lightest, would fly off from the wind; and in the case of her being ‘down by the head,’ the reverse effect would take place. This has to be counteracted by the helm, but, in an extreme case, and a heavy breeze, it would be insufficient. In nautical language, a vessel that is down by the stern carries ‘lee helm,’ and when down by the head, ‘weather helm.’ When possible, also, the rope should not be placed entirely in the bottom of the vessel, for if the centre of gravity of the vessel and of the load is low down, the rolling of the vessel will be greatly increased. It is a well-known fact amongst sailors, that although the loading of a vessel with iron, or other heavy material, placed near the bottom, will render her stiff under canvas, it will produce heavy rolling when at anchor, or under steam; whilst a vessel that is ‘crank’ is invariably easy and rolls little. Heavy masts will reduce this tendency to roll, and this is so well known, that merchantmen laden with iron, when rolling heavily at anchor, will frequently hoist their topsail-yards to the topmast-heads, to raise the centre of oscillation. In loading railway bars, they are stacked up so as to raise the centre of gravity as much as possible. The Atlantic cable, coiled on board H.M.S. ‘Agamemnon’ last year, being entirely on the flooring, caused her to roll heavily in a moderate swell, and this was still further increased, by the fact of the masts of a frigate having been placed in her. In the U.S.S. ‘Niagara’ there were six different coils, only one of which was on the flooring, whilst two were on the upper deck; the weight was thus exceedingly well distributed, and as she had heavy spars, there was not the same tendency to roll.

It is also essential, that some good mechanical means should be provided, for applying a retarding force to the cable, so as to prevent it from running out, by its own weight, faster than is desirable. The break used in the case of all those cables, with a metal sheathing, which have been successfully laid, with the exception of that from Dover to Calais, has consisted of a cast-iron drum 6, 7, or 8 feet in diameter, round which the cable takes four, or five turns. This has one, or two break-straps, which can be tightened up by a lever, so as to cause a certain amount of friction, and thus retard the revolution of the drum. The friction of the cable upon the cast-iron prevents it from slipping, or surging round the drum, when the latter has a retarding force applied to it, by means of the friction-strap. A cast, or wrought-iron piece, called the knife, or plough, which fits close to the drum, is fixed just over the place where the cable leads on to the drum. This keeps continually shifting the ‘turns’ on the drum horizontally, thus keeping a clear part on the drum for the cable to lead on to. This arrangement prevents one ‘turn’ of the cable from riding over another. This kind of break was first employed, for submarine purposes, by Mr. Newall, in laying the Holyhead and Howth cable. Fig. 13 represents the break on this principle, designed and used by the Author, in the North Sea and Irish Channel. In this case, the break-strap pressed directly upon the casting, producing friction of iron against iron; but in others, wooden blocks have been inserted round the drum. In laying several of the heavy cables, two breaks have been employed, one before the other, and each with two break-straps. The cable is guided on to the drum by various means. The Author has adopted three rollers, which can be adjusted so as to leave an opening suitable to any size of rope. They are also so arranged, that should a kink, or any other small obstacle, present itself, they will open, and allow it to pass.

Fig. 13

In the case of the failures of the two cables in the Mediterranean, in 1855 and 1856, under Mr. Brett, the break power does not appear to have been sufficient. Sudden and alarming flights of the cable are described by Mr. Brett to have taken place. In one instance 2 miles of cable ran out in five minutes, flying round even when the drums were brought up dead, showing that the number of ‘turns’ round the drum was insufficient. The drums, the Author believes, were about 8 feet in diameter, and each had two straps.

Fig. 14

The break designed for the Atlantic cable by Mr. Bright, Fig. 14, presents some advantages, and some disadvantages, as compared with the old plan. The advantages may be thus stated. First, the ‘turns’ being taken round separate sheaves, there is no necessity for a knife, which, it is alleged, has sometimes caught protruding broken wires, though the Author has never known such a thing to occur. Secondly, the V sheaves have a firmer grip of the rope than a flat drum has, as a greater amount of surface, per foot-run of the rope, is touched by the cast-iron, whilst the rope is less liable to be crushed flat, being supported (in section) over nearly half its circumference; whereas on a flat drum the rope touches the drum, theoretically, at one point only of its circumference. Thirdly, the sheaves, with their bearings, being both supported on one side of the wheel, the cable can be taken off and on without having to obtain either end. The disadvantages may, in the Author’s opinion, be thus stated. First, the great weight of metal to be set in motion, and the resistance caused by such a series of spur-gearing. Secondly, the time taken to release the break-chocks, B, Fig. 15, from the break-pulley C, requiring as it does several turns of a small hand-wheel A. Thirdly, the break-chocks being unattached to the castings, and the castings D unconnected with the tightening gear E, when the break has been put hard on, and it is requisite to release it, the tightening gearing is drawn away from the pulley, but the chocks remain wedged up between the pulley and the top and bottom of the two castings, in which the chocks are supported, thus causing considerable friction.

Fig. 15

In shoal water, the break is not of very great importance; but in deep water, where the strain on the cable has to be kept as high as possible, consistent with safety, it becomes essential that it should work freely. The motion of the vessel causes unequal strains to come on the cable, which must be relieved from them, by quicker delivery into the sea. The break, therefore, has to revolve faster, and the heavier it is, the greater, of course, will be its resistance to an increase of motion, or in other words, its vis inertia brings an extra amount of strain on the cable. When the strain decreases suddenly, the break is liable to stop. In such a case a light and active break is of the utmost consequence; for, if the machine is heavy, and presents also considerable friction in its parts, its resistance to being dragged suddenly into motion, is enormous. If to this is added the resistance of the friction appliances, which from any cause cannot be immediately released, the total strain on the cable will be extreme.

The attendance on the break, in shoal water, is naturally an unimportant duty, but in deep water it becomes a position of considerable responsibility. The rate at which the cable is being payed out must be, as far as possible, compared with the rate at which the ship is going over the ground, and the strain should be regulated accordingly. The irregular strains which come on the cable must be watched, and the break be released. It is true, that several means can be devised for compensating this irregularity of strain; as, for instance, by leading the cable round three sheaves, one of which shall be kept in position by springs, weights, or, by the piston of a cylinder filled with steam, or air. But all these arrangements present, besides many practical difficulties, more, or less complication, which on board ship is to be avoided. These may, perhaps, in large ships, be carried out, and answer to some extent; but the quick release of the break by hand is a necessity, which, in the Author’s opinion, will never be entirely dispensed with by any contrivance.

In paying out a cable, it is necessary to keep up the most accurate and constant tests for insulation. Thus, if a fault goes over, it is immediately detected, and steps are taken, either for the recovery of the cable, or for suspending the operations. Merely ‘signalling through’ should not be relied on, as a test of insulation, particularly where intense induction currents are used, and still more, where the cable is divided into sections, which are only added to the circuit as they are required for paying out. For, if a fault passes over, which offers a certain amount of resistance, some of the current from the shore battery may still traverse the wire beyond it, and be sufficient to work the relays on board. When, however, the next section is added, the increased resistance in the wire will cause more of the current to pass out at the fault, and when sufficient resistance is thus added, the proportion of current passing to the relays may be insufficient to work them. Thus, many miles of cable may be payed out, before the fault, which will be many miles astern, is noticed. Even where the cable is in one section, a slight fault may be payed out unnoticed, if ‘signalling through’ is the only test employed, which in time will become worse, and the cable fail a few days after it is laid. The Author has been given to understand, that but few regular tests for insulation were made during the attempted paying out of the Atlantic cable in 1857, but the reception of a current simply, was relied on. Professor Morse observes[7]

“We got an electric current through till the moment of parting, so that the electric connection was perfect; and yet the further we payed out, the feebler were the currents, indicating a difficulty, which, however, I do not consider serious, while it is of a nature to require attentive investigation.”

[7]  Vide ‘Journal of the Society of Arts,’ vol. v. p. 632.

The difficulty appears to the Author to have arisen from the loss of current by bad insulation, the cable being then in the act of being submerged, a process which would bring into play any imperfections, or damages.

In paying out the cables in the Mediterranean, between Cagliari, Malta, and Corfu, they were tested by the Author, and by others, every twenty minutes; the wire was so perfect, and the tests were so delicate, that immediate indications were given of the passage of a shower of rain, over the mile of land wire, which connected the end of the cable with the office at Cagliari. The length, in circuit, was sometimes 900 miles.

No cable should, in the Author’s opinion, ever be taken to sea, without its insulation having been tested, whilst the cable was submerged in its entirety, for all the most delicate tests would not prove the cable, unless made under those conditions. It is true, that the total conduction by the gutta-percha, as a medium, is to some extent measured by testing it on a damp day, when coiled in an open yard; but it is quite possible, that an actual mechanical injury to the gutta-percha, such as a puncture, may exist, which will not, so long as the gutta-percha is dry at that particular spot, give any indications. When, however, this is submerged, the water penetrating to the copper wire, immediately produces a fault, which on a long line, where the resistance is great, it would be impossible to work through. The Author has known two, or three such cases to occur in his experience. The cable had been tested on shore, and was even wetted, but not submerged, and had tested well. When, however, it was payed out, a fault suddenly occurred, and on recovering the cable, by winding back and finding the fault, it was proved, by its nature, to have existed before the testing ashore took place.

The Atlantic cable was not tested under water, for fear of the outside wires rusting, and thus impairing its strength. Had these been galvanized, the cable might have lain in the water for a year without fear of rust. An impression exists, that galvanized is not so strong as ungalvanized wire. That this feeling is, to a certain extent, erroneous, is proved by a series of experiments, made by Mr. R. Curle, with a chain-testing machine, at the works of Mr. M’Vicars, on the 27th November, 1855. These experiments were not undertaken for the purpose of ascertaining the relative strength of galvanized and ungalvanized wire, but simply to prove the absolute strength of the materials employed by Messrs. R.S. Newall and Co., and that the breaking strains given by them were correct.

Per R.S. Newall's Card Ungalvanized. Galvanized
Circumference
of Rope.
Weight
per
Fathom.
Caliper
Circumference
of Rope.
Broke at Caliper
Circumference
of Rope.
Broke at
inches.
4  
 
3 3
4
3 1
2
3 1
8
2 5
8
2  
 
1 5
8
lbs.
14  
 
12  
 
10  
 
8  
 
6  
 
3  
 
2  
 
inches.
4  
 
3 3
4
3 1
2
3 1
4
2 5
8
2 1
8
1 1
2
tons.
28  
 
23  
 
19 4
20
18 4
20
14 12
20
7  
 
4  
 
inches.
4 1
4
4  
 
3 1
2
3 1
4
2 3
4
2 1
8
1 5
8
tons.
24 5
20
26  
 
21  
 
18 4
20
11 10
20
7  
 
4 15
20

But, even allowing that the strength was slightly impaired by using galvanized wire, it is believed that this would be more than counterbalanced, by the advantage of proceeding to sea with a knowledge that the cable is perfect; a fact which must remain doubtful, as long as the cable is not submerged when tested, and particularly where its insulation is disguised by a thick coating of tar and oil, both of which are tolerably good insulators.

The steering of vessels across tide-ways, for the purpose of paying out submarine cables, differs greatly from the steering of a vessel on an ordinary passage. In the former case, a straight course over the ground is to be aimed at; and in the latter, the shortest possible passage as regards time. When the master of a steamer is about to set the course for a passage across a ‘beam’ tide, he first takes into consideration how many hours’ ebb, and how many hours’ flood, will take place during his passage. If his passage occupies about twelve hours, it is clear that he will, in ordinary cases, have six hours’ flood, and six hours’ ebb. Thus he would be carried out of his course by one tide, but back again by the other. He therefore steers the actual magnetic course for his port, and thus makes the land at the same spot as he would do if there was no tide at all. Further, he will only have run the same distance through the water, that he would have done had there been no tide at all, and he will have made the shortest passage possible. But, if the line the vessel has traced over the ground is laid down, it will he found, that he has made a considerable deviation from a straight line. Supposing he started at young flood, or ebb, this line will be a kind of wave, similar to Fig. 16; the first part being a curve, caused by the accelerating velocity of the tide, the next a straight line, during the uniform velocity of the tide, then a curve during the slacking and changing of the tide, &c. The versed sine of this wave, or deviation from the straight course, will be the velocity of the tide, multiplied by the time of its running, six hours, and the chord will be the vessel’s speed, multiplied by twelve hours, the assumed time of passage. If, instead of starting at young flood or ebb, the vessel started at any other time of the tide, a different curve will be traced; but the result, as far as time, distance, &c., are concerned, will be the same (Fig. 17).

Figs. 16 - 18

If, in paying out the third Hague cable, which occupied twenty-five hours, an E.S.E. course had been steered throughout, the ship would have made the land correctly at Scheveningen, but the cable would have been laid on the ground in a double bow, or wave (Fig. 18). Owing to the slow speed of the vessel whilst paying out, compared with the velocity of the tide, the versed sines of these bends, in proportion to their chords, would have been very great, and the quantity of cable required would have been 135 miles, although the actual distance is only 114½, whilst the actual quantity used was 119 miles. In order to lay the cable straight, it was necessary to steer the ship in such a direction, by constantly changing her course, and to keep her going at such a speed, as would counteract the effect of the tide; so that the direction of the resultant of the two forces,—the tide and the way of the ship through the water,—lay exactly on the proposed line of the cable, which, as before stated, was E.S.E. For this purpose, before paying out the Hague cable, the Author drew up a table from the Admiralty Tide Tables, with the set and rate of the tide for every hour of the passage. During the operation of paying out, one of the accompanying vessels occasionally let go a stream-anchor, and took the rate of the tide, which was then signalled to the ‘Monarch.’[8]

[8] Since the above was written, a cable has been laid from Zandvort, in Holland, to Dunwich, in England. The distance across is 122 statute miles, and the amount of cable payed out was 138 miles, although great strain was applied to the breaks. Thus 16 miles of extra cable were payed out, being 10 miles more than the average extra length used on the Hague cables, allowing for the increased distance across; so that not only has extra expense been incurred, but the resistance of the circuit has been permanently increased, by a length of 10 miles. This result is no doubt due to the want of proper allowance on each change of the tide; for it appears that the time occupied in the operation was forty-five hours, or nearly four tides, so that the cable probably lies upon the ground in three or four contrary curves, as described in the paper.—F.C.W.

The Author has also adopted a quick practical method, suggested to him by Capt. Burstall, R.N., of obtaining the required rate and direction in which the vessel should be steered across a tide, to make the resultant any given course. The card, or diagram, used is shown in Fig. 19; in which the distance between each circle represents a knot.

Fig. 19

Suppose, that it is required to make good an E.S.E. course; that the tide is setting to the S.W., and that its rate is three knots per hour; required to know the direction in which the vessel should be steered, and the speed at which she should be made to go through the water, in order to pay out the cable at the rate of four knots per hour. From the intersection of the three-knot circle on the S.W. line, draw a line D C to the four-knot circle on the E.S.E. line. Parallel to this line, draw a line A B through the main centre. This line will give the direction to be steered. Through the point C draw a line parallel to the S.W. line, so as to intersect the line A B, then the distance between F, the point of intersection, and the centre A, will give the number of knots at which the vessel should be made to proceed through the water. The result, in this example, will be,—course to be steered, E. A N.; and the rate the ship should go through the water, 5½ knots per hour. The actual application of this rule is, as will be readily seen, very simple, for it only requires two parallel lines to be drawn, and the measurements to be taken.

In bringing up a ship, when paying out a cable, great care should be taken to relieve the break from strain, the instant the anchor is let go; for, should any strain be kept on the cable, the vessel will be prevented from swinging freely to her anchor; and the tide, or wind, acting on her, thus laid athwart it, will cause the ship to drag her anchor, or to break the telegraph cable.

In the process of paying out, accidents will sometimes occur, which completely stop the operation for a time, although the cable is not actually broken. In such cases, in certain depths of water, it is quite practicable to buoy the end of the cable. It is necessary, therefore, that proper buoys, with suitable moorings should always be carried with the ship, and should be so arranged, that they can be let go at an instant’s notice, or they are otherwise useless. In several cases where telegraph cables have been lost, a proper arrangement of buoys would have enabled the operators to leave the work, and to complete it at a subsequent period, with perfect ease; thus property of immense value might have been saved. In the cases of the Donaghadee cable, where the end was let go in a depth of 80 fathoms, the Newfoundland cable in 200 fathoms, and Mr. Brett’s Mediterranean cable in 300 fathoms, each of these ends could have been buoyed. In the two first of these cases, the cables did not part, but were cut. In the last case, the supply of cable ran short, and the ship hung on for several days, and finally broke it.

The operations which the Author has carried on in the North Sea, and which are described further on, amply prove the utility of buoys, when laying down shoal-water telegraphs. On several occasions, when an accident has occurred in paying out, or a fault of insulation has passed, the Author has cut and buoyed the cable, even in extremely heavy weather, on a pitch-dark night, at 40 or 50 miles from land, with confidence that the work could be resumed again, even after the lapse of several weeks. The water in these cases was only 30 fathoms in depth, or much less than in the examples cited. Still, in the greater depths mentioned, it would be quite possible to buoy. Larger buoys would have to be employed, and when the depth is greater than 60 or 70 fathoms, hempen ropes should be substituted for the chains. In no case, however, should the buoy ‘ride’ by the telegraph cable; but it should be attached either by a chain, or by a hempen rope, to an anchor, mushroom, or sinker; and this again be connected to the telegraph cable.

In the Atlantic Telegraph expedition of 1857, the buoying arrangements (which were intrusted by Mr. Bright to the Author) consisted of three buoys for each ship, for buoying with chain in 30 fathoms, two for buoying with chain in 60 fathoms, or with hemp rope in 200 fathoms, and two for buoying with hemp rope in any depth above 200 fathoms up to 2,000 fathoms. The success of buoying in the latter depth would, of course, have been doubtful, but still it was a precaution, which it would have been imprudent to neglect. Should the vessels, in starting from the centre, make either coast in a gale of wind, blowing on the land, and thus not be able to approach the coast, or should they run short of cable, in reasonably shoal water, say 300 fathoms, the buoying of the end could, by careful management, be easily effected. The Author has buoyed ends at 3 or 4 miles from the land, as far back as 1853, and at a distance of 50 miles as far back as 1856.

In paying out, it sometimes becomes necessary to stop the egress of the cable, for a certain period. When this occurs, in a strong beam tide-way, or heavy beam wind, the situation becomes exceedingly critical, even if the first act of stopping the vessel is performed in time; for a vessel, when stopped, tends to lie ‘beam to wind,’ and thus drifting bodily to leeward, will speedily bring a strain on the cable, whilst neither moving ahead, or astern, will, in that position, ease the strain on the cable, which is growing out ‘broad on’ the beam, that is, at right angles to the axis of the vessel. If the cable is strong enough, the vessel will be gradually brought round stern to wind, or tide, and hanging stern on, will exert considerable strain on the cable. If in this position, the vessel could be eased up to the cable, by her engines, the strain might be relieved, but steamers (whether screw, or paddle) have a peculiar propensity to turn their sterns invariably towards the wind, whenever they gain stern way. This they do in spite of their helm. If, therefore, the wind and tide do not coincide, the vessel, on gaining stern way, will sheer clean away from the direction of the cable, and thus, perhaps, actually increase the strain. But even if there is no wind, but a strong tide, it is almost impossible to steer a vessel up to the cable, stern foremost, for the helm when once put hard over cannot, whilst she has stern way, be forced out of its position, and can only be kept amidships with great difficulty and danger to the men at the helm. In very deep water, it is some time before the vessel will drift sufficiently to cause this strain, and as the streams in deep water are very gentle, this difficulty is then happily lessened. In shoal water, with a strong tide, the best means of overcoming the difficulty, if a stoppage must occur for more than a few minutes, is to sever the cable, and get the end forward, when the ship can be steamed and steered up to it, when once she is checked round end on, so as to relieve all strain. If the cable is not sufficiently strong to check her round, warps should be bent on, so that the end may be eased away, whilst the vessel is being steamed round. For the purpose of getting the end forward, a hawser should always be kept in readiness, led from a sheave at the bows, and leading outside all, to the stern of the ship. This is then attached to the cable abaft the paying-out gear; the cable is then cut, and lowered over the stern, and hauled up forward. Or another plan is, to cut the cable and buoy it. Both these operations have been performed by the Author. To get the cable into position again, when the stoppage is over, the cable from the hold should be hauled round the breaks, over the stern, and passed outside all, to the fore part of the vessel. The splice is then made to the sea end. This is then lowered over by a rope passing over the bowsheave, until the strain comes on the stern, when the hempen rope is cut away. Thus, up to the very last minute of the change, the cable has been hanging from the head of the vessel.

When a stoppage in the paying out occurs, with a leading wind, or tide, the primary stoppage of the vessel cannot be so promptly effected, as with a head wind or tide. For this reason, the operation of paying out is far more dangerous under the former, than the latter condition. A heavy-rigged vessel, like the ‘Agamemnon,’ would, with a following gale, go through the water at the rate of 6 or 7 knots, without steam, or canvas. The reversing of the engines would not stop her under such circumstances, and hence there would be great danger in paying out, with a heavy leading wind. What would most probably be done, under such circumstances, would be to heave-to and drift, at 3 or 4 miles per hour, endeavouring, as far as possible, to make good the proper course over the ground.

The effect of a ship being out of trim, thus causing her head, when she is much ‘by the stern,’ to fall off before the wind, has already been alluded to. There is, however, in the case of paying out a submarine cable, another cause which tends to affect the steering of the ship. The cable, when confined over the stern, and when the ship and the egress of the cable are stopped, will, if it is strong enough, as before described, swing the ship stern to wind, or tide. When the vessel is progressing, and the cable paying out, it still has a tendency to swing her, if there is a wind, or surface-current acting at right angles to the ship’s course, or if a tide, extending even to the bottom, is running at right angles to the course, but the cable is being payed out under tension at the bottom. The cable will, in each of the three cases described, assume a direction nearly coincident with her course over the ground, and will thus form an angle with the ship’s axis. In this position, the cable acts as a check-rope, tending continually to check her stern to wind, or tide. The problem may be thus explained, geometrically. Let A B, Fig. 20, represent the vessel’s way through the water, in the case of a tide, or current, and her progressive movement ahead, in the case of the wind alone; and let A C represent the effect of the tide, current, or wind on the vessel, acting through the centre of displacement, and at right angles to the course she is steering; then A K will represent her course, made good, over the ground.

Fig. 20

Now, the horizontal direction the cable will assume, immediately on leaving the ship, will be as follows:—In the first case, in which a surface-current extends to a shallow depth, and the cable is payed out without loss, and without tension at the bottom, the cable will assume a direction very nearly parallel to the ship’s course over the ground, but dependent on the proportion of the depth of the current to the total depth of the sea. In the second case, and the tide extending the whole depth of a shallow sea, with a cable being payed out under tension at the bottom, the cable will assume a direction approaching the course over the ground, but dependent on the amount of tension applied. In the third case, of wind alone, the cable will assume a direction exactly coincident with the ship’s course over the ground, which will also be in the exact line of the ship’s ‘wake.’ The direction the cable will assume in all these cases will, therefore, be at an angle with her axis. Let D E, Fig. 20, represent the tension on the cable; this may be resolved into two forces, D G, which is merely retarding the progress of the vessel, and D F, which is acting in the opposite direction to the tide, wind, or current. There is then the force C A (wind, tide, or current), and the force, D F, acting at right angles to, and at the two extremities of the line D A, thus tending to turn it round, to a position in which the two forces, D F and A C, act in exact opposition to each other. They are, therefore, continually tending to turn the vessel round, and will do so, if not counterbalanced by some other force.

Fig. 20a

It will be seen by Figs. 20a and 20b, that no change in the position of the vessel can destroy this tendency, so long as the cable forms an angle with the ship’s axis, that is, so long as it ‘grows on the quarter.’ For the tension on the cable, when it forms an angle with the ship, may always be resolved into two forces, D M and D L, the former of which is retarding the progress of the vessel through the water, and the latter acting at right angles to the line D A. In the same manner, the momentum of the ship, represented by A K, may be resolved into two forces, A O and A N, the former in the direction of the axis, and the latter acting at right angles to it. It will be evident, that D L and A N still act at right angles to, and at the extremities of the line, A D, and consequently tend to turn the ship round, to such a position, that D E, D A, and A K, shall form one straight line.

Fig. 20b

In the case of a tide, where the cable is being payed out without tension at the bottom, and where the tide is acting throughout the whole depth of the sea, the cable will not assume the direction of the ship’s course over the ground; for each particle will drift as fast as the ship, and thus the direction the cable will assume, from the ship to the ground, will be the same as in still water, namely, in the vertical plane of the ship’s motion through the water, and the cable will be deposited on the ground in a line parallel to the ship’s course. In practice this can rarely occur, for in a sea shallow enough for the tide to extend to the bottom, the cable would, in all probability, be payed out under tension at the bottom; and it would then, as before remarked, tend to assume, from the ship to the ground, the direction of the course made good. It will be seen by the projection, Fig. 21, why the cable, when a surface-current is running, will also assume, very nearly, the direction of the course over the ground.

Fig. 21

Let A O represent the ship’s way through the water, A R the action of the surface current, and A P the course made good. If balls were dropped at equal intervals of time, the line joining them at any instant, would assume the line, ABC, and any one ball would trace, in falling, the dotted line, A E F. In the case of a cable, however, as it would be under considerable tension at the point, B, at the entrance of still water, it would be impossible for an angle to exist there, as there is nothing to keep the cable in that position, but the resistance of the water to the cable to move laterally, either in the current, or in the still water. This would be nil, when the lateral motion was nil, and therefore the cable would, if it had ever assumed the position ABC, be moved laterally, by the tension, to a direction, A K L, nearly coincident with the course over the ground, forming a slight angle, or rather curve, at K, dependent on the proportion between the depth of the current and the depth of the sea. The deeper the current, the nearer would the line, A K, coincide with A B; and if the current extended entirely to the bottom, it would, as before stated, coincide exactly with A B. This is, of course, on the supposition that the cable is being payed out free from tension at the bottom.

An under current may also carry the cable out, at an angle with the ship’s course, this being almost the reverse of a surface current; but whatever may be the cause of the cable assuming an angle with the ship’s axis, whether it arises from wind, tide, surface, or under current, as long as there is that angle, and the cable is being payed out from, and is confined over, the stern, the tendency to swing must continue. When the ‘cable grows on the quarter,’ the steerage of the vessel must be affected, more or less, and there will always be a tendency to swing. This tendency has, therefore, to be counteracted by the helm, and a disposition of the sails, which would, however, in an extreme case be insufficient. When the cable, from tide, or current, but without wind, ‘grows on the port quarter,’ the vessel will carry starboard helm, and vice versa. When the cause of the ‘grow of the cable’ is wind alone, it will of course grow on the weather quarter, and the vessel will require lee-helm, or after-canvas; and, in fact, whenever the cable grows to windward, from whatever cause, the above rule will hold good. If a strong weather-tide should cause the cable to grow on the lee quarter, the vessel will carry weather helm, and may require head-canvas.

Fig. 20c

This action of the cable on the ship’s steerage must not be confounded, as the Author has occasionally known it to be, with the case of a spar, or boat, towing by a rope, fast to the quarter of the vessel, from which it differs essentially. It may easily be seen, that although the tow-rope might, and would in most cases, ‘grow’ straight astern, yet if towing from the weather-quarter, the tendency would be to make the vessel carry weather-helm, and if towing from the lee-quarter, lee-helm. Let D E, Fig. 20c, represent a force, as a rope towing a boat, or spar, acting parallel to the ship’s axis, from her quarter; and let A C represent the ship’s way, acting through the centre of motion, then the force D may be resolved into two forces, D M acting in the direction of a line passing through the points D and A, the centre of motion of the vessel, and D L acting at right angles to D A, consequently tending to turn the point D round the centre A. If D is on the weather-quarter, the vessel’s stern will be impelled by the force D L to leeward, and she will consequently require weather-helm, to counteract this force. With a light cable and a large ship, this check will scarcely be felt; but with a heavy cable and in deep water, where a great strain is applied to the cable, it becomes very dangerous. The Author believes, from the accounts and evidence he has gathered, that this was the principal cause of the loss of the Donaghadee and the Newfoundland cables. In the heavy Mediterranean cables, this tendency to swing would have been exceedingly dangerous, had there been much beam-wind, or current.

Fig. 22

The first plan suggested for overcoming this difficulty, was to let the cable pay out freely in any horizontal direction, from a point at the centre of the vessel, imitating the practice which all tugs adopt, of having their tow-ropes fast only at the centre of the vessel. In Fig. 22 the forces A C and D F would be always acting through the same straight line, and no tendency to swing would take place. This, however, is difficult to carry out in practice, to its full extent, for many reasons. A modification of the principle has been adopted in the fittings of the ‘Monarch’ steamer, belonging to the Electric Telegraph Company, the free point being 12 feet forward of the taffrail, thus benefiting, to some extent, the steerage of the vessel across a tideway. It may be easily understood how dangerous this tendency to swing must be; for, with the wind, or tide, at right angles to the course, should the tendency to swing overcome the counteracting effect of the helm, or canvas, the vessel will gradually swing completely round, and thus proceed in a totally different direction to that required. (Figs. 20, 20a, and 20b.) To recover her position, the only chance is to go faster through the water, and thus to make the helm act more energetically. But the tide, or wind, is now in her favour; and as there is a limit to the rate at which the cable can safely leave the ship, the speed must be slackened instead of increased, and all hope of recovering her position is gone.

Fig. 23

The cause of the loss of cables in deep water may be thus roughly traced. Suppose a cable leading from A, at a ship’s stern, to the ground at B, Fig. 23. For that portion to be laid without loss on the ground, the point A must move to C, and in doing so, must move laterally through the water a distance D C, the versed sine of the angle ABC. The resistance of the water will, however, oppose this horizontal motion, and unless great tension is applied, the point A will fall to some spot nearer to D. If the lateral motion is entirely obstructed, the point A will fall to D, and the loss will then amount to a length equivalent to D C.

Fig. 24

With regard to the curve assumed by the cable in paying out, the Author thinks that it will be concave towards the ship in every part, but approaching nearer to a straight line as it reaches the bottom (Fig. 24). The angle of the cable as it leaves the ship is found, by actual measurement, to be 9° or 10° with the horizon.[9] If the rope continued at this angle all the way down, the loss, or waste, could not amount to 30, 40, and 50 per cent., as it now sometimes does; for the versed sine of an angle of 10° would not be more than the one hundred and fifteenth part of the radius. The rope is, therefore, steeper as it approaches the bottom, and thus brings the lower part to such an angle, that its versed sine may bear the same proportion to radius, as the per-centage bears to the amount of cable payed out. That the curve cannot be convex, at the lower part towards the bottom, is argued from the fact, that the rope is slack on the ground, and there can, therefore, be no tension to draw it into such a curve. Any point in the cable at the top of the curve, if it sank perpendicularly, would give a loss less than that at the bottom, which is impossible. The cable at the top must, therefore, move backwards, to make the loss at the top agree with that at the bottom. This is found to be actually the case. The weight of the lower parts of the curve sinking perpendicularly in the water, causes the top parts of the rope to slide backwards, and any point in the rope probably describes a curve somewhat similar to that shown by the dotted line.[10]

[9] This the Author ascertained by means of a sextant, when paying out the Atlantic cable from the ‘Agamemnon,’ in a depth of 2,000 fathoms.—F.C.W.

[10] The beautiful investigation by Messrs. Longridge and Brooks on the submerging of telegraphic cables, tends to throw considerable light on this branch of the subject; still the Author prefers to leave his own rough reasonings unaltered for publication. That investigation shows, that the cable can fall to a position within the point D, (Fig. 23,) and that the loss of the cable can be considerably more than the versed sine of the angle, when insufficient tension is applied; consequently, it does not necessarily follow, that the cable must assume an angle at the bottom, the versed sine of which shall be equal to the loss of cable, as argued by the Author. With regard to the curve being concave towards the bottom, they have admitted that, as the measured angle of the cable with the horizon from the ship to the surface of the water is less than the calculated angle of the cable with the horizon below the surface, at any ordinary velocity, there must be a change in the direction of the cable after it enters the water; and as it is impossible that the cable can, under such tension, form an abrupt angle, it must form a curve concave towards the bottom. The correctness of the Author’s diagram (Fig. 24), therefore, will depend on how far down this curve extends, before it merges into a straight line. This has not been calculated, or shown by any one. It has, however, been assumed that this curve extends but a very limited distance.—F.C.W.

The mechanical arrangements for paying out a cable are no doubt important, and should be carefully attended to; but undue consideration is sometimes given to them, whilst the value of an organised and efficient staff for superintendence is perhaps under-rated. No amount of mechanical contrivances can dispense with the forethought, the quick detection of cause from effect, and the prompt decision, which is necessary in all nautical operations. Persons unaccustomed to the sea, however mechanical-minded they may be, are apt to limit their observation, when placed on board ship, to what occurs inboard. Thus, the silent effects of tide, current, leeway, &c., are neglected, because the evidence of their presence is unperceived, until the effect of the forces thus brought into play leads to some mishap. A ship is, in fact, in almost every circumstance, except in a dead calm, and when there is no current, or tide, perpetually moving at a different velocity, and in a different direction, to what, to the casual observer on board, she appears to be. A surface current, or a tide, may be running with the ship, and the ship, consequently, proceeding faster over the ground than she is going through the water. She thus requires, and takes, a greater delivery of cable, than that due to her velocity through the water. This might lead an inexperienced person to imagine, that the cable was running out faster than it should do, and induce the application of greater break power, when in reality it was unnecessary.

When a cable suddenly parts outside the ship, in deep water, no evolution can, of course, remedy it. There are, however, situations, such as those previously described, and many others, where the handling of the ship, or the prompt execution of an operation, such as cutting and passing the end forward, cutting and buoying, stopping, &c., is necessary for the safety of the cable. Correct judgment as to the proper instant at which such evolutions are necessary, and a knowledge of the manner of carrying them out, can only be obtained by long observation of the various actions to which a ship is subjected.

Again, if a difficult evolution is even schemed deliberately beforehand, its performance requires that every man and rope should be in position, and that each successive part of the operation should be performed exactly at the proper instant. This requires a kind of skill, the practice of which is rarely called for in land operations. These delicate evolutions require also an organization, discipline, and centralization in the staff employed, extremely difficult to procure in a body collected from various quarters, to carry out a work of such short duration as the submerging of a single telegraph-cable.

In the Atlantic expedition of 1857,had the ‘Niagara’ payed out her portion of the cable successfully, a most difficult operation would still have had to have been performed. The end of the cable on board the ‘Agamemnon’ would have had to be passed to the ‘Niagara,’ and spliced to that ship’s cable; or the ‘Niagara’ would have passed a short piece to the ‘Agamemnon,’ and two splices would have been made. The ‘bight’ of the cable in the ‘Niagara’ would then have required to have been passed up, out of the hold, and freed from the break, which was, it is true, exceedingly well adapted for this operation. The ‘bight’ would then have had to have been lowered over the stern, until the strain came on the ‘Agamemnon’s’ rope, when that ship would have commenced paying out. Here, there is an example of an evolution to be performed, in deep water, requiring the utmost care and precision in its execution, and on the completion of which, would have depended all the previous success obtained. The naturally unorganized state of a staff, collected simply for this one work, most part of whom had never been engaged on submarine telegraphs before, would have rendered the safe execution of this dangerous manoeuvre extremely doubtful, even in fine weather. In foul weather it would have been utterly impossible.

It is but justice to the engineers engaged, of whom the Author was one, to state that they one and all strenuously opposed the proposition of starting from the shore, instead of from the centre, in which latter case the operation of splicing would have been simple, and no risk would have been incurred. Their opinion was, however, overruled by the decision of the Directors. It is some consolation to hear, that in the next attempt, it has been decided to start from the centre. It would, however, for many reasons, be advantageous, if the whole cable could be carried in, and be payed out from, a single ship.

It may be argued, that the evolutions described are peculiarly the duty of the officers who have charge of the ships, and that the engineering staff need possess no nautical knowledge. This may, to some extent, be correct, but still a considerable amount of nautical knowledge is requisite, in all those who superintend an operation so eminently nautical; whilst a too great division, between the superintendence of the ship and the cable, cannot, it is evident, work well. The movement of the ship affects the cable, and, in some cases, the cable affects the ship; the action of both must, therefore, be regulated by persons acquainted with the peculiarities of each.

On Repairing Submarine Cables.

Telegraph-cables in the North Sea, and the English and Irish Channels, and indeed in any shallow seas, will always be exposed, more or less, to the possibility of damage from ships’ anchors, and of ultimate decay from corrosion. The shortest and safest means of repairing them becomes, therefore, a subject for the attentive, consideration of engineers. Some account of the repairs executed by the Author upon the Hague, and other cables, may, perhaps, be interesting.

The first step to be taken, in the case of a stoppage of the communication, is to ascertain, as nearly as possible, the position and nature of the fault, in order to judge at which end of the cable operations should be commenced. This must be done by means of electrical tests. The first point to be ascertained is, whether the cable is completely severed. If the slightest deflection cannot be obtained, on a delicate galvanometer, at one end, when a powerful voltaic current is sent from the other, and the wire also shows, by tests for insulation, that it is in contact with the sea, it may pretty correctly be assumed that the gutta-percha and the copper wire are severed, and most probably the outside wires also. The principal means for ascertaining the distance of the fault, when the ends of the copper wire are in contact with the water, is by the resistance of the copper wire to a weak current.

During the construction of the Hague cables, they were tested for resistance at every ten miles additional length, and from these tests, the following Table is constructed. It is taken from a mean of four different cables. The first column shows the length of No. 16 wire in circuit; and the second column shows the deflection on a horizontal galvanometer, with the length of wire in the first column in circuit. The galvanometer on short circuit stood at 52°, with one pair of plates 3 inches by 4 inches:—

Length of Wire In Circuit.
Statute Miles.
 
0
10
20
30
40
50
60
70
80
90
100
110
120
Deflection on Galvanometer
with one pair of Plates.
°
52
39
33
29
26
24
22
20
18
16
14
12½
12

These tests are made with a perfect connexion between the end of the cable in circuit and the earth. If this could always be insured, in testing the cable when in the sea, the distance of the fault could be positively ascertained, within a mile or two, by comparing the deflection obtained with the cable in circuit, Fig. 25, with those shown in the Table, or by dividing a current between the cable and a series of resistance-coils. The latter should be added to, until they balance the resistance of the cable, which may be ascertained by placing two similar galvanometers in the circuits, Fig. 26, or by a differential galvanometer. When the deflections of these are equal, the resistance of the coils is equal to the resistance of the wire in the cable. The distance of the fault may be calculated, by knowing the number of miles of No. 16 copper wire which will equal the resistance of the coils. In practice, however, a perfect connexion between the end of the cable and the earth is rarely obtained. The end of the copper wire, which is bare in the sea, does not make a perfect connexion with the earth, unless it touches the outside iron wires;—and sometimes not even then. Salt water, being an imperfect conductor, requires a considerable surface of copper wire to be in contact with it, to compensate for its bad conduction. When the amount of copper wire exposed is insufficient, a resistance is caused in the circuit, producing the same effect on the galvanometer, when testing for resistance, as so much more copper wire in circuit. Thus the fault appears always further off than it really is. The Author has generally found the resistance of the end to equal about 8 or 10 miles, and he now usually allows that length, in judging of the distance of a fault.

Figs. 25, 26

The return current likewise gives intimation of the length of wire in circuit, but this is affected by the resistance of the end. Still, a comparison of the return current and the resistance is exceedingly important, as the Author will endeavour to show. If the end always made a perfect earth, the return current, of wires of a given pattern and of given lengths, would it is evident be, with a given battery-power, always the same; and, with wires of different lengths, the return current would increase in some definite proportion, which might be ascertained by experiment on different lengths of wire, and tabulated. Where, however, the resistance is increased at the end alone, without prolonging the whole pattern of the wire, and thus increasing also the surface under induction, the effect of the return current will not increase in any definite ratio with the resistance, as it would on the first supposition; for the effect of the return current on the galvanometer, is the discharge of a certain portion of the total statical charge in the wire, one portion passing out at the further end, and another portion passing through the galvanometer. The effect, or value of the total charge, is evidently dependent on the product of the surface charged, and the intensity of that charge at every point; and this intensity is dependent on the resistance between that point and the earth, and on the intensity of the battery. In the case of a wire a mile long, C D, Fig. 27, with a resistance beyond it, equal to 20 miles, D E, caused by the end, there will be a charge over the surface of the mile of wire, equal in tension to the charge over the first mile of a wire 21 miles long, which may be represented by A B C D. This total charge will divide itself in discharging, very nearly in inverse ratio, between the resistance at the end, D E, 20 miles, and the resistance of the galvanometer G. If a wire 21 miles long, or C E, is taken, the first mile will be under precisely similar circumstances to the one-mile length above cited, whilst the charge over the surface of the remaining 20 miles, which will decrease in intensity to nil at the further end, will also have to divide itself, in discharging between the mean resistance towards the galvanometer plus the resistance of the galvanometer, and the mean resistance towards the sea end. The total charge may in that case be represented by the triangle A C E. The total effect on the galvanometer will therefore be much greater than in the first case; whilst the resistance, in the two cases, will be equal, and the result of the tests for resistance similar.

Fig. 27

It follows from this, that the effect which is obtained from the return current, when compared with the resistance shown, will indicate whether that resistance is caused entirely by mere length of wire under induction, or is partly due to a bad ‘earth.’ If there were accurate tables of the resistance and the return current due to given lengths of wire of a certain pattern, it would be possible to make considerable use of this effect, but even without such tables, an extreme case will make itself apparent. Supposing, for instance, the wire was broken at a distance of a very few yards from the galvanometer, it is evident, that even if the resistance of the end was equal to, say 200 miles, the effect of the return current would be very feeble; for, although the intensity of the statical charge would be high, (which cannot, until the resistance is infinite, equal the intensity at the battery,) and a great portion of the total charge would pass through the galvanometer, yet, from the very small amount of surface under induction, the total effective charge would be extremely small. Such disproportion between the resistance and the return current, would at once show, that the resistance was not due to the mere length of wire capable of acting under induction.

When the wire, though broken, shows no contact with the earth, the return current will, of course, be equal to the full amount of the charge of that length of wire, of an intensity equal to that of the battery, all of it being forced to pass out through the galvanometer. When such a case occurs, the charging and discharging of the wire, compared with the charging and discharging of known lengths of wire of the same pattern, with a given battery-power, forms a very approximate method of ascertaining the distance of the want of continuity.

The mode of measuring the dynamic effect of the static charge, or in other words, the return current, is at present very crude. The Author employs the following method:—A key, Fig. 27a, is so arranged, that when pressed down, it shall put the cable, c, in connexion with one pole, z, of a battery, the other pole of which is connected with the earth. The cable thus becomes charged. When the key is released, it is pressed by a spring into the position shown in the figure; thus breaking contact with the battery, and connecting the cable immediately to a galvanometer, which is again connected with the earth. The return charge, therefore, passes through the galvanometer to the earth. This galvanometer has a dial-face, which can be moved round, carrying the coil with it. This dial has a pin fixed on it, against which the needle just touches, when it is at rest. The face of the dial is graduated in degrees, and a permanent pin in the frame of the instrument points to zero, when the needle is at rest. When a discharge current is sent through the coil, the needle, of course, is thrown away from the pin, with more or less violence; the dial is then turned round, and the pin carries the needle to an angle with the vertical. A second discharge is now sent through the instrument, and if the needle still moves from the pin, it is moved to a greater angle, and so on, until the movement of the needle is scarcely discernible. The angle that the zero of the dial now forms with the permanent mark on the frame is then read off, and gives, by comparison with other tests, a rough measure of the electro-dynamic force which has moved the needle.

Fig. 27a

It would be of considerable interest and use, if those who have charge of the testing of long lengths of cable during manufacture, would test the cable for resistance at every 10 miles, increasing the battery-power used, by one pair of plates, say at every 100 miles. Also, for the return current, with the end to the earth, decreasing the battery-power used at every 100 miles; and for charge with the end sealed up, also decreasing the battery power used at every 100 miles. If these tests were carefully tabulated, and the pattern of the cable, the nature of the batteries and the galvanometers specified, the information would be of great use to those engaged in detecting faults.

Fig. 28

When the tests, thus made, indicate that the fault is not far from the shore, the cable may be under-run, in a tug-steamer, to the broken end. To execute this, the cable must be grappled up near the shore, and the ‘bight’ be passed over a sheave, hung at the bows of the vessel, under the cat-head, Fig. 28; or it may be got over the vessel, as shown in Fig. 29. In no case, however, should the cable be passed completely fore and aft, if there is any beam-tide, or wind, as the vessel will then be moored athwart the tide, whereas with the hanging sheave (which is the method the Author prefers when a steamer is used) the vessel can take up her proper position, to counteract the effect of tide, or wind. The end thus arrived at is then buoyed, and the sea-end dredged for; and, when obtained, a piece of cable is spliced on to it, and payed out to the buoy on the shore-end, which is then raised, and the final splice made. This method was first executed by the Author on the Hague cables, in 1853, on the Dutch coast, the fault being 2 miles out; and since then, several faults have been repaired in the same way. The Author has under-run, off the coast of Ireland, as far as 14 miles, into a depth of 40 fathoms; but the strain on the cable, caused by under-running, renders it advisable to adopt this process, only for shallow depths and short distances.

In the severe winter of 1854-55, a fault occurred at 10 miles from the Dutch coast. This the Author was requested to attempt to repair, in a Dutch fishing-schuyt, as it was said to be imprudent to risk a tug on that coast during the winter months. The vessel being capable of locomotion only when there was wind, became, of course, useless in the finest and smoothest weather. The ice, which was very thick on the beach, carried away the buoys, and for some time endangered the vessel, which was the only schuyt that was not housed on the beach; and on one occasion the work of many weeks was completely obliterated. The cable had been under-run to the end and buoyed; but before the other end could be obtained, the weather became so bad, that all work had to be stopped for about ten days, and during this time, the ice carried away all the buoys. Thus the work had to be recommenced. Finally, after three months’ tedious and anxious work, during which the operators were sometimes exposed, for three, or four days and nights, to the bitterest cold and the severest weather, the work was at last successfully accomplished. The cable was under-run, a second time, the whole distance, and the other end dredged up, and a piece spliced in. The cable was led over a V sheave at the head of the vessel. Fig. 29, and was passed from thence over the rail, forward of the rigging, on the weather side, or, when the tide was stronger than the wind, on that side against which the tide was running. By this means the cable could be released at will, and the ship allowed to take up her proper position to counteract the tide; whilst it was also possible to apply manual power to the cable, when there was not sufficient wind to propel the vessel.

Fig. 29

In grappling for an end, the cable must be hooked by the grapnel, neither too near nor too far from the end. For if too near, the end will slip over the grapnel; and if too far, the cable will not come home over the ground, and thus a great strain will be put on the cable, by attempting to lift the ‘bight’ to the surface. In the latter case, (or when a cable, that is not broken, has to be grappled up,) it is lifted from the ground for a distance of about half a mile on each side of the vessel. The strain on the grapnel-rope is then considerable, whilst that on the cable is excessive. The proportion of the strain A B, Fig. 30, on the cable is to that A D on the grapnel-rope, as half the secant A K of half the angle C A B formed by the two parts of the cable, is to radius A D.

Fig. 30

It was thus proved to be practicable, to under-run the cables, for short distances, from land; but even these operations were attended with considerable risk, from the strain caused by the cable being lifted in the manner described, which on one occasion resulted in the cable parting. When one cable overlays another which is being under-run, the operation of under-running is necessarily stopped, until hold can be obtained of the cable which is being under-run on the sea-side of the cross cable. This is effected by grappling for the cross-cable, and lifting it to the surface, when the cable which is being under-run can be procured seawards of it, and the cross-cable let go. This operation exposes the cross-cable to considerable risk, from the strain it is subjected to in being lifted.

Fig. 34

In November, 1855, one of the Hague cables broke, and the tests showed the fault to be about 50 miles from the English coast. To have under-run such a distance would have been injudicious, if not impossible. The means, therefore, proposed and carried out by the Author, were to wind the cable into a steamer, as she proceeded along the course, of the cable, and when the broken end was arrived at, to lay down buoys, and dredge for the other end. As it was extremely probable that bad weather might occur during the picking up, it was necessary to take such precautions, as would allow of the work being abandoned and resumed at pleasure. The Author feeling confident that buoys, if properly arranged, would enable this to be done, and that the cable could thus be cut, and left when bad weather came on, organized a system of buoying, which was employed with success. The ‘Monarch’ steamer was accordingly fitted in the following manner: A piece of machinery, with a drum, 6 feet in diameter, for winding in the cable, Fig. 34, was fixed on deck abreast of the foremast. This was worked by a 4½ H.P. engine. A sheave, 3 feet in diameter, was fitted over the bows, between two balks of timber, 12 inches square. The cable was led in over this sheave, made three turns round the drum, and was then led away into the hold.

Fig. 31

Fig. 33

For buoying the end, a nun-buoy, 5 feet 6 inches long, Fig. 31, was placed in each sponson. These were each connected, by 50 fathoms of chain 3/8ths of an inch in thickness, to a mushroom-anchor, slung at the side of the vessel; an additional 15 fathoms of chain being attached to the mushroom and led forward, and the end kept in readiness near the bow sheave. When it was desirable to buoy the end of the cable, this stray chain was attached to the cable. The vessel then had stern-way given her, and the cable was allowed to pay out. When the strain came on the mushroom-anchor, it was let go; and when the buoy-chain was in a similar manner laid out, the buoy was let go, and the vessel thus dropped clear of it. When about one-eighth of a mile of cable had been payed out, another buoy was attached, in a similar manner, and the cable then cut, and let go. The buoys being thus moored by the mushroom-anchors, no strain could come on the telegraph-cable, which lay slack on the ground, and unaffected by the swinging, or movement of the buoys; but still connected to the mushroom-anchors by the stray-chains, Fig. 32. By having two buoys, when one is sunk by a vessel, or carried away, the other is sufficient to procure the end, and the risk of the loss of the end is reduced.

Fig. 32

These buoys do not project more than 2 or 3 feet above the water, and are thus only visible at a distance of about 1 or 2 miles. To find them, when placed at 50 miles from land, would be difficult. Two more conspicuous buoys were therefore always moored about a mile on each side of the nun-buoys. These buoys, Fig. 33, have a pole passing through them, which is ballasted at the lower end, and carries a mast 18 or 20 feet high on the top, with a large flag.[11] These buoys can be seen at a distance of 5 or 6 miles off on a clear day. As they are inconvenient for weighing in rough weather, they are not adapted for attaching to the cable. It was frequently found possible, in moderate weather, to weigh the nun-buoys, and procure the end, and thus proceed, whilst the flag-buoys could not be weighed without danger of destroying them. A small nun-buoy was also suspended from the bows of the vessel, with its moorings in readiness. When, in picking up the cable, the broken end was arrived at, this buoy was instantly let go, and thus marked the position of the other end, until the flag-buoys could be launched, during which time the ship would get considerably out of position.

[11] Buoys on this principle have been in use, for many years, on the Admiralty marine surveys. The Author has, he believes, improved and modified the details of their construction.—F.C.W.

The arrangements being thus made, on the 6th of May, 1856, the cable was cut outside the Orfordness large cable, and picking up commenced. The cable was wound in at a rate varying from 1 to 2 miles per hour, depending on the depth to which the cable was buried in the ground. A constant look-out was kept on the cable, and the ship was steered so as to keep as little strain as possible on it. The deck-engine had also to be constantly regulated, to suit the varying strain on the cable. At intervals of 10 miles the cable was cut, and the section on board was tested for insulation whilst wet. A length of 17 miles had thus been hove in, when, in a breeze that had been freshening, the vessel lifted so as to break the cable. Sufficient experience had not been then attained to know when it was prudent to cut and buoy the cable. Three flag-buoys were let go and moored near the spot. The wind freshened to a gale, and the ship had to return to harbour. A week afterwards grappling for the end was commenced. After three weeks’ tedious work, being driven away continually by wind and sea, the end was at last obtained. The cable was, however, fast under a wreck, and was with difficulty freed. The difficulty was augmented by the three other cables being all near the one that was being grappled for; for although they were constantly hooked, it was impossible to heave them to the surface without danger of breaking them. Thus, although the right cable must also have been caught seawards of the wreck, yet from its tightness, caused by its being pinned down by the wreck, it could not be distinguished from the others. It was not until the boats dredged over the short end, which protruded shorewards of the wreck, that it became evident they had the desired cable.

The end being once more on board, the operation proceeded; and when 50 miles had been picked up in all, the distance from land being 53 miles, the broken end came on board. The flag-buoys were launched, one of them having a lantern lashed to it, the ship was kept steaming round it, as there was too much sea on to anchor. At daylight on the following morning, the other end was dredged for, and obtained in the afternoon. Two nun-buoys were attached to the end, and three flag-buoys were moored near them. The 50 miles of cable were then taken to Lowestoft, landed, and thoroughly repaired. The ship then proceeded to the buoys, and on heaving up the end, and testing it, a second partial fault was found to exist, though the cable was not severed. Picking up was again commenced, and the second fault came on board at 40 miles from Scheveningen. The end was again buoyed, as before, and some repairs on one of the other cables, which was broken in two places, at 8 miles and 14 miles from land, were proceeded with. When these were completed, it was decided that the rest of the other cable, buoyed at 40 miles from Scheveningen, should be picked up, and the whole of it re-laid to the northward of the other cables. The winding in was then proceeded with, and when about 10 miles more had been obtained, a gale coming on, it became necessary again is cut and buoy the cable, and to run for Helvoetsluis. Subsequently the buoys were again reached, and in 26 hours the rest of the cable was obtained. Thus, the whole of a cable, 120 miles in length, was recovered from the sea, without losing a single inch; the cable having been severed in four different places, and the ends buoyed three times for periods varying from 7 to 42 days, the whole operation extending (through bad weather and the performance of other work) over a period of 11 weeks.

The cable was then re-coiled on board and re-laid, but, owing to a defect in the compasses and strong easterly winds, the course of the cable was found to have crossed the line of the others, about 60 or 70 miles from the English coast. In the summer of 1857, the same cable having been broken at about 45 miles from the English coast, it was determined to pick it up from the Dutch coast, towards England, to a distance of 30 miles from England, and then to relay it to the northward. This was accomplished, and the cable was laid in the direction shown in Fig. 35.

Fig. 35

Two of the other cables having been broken at 20 miles from the land, and having a third cable lying on them, it became necessary to remove the upper one for a distance out beyond the fault in the lower ones. The upper cable was accordingly picked up, for a distance of 30 miles, and re-laid in a bow to the southward (Fig. 35). The other two cables were then picked up to their faults, a distance of about 20 miles, and re-laid. In this operation, some parts of the cables that were operated on had been lying undisturbed for four years, and were, consequently, in some places so buried in ridges, or sandbanks, that it became impossible to remove them. This involved a tedious series of operations to procure the cable outside the bank again.

Fig. 36

When the utmost strain that it was safe to apply had failed to produce any effect, the cable was cut and buoyed, say at A, Fig. 36. If at night, a buoy with a lantern attached to it was moored near it. The cable was then grappled for, about a mile or a mile and a half seawards of the buoy. When the cable was hooked, say at C, it was hove up, cut, and tested each way, when it invariably proved to have parted, from the strain caused in raising it, seawards of the vessel, sometimes a mile distant, say at D. The end C of the piece, leading back towards the bank, was then buoyed, and the piece beyond C" D picked up. Great care was taken to steam and steer the ship well up to the cable, so as not to drag home the end D. When it appeared, a buoy was instantly let go, which thus marked the position of the sea end D". The sea end D" was then dredged for, and when obtained was immediately buoyed. The ship then proceeded to weigh the buoys at C, and the cable was picked up, back towards the bank, until no more could be obtained, when the cable was finally cut, and the end let go at B. Thus a piece, A B, was left in the bank, which was generally about 200 or 300 yards wide, and having about 4 or 5 fathoms less water than the part immediately around it. The buoys at A were then successively picked up, and the ship proceeded, until the sea-end was procured. The bank being thus passed, the process of picking up was continued. The cable was frequently buried to a great depth in other places, though not so much as to prevent its removal. To drag it out without breaking required much patience and care, sometimes taking two hours to obtain five yards.

When the cable was thus buried, it was found to be in perfect preservation, the galvanized surface being as clean and good as when the cable was first laid; whilst in those places where it lay on the surface, it was eaten through, extending, in many parts, for 1 or 2 feet in length, to such an extent, that it was only held together by a portion of one outside wire. The gutta-percha and hemp were undisturbed, retaining the impression of the outer wires, thus showing that the decay was not caused by mechanical injury, but by oxidization, probably the result of a galvanic action. The galvanized surface had also entirely disappeared, where the cable was not buried. This decay, where the cable lay on the surface, fully accounted for the invariable fracture of the cable seawards of the ship when it was grappled for, and lifted outside the bank. On the Dutch coast, where the cable appears to bury itself to a less depth, but more regularly, and where the water is less salt, in consequence of the discharge of the great quantity of fresh water from the Rhine, the cables are in as perfect a state as when they were first laid.

Thus by the system of picking up, buoying, dredging, and grappling, cables, where the depth is not very great, can now be regularly repaired; and submarine telegraphs, in shallow seas, become a less precarious property, than they were at first believed to be.

The system, above described, first suggested, organized, and carried out by the Author, for the repair of the Hague cables, is still continued without any particular modification, or improvement, either in the mode of operation, or of the gear employed.

Since these operations in the North Sea, the Author selected, and fitted out, the ‘Leipsig’ paddle-wheel steamer, for the Atlantic Telegraph Company, for the purpose of recovering some of the Atlantic cable. He was unable to attend to the operation himself, but his late assistant, Mr. Suter, was employed, and succeeded in recovering 53 miles, when the cable parted in a depth of about 200 fathoms. Little hope exists, therefore, of the recovery of the remainder.

The Paper is illustrated by an extensive series of diagrams, from which Plate 8 has been compiled.

[Note:The diagrams of Plate 8 have been placed in the appropriate place in the body of the paper.]

March 2, 9, 16, and 23, 1858.

JOSEPH LOCKE, M.P., President,
in the Chair.

The Discussion upon the Papers, No. 975, “On Submerging Telegraphic Cables,” by Messrs. Longridge and Brooks, and No. 955, “On the Practical Operations connected with Paying out and Repairing Submarine Telegraph Cables,” by Mr. F. C. Webb, occupied four evenings, to the exclusion of any other subject.

Professor Airy said, amongst the numerous subjects treated of in the two Papers now before the Institution, there were many into which he could not enter: indeed upon those points on which he intended to offer a few remarks, he must avow that his conclusions could not be given, as possessing accurate and absolute mathematical certainty. Still, he might show the necessity for caution, in accepting some of the deductions in those two Papers; and he should endeavour to do this by using what was known accurately, to illustrate what was not known with the same accuracy. In the first Paper, the opinion was cited (not as that of the Authors of the Paper, but as the opinion of other persons), that in laying out a cable, “the successive portions payed out arranged themselves into wavy folds, since the actual length of the cable was greater than the horizontal distance;” and in the second Paper it was asserted, that ‘the waste,’ by which term he understood the extra length of cable laid down above the actual distance traversed, or more than the length of a straight line between the two places so connected, “varied from 30 to 50 per cent.” He considered, that there was not sufficient foundation for either of these two conclusions; and indeed that both were totally erroneous, if they were to be understood as implying the necessity for such waste. He believed, that the case of a cable laid down without any resisting influences whatever in falling, might be taken as a very close approximation to the laying down of a cable with resistance from falling in the water. He showed by a model, that a small chain, when uncoiled from a roller, round which it was wound, was delivered in a straight line, as the roller travelled along the upper plane of the model. Assimilating that to a ship, the cable must be supposed to be payed out as fast as the ship itself progressed. The chain might be so far unwrapped from the roller, that it would be delivered in a direction absolutely perpendicular to the horizon, and yet the chain was laid on the floor of the model in a perfectly straight line, without the smallest loss of length. If the chain was wound up, so that it took any degree of inclination, as the roller travelled along the upper plane, the chain was still deposited, without the slightest loss of length, and everywhere the chain was laid down straight. This was not a mathematically accurate representation of what occurred in depositing a cable in water; for, as the vessel was going through the water, the cable, in dropping down, experienced some little resistance, but of that resistance a very insignificant, probably an imperceptible, part was horizontal. The resistance which the chain experienced was almost a vertical resistance, and amounted to nearly the same thing, as a little decrease of its specific gravity. He considered the same results stood very approximately, in the case of depositing a cable in water, as in the case of depositing a cable in air. Hence, he argued, that, allowing for the greater buoyancy, or the reduction of specific gravity, by immersion in water, a cable could be laid at the bottom of the sea, in the track of the ship, exactly as it left the coil in the hold, without any loss of length. It was impossible that the cable could take the form represented in one of the diagrams, with its concavity downwards; in conformity with the remarks just made, it must be a very close approximation to the common catenary, touching the bottom of the sea at its lowest point. It was also impossible that, in the mode of delivery supposed above, there could be the wavy folds represented in another diagram; for there was a horizontal tension on the cable where it reached the ground, which would keep in a straight course every part of the cable as it was deposited. Instead of the ‘waste’ being 30 per cent. or 50 per cent., it need not be 1 per cent., or any perceptible per-centage; although, of course, it might be made large by an unreasonably wasteful delivery. Apprehensions of the kind mentioned had led to the cable being payed out in a direction very little inclined to the horizon; and it had been stated, that the Hague telegraph-cable made an angle, at the ship, of 9° or 10° with the horizon. One thing, so far as he could ascertain, was not taken into account. There were no calculations, on theoretical principles, of the tension which the cable must experience. Speaking under the qualification of not saying exactly what was the case with bodies in water, but stating exactly what was the case in air, as a very near approximation to the state of things in water, and taking known principles (the accuracy of which he guaranteed), he was enabled to compute this tension. It was clear, that the minimum tension was that which took place when the cable hung vertically, and was the weight, in water, of a piece of cable, whose length was equal to the depth of the sea. Taking that as a unit. he had calculated, in terms of that minimum, what the tension would be, when the cable was payed out from the ship at given angles with the horizon; and the results were given in the following table:—

Angle made by the Cable,
at the Ship,
horizontal line
°
5
10
15
20
25
30
35
40
45
50
55
60
Tension on the Cable, at the
Ship, expressed in terms
of the minimum tension
 
262.8
65.8
29.4
16.6
10.7
7.47
5.53
4.27
3.41
2.80
2.35
2.00

Thus, when a cable was payed out at an angle of 10°, there was a strain upon it of nearly sixty-six times the minimum tension, or sixty-six times the depth of the sea at that place. When this was considered, it seemed to him, that in all the annals of engineering, there was not another instance, in which danger was incurred so needlessly. There might be no great risk in the shallow waters of the North Sea, south of the Dogger Bank; but in soundings of 2,000 fathoms, in the Atlantic, it was madness to imitate the style of delivering the cable, which might not be attended with danger in other places. If it was certain that the cable would always be deposited pretty exactly in the way in which it was deposited in the air, (of which he had no doubt,) there would be no objection to saying, that the angle at which it should be payed out should never be less than 45° with the horizon; and there would then be very great security for the cable. Some rule of that kind must, he believed, be adopted, in order to insure the comparative safety of the cable in traversing deep seas. The Table was not given as an accurate representation of what occurred in water, but it was a close approximation. The principal object of his remarks was to draw attention to the enormous increase of strain, when the cable was payed out at an angle so near the horizon, as appeared to have been the case in the instance mentioned in the second Paper.

With regard to the mode of construction of the cable, he remarked, that the external iron-wire covering, which was supposed to form the strength of the whole fabric, being laid spirally, would yield when the strain came upon it; whilst the gutta-percha core, containing the conducting wire, being straight, would have to bear the strain, and hence would be liable to fracture. Thus the insulating medium being cracked, the cable would become useless. He had seen a number of defective gutta-percha-covered wires taken up, and it appeared, that there was an inherent tendency in the gutta-percha to crack all round across its length. This tendency would probably come into play, when the longitudinal strain came upon the gutta-percha; and in this arrangement, with the iron wire laid spirally and the gutta-percha straight, he believed the gutta-percha was exposed to be fractured. This, however, was merely thrown out to elicit from practical men, information as to whether any means had been devised in manufacturing cables, to throw the tension upon the external wire covering; and in which the strain should not in any case come immediately upon the gutta-percha.

Mr. Longridge remarked, that the whole of the conclusions given in the Paper with which he was associated, were based upon the mathematical reasonings given in the Appendix. With reference to the two methods which had been suggested, for catching the cable in case of fracture, the Authors did not insist upon them, as things in which they had great confidence; indeed, he himself saw difficulties in the way of their application. They were simple suggestions; but where so much capital was at stake, and where a single accident might risk the whole venture, some mode of catching the cable in case of fracture was deserving of attention. Whether the means suggested were sufficient, or whether they were practically useful, he did not pretend to say.

Mr. Webb said, in order not to increase, unnecessarily, the length of his Paper, he had left unnoticed many details connected with paying out and repairing cables. He would not now attempt to dwell on these points, and merely adverted to the subject, in order to acknowledge their omission. The process of paying out was far more simple and straightforward than that of repairing. The former was, generally, a continued repetition of the same operation, whilst the latter involved a series of different operations, constantly changing in character, with the various circumstances that arose. If, therefore, he had fairly dealt with each branch of the subject, that of repairing would have occupied by far the greater portion of the Paper. As it was impossible to treat both questions fully, greater prominence was given to that of paying out, as being more generally interesting.

With reference to the historical portion of his Paper, Mr. Crampton had reminded him, that in laying down the experimental wire between Dover and Calais, the vessel was not stopped whilst the lead weights were put on, as he had described, but was merely eased.

Mr. C.H. Gregory thought many of the cables, of which specimens were exhibited, were so formed as to resist a considerable amount of strain, without injury to the conducting-wire. If it was held to be essential for the strength of a rope, that the materials should be in the line of direction of the strain, no rope, whether of iron, or hemp, manufactured as had been suggested, would meet these conditions.

Mr. Charles May said, if he understood the remark correctly, it was this, that there was a twisted outer rope round a straight gutta-percha covering to the inner twisted rope, and the twisted ones would stretch, whilst the straight gutta-percha must bear the strain, or crack.

Mr. Crampton explained, that it might be desirable to have the external wires of a cable laid straight, if it were practicable, without much difficulty; but this was not the case. The impression, that the internal conducting-wire and the gutta-percha covering were straight was incorrect. Mr. Wollaston had suggested that they should be, and they were, laid together in a spiral form, which conferred on them a greater degree of elasticity than was possessed by the outer wires, and these would, therefore, be ruptured before the inside wire was injured. When the Dover and Calais cable was torn asunder, by a ship’s anchor, in January, 1857, it was found that the inside, gutta-percha covered, copper wires were much longer than the external iron wires, proving that the latter had broken first. The outside wires were drawn out to nearly one-eighth of an inch diameter before breaking; but the gutta-percha and copper were drawn more than that before the fracture took place. It was ascertained that the gutta-percha was uninjured except at the point of rupture.

Professor Airy said, at the time of his remarks, he had in his mind the Atlantic cable, in which there was not an involved coil of gutta-percha, but only a single strand.

Mr. Cyrus W. Field remarked, that in the Atlantic cable, although the outside strand would stretch on sustaining a great strain, yet the inside would stretch much more, indeed the core would stretch more than 10 per cent. of its length. The core consisted of seven No. 22 gauge copper wires laid into one strand, and covered with three coatings of gutta-percha. This was served with tarred yarn, and was finally covered with eighteen iron strands, each formed of seven No. 22 gauge wires of the best charcoal iron. In every instance he was acquainted with, where the cable had been tested for strength, and they were many, the iron wires had been broken before the copper wires began to yield, to an extent to impair, materially, the efficiency of the conductor.

Mr. C.T. Bright thought, that all persons, who, like himself, were interested in the extension of telegraphic communication, must be deeply indebted to the Authors of the Papers which had been read. The first Paper might be regarded as a survey of the question from a theoretical point of view, whilst the second Paper was a fit companion to the former one, from the practical information which it furnished.

Reverting to the former Paper, he would allude, in the first place, to the subject of resisters, which was an important point, relative to which frequent suggestions had been made. There was no doubt, that advantage would result, from the employment of any appliance, which would retard the egress of the cable, so as to allow of heavier and better-protected cables being laid in deep water. From the difficulties attending the use of buoys, the Authors appeared to have discarded them from consideration, as being too bulky to be carried, but suggested resisters instead. He had himself been engaged in some experiments in this direction, and he had found that a plane surface 1 foot square offered a resistance, when drawn through the water at the rate of one mile per hour, of about 3¼ lbs.; and with a surface of 4 feet, the retarding strain, at the same speed, would amount to nearly 56 lbs. Colonel Beaufoy’s nautical experiments,[1] upon the comparative resistance of solids moving-through water, gave a similar result, and he had no doubt of the accuracy of that conclusion. At the rate of two miles an hour, the resistance experienced with a disc having an area of 1 square foot equalled a pressure of a little more than 13 lbs.; at three miles it amounted to 29 lbs.; and at eight miles to 203¾ lbs. But there was a disadvantage in the use of resisters, in the form generally proposed; especially in the case of the sudden pulling up of the ship from any stoppage in-board, from the occurrence of a kink, such as was met with in paying out one of the cables most recently laid. Should the egress of the cable be suddenly stopped, from this, or any other cause, it must be subjected to a great strain, until the vessel could be brought up, or else it would part. The pitching of the vessel in a heavy sea was also an important feature, in considering the advisability of employing resistance floats; a remedy for these objections might be found, in so attaching them, that they would be released upon receiving pressure in the direction that would be prejudicial, but it would be far better to do without them altogether.

[1] Vide “Nautical and Hydraulic Experiments, with numerous scientific miscellanies.” By Col. M. Beaufoy. 4to. London, 1834.

Another point suggested in the first Paper was, the possibility of catching the end of the cable, should it break near the paying-out vessel; relative to which the proposal had been made, that auxiliary vessels should follow the paying-out steamer, with the cable passing through a ring. He did not think that plan would be successful, but, on the contrary, he believed it would add to the chances of losing the cable.

Another question was that of compensating for the rise and fall of the ship. The Authors appeared to hold the opinion, that it was of comparatively minor importance, because it would merely alter the form of the curve in a small degree, and only “increase the abscissa of the catenary, by the amount due to half the height of the wave.” With a fragile and light cable, he considered it of more consequence than had been allowed, to compensate for the pitching of the ship; but, he thought, it was far from necessary, or even desirable, under all circumstances, especially where it was possible to use light paying-out machinery; and there was a difficulty in carrying it out, so as to meet every requirement, without adding materially to the weight to be set in motion.

The chief subject suggested in the first Paper was, the comparative advantages of light and heavy cables; and this, he apprehended, would form the leading feature of the present discussion. He fully subscribed to all that the Authors had argued, with reference to the importance of light cables for deep waters. The lighter, the stronger, and the cheaper they could be constructed, the better; but he had not yet seen, what might be termed, a very light cable, that fulfilled all the conditions required of it. In the Atlantic cable, a considerable step had been taken in that direction, as far indeed as it was prudent to go, in the present state of experience. He thought a great deal more had been done, in the matter of light cables, than was known to the Authors. The first experimental cable, between Dover and Calais, was composed of a copper wire, coated with gutta-percha; it was as strong, in proportion to its weight, and the depth in which it had to be submerged, as the cable alluded to in the Paper was to such depths as had to be traversed in the present day. Then there was the cable attempted to be laid between Portpatrick and Donaghadee, in 1852, by a Company which had since been wound up. The insulating substance was gutta-percha with india-rubber, covered with common rope. Then there was another— properly to be called a light cable, although it had iron wires laid spirally outside the core—between Holyhead and Howth, which weighed about a ton to the mile; also, the cable laid between Varna and Balaklava, in 1855. Thus, a good deal had been done, practically, in light cables; but the results had in all cases been most unfortunate. The Dover and Calais experimental cable, in 1851, failed after raising the hopes of all concerned in it. The hempen cable in the Irish Channel never reached the shore. The light cable from Holyhead to Howth broke down shortly after signals were sent through it. And the Government wire in the Black Sea, although it remained in good working order during the period of the war, broke down when peace was proclaimed, as if aware that its mission was over. The Authors recommended another form of cable altogether, and, to some extent, censured those who had had to do with the Atlantic cable, for not having adopted it, because it had been invented three years before. Mr. Bright need not advert to the many experiments by which the form of the Atlantic cable was determined; but the kind which had been advocated in the Paper was considered among others, and rejected, as it always must be, by any one practically experienced in the working of telegraphic wires. The Authors prefaced their remarks upon cables with the observation, that the conducting and insulating powers of cables did not come within the scope of their arguments. But it seemed to him, that this embraced the whole gist of the matter. The cable which they so strongly supported, was composed of a strand of iron wire, covered with its insulating substance, combining the conductor and the strength in one. That seemed, at first sight, to present many advantages; but when the conductor and the insulator were examined, which were, in fact, the very first considerations in constructing a cable, it would be found that the principle adopted was wholly defective, the conductor being altogether unfit for the purpose. The conditions, with regard to overground wires and underground and submarine wires, were very different. The possibility of working through an overground circuit, whatever the loss by defective insulation, was limited only by the resistance of the conducting power of the wire. In submarine and underground wires, on the other hand, the resistance was increased enormously, by the induction of the surface of the metal, and the surface of the gutta-percha. He regretted the absence of Professor Faraday, who had explained the phenomena attendant upon long, gutta-percha-covered wires so clearly at a former meeting. The great object was to get a good conductor. To employ an iron conductor, such as was recommended, would be to increase, to an enormous degree, the difficulties experienced in lines of this kind. According to Becquerel, the conducting power of copper was 100, of gold 93, of silver 73, and of iron only 15; so that iron was possessed of six times less conducting power than copper. To produce an equal effect, it would, therefore, be necessary to enlarge the dimensions of the conductor; whilst the difficulty of obtaining perfect insulation being then greater, by having so much more surface to protect, the chances of leakage would be proportionately increased.

Before concluding his remarks upon the first Paper, he would allude to an intimation, partially thrown out also in the second Paper, that sufficient publicity was not given to operations of this kind. He could only say, that no ground of complaint could exist on that score, with regard to the Atlantic cable, inasmuch as the details of the manufacture of every portion of that cable had been published as fully as possible; and as far as he was personally concerned, he had at all times most freely given all the information it was in his power to afford. It had been stated in the second Paper, that the Atlantic cable was not tested under water, from the fact that the wire used was not galvanized. Now, the core of that cable was regularly tested from the beginning, by Mr. Whitehouse, the laborious and careful nature of whose electrical experiments were well known. It was tested with a battery series of five hundred cells, and the most sensitive instruments that could be obtained from the instrument-makers of London and Berlin, who were most celebrated for delicate electrical apparatus. The wire, when tested under water, where it had remained for some days, was sent to the works upon drums, so protected that it could not possibly receive any injury, and was immediately covered with a serving of hemp, prior to receiving the outer strands. It was not possible to test the whole cable, with the outer strands, under water; because, in a large coil of cable, the water would not reach to the centre of the coil, unless it was placed in a vessel large enough, and strong enough, to enable the requisite pressure to be applied to so enormous a coil; nor would it have been of any value if it had been practicable. The Table giving the breaking strains of common and galvanized iron wire, would probably lead to further discussion. He had always understood, that galvanizing the smaller sizes of wires injured them, and he imagined, that the Table must apply to the larger sizes, although it was not stated, whether the circumferences given applied to bar iron, or to made rope. The diary of Professor Morse had been alluded to, and it appeared, the assumption put forward in the second Paper, that the cable was not properly tested, was grounded on it. It was true, that as they got further from the shore, in the Atlantic undertaking, the signals were weaker; but as Professor Morse was not concerned in the practical part of the undertaking, he was not acquainted with the conditions, or the reasons for what was done,—indeed he was confined to his berth nearly all the time. A very low battery-power was purposely employed, for objects that would be evident to all acquainted with testing wires. It was desirable to continue signalling to the shore, at the same time that any falling off in continuity, or in insulation, should be discovered, by the effect on the instruments; as new lengths of cable were added, from time to time, to the circuit, there was of course some variation of signals, and the adjustment of the instruments was sometimes interfered with, by the motion of the ship; but there was nothing uncommon in the occurrence, and the supposition that a fault had passed unnoticed was altogether groundless.

Mr. Crampton said, he had but little to add to the remarks on the same subject which he had made during the last Session.[2] The Dover and Calais cable, though made and laid under many disadvantages, had only sustained one accident, during the five years it had been down. The Admiralty authorities, it was true, had given him every facility, and had allowed him to select a vessel from any of the Royal Dockyards, to be employed for submerging the cable; but it happened, that there was not any vessel, with engines on board, which had space for the cable, so that, eventually, the choice fell upon an old vessel from which the engines had been removed. The cable was then placed in this vessel, which was towed by a steamer. In taking that course, he was aware that there was a liability of the tow-rope breaking, which did actually occur; but he was obliged to do as well as he could to save the concession. The operations were commenced on a calm day, but, shortly, a three-quarter gale came on, and the tow-rope parted; the vessel, holding on to the cable on board, drifting about 1¾ mile down the Channel, before it was again taken in tow. The laying the cable was then proceeded with, when it was ascertained that there was not sufficient cable to reach the shore by about half a mile. This was immediately manufactured, was spliced on, and the communication was perfected. The loss of cable sustained in that case could form no guide for the future. They certainly had not sufficient break-power, but, with that exception, everything else was foreseen; and if he had to do the same work again, he would, with the single exception named, follow precisely the same plan. The break was calculated upon the supposition, that a perfect piece of workmanship had to be controlled, but it was not adapted for allowing broken strands to pass freely. It should be borne in mind, that there had been no previous experience to act upon.

[2] Vide Minutes of Proceedings Inst. C.E., vol. xvi., pp. 205-207.

Mr. Webb maintained, that the Atlantic cable, in the real sense of the word, when covered with yarn and iron wire, had not been tested under water. The gutta-percha-covered wire was, for all cables, always tested under water, as a matter of routine, at the Gutta-percha Company’s works; but, in the case of the Hague cables, they were further tested, under water, after being covered with iron wire. The gutta-percha core was exposed to many casualties, previous to the outer covering being applied, and even whilst it lay in a coil in the yard, which might affect the insulation. He therefore considered it essential, that all cables should be tested under water, previous to being shipped for their destination.

With reference to the Table giving the comparative strength of ordinary and galvanized iron wire, he acknowledged that it would have been more complete, if the gauge of the wire had been given; but it might naturally be concluded, that the smaller ropes were composed of the smallest wire. It would be noticed, that the smaller ropes agreed more nearly in strength than the larger ones. In the ropes 4 inches in circumference, the ordinary wire bore the greatest strain, but in the ropes of 2 inches and 1 5/8-inch in circumference, the breaking strains of the galvanized and ungalvanizd wires were equal.

Mr. John W. Brett believed, that the first practical application of the system of twisting the gutta-percha core, in conformity with the lay of the external iron covering, was due to Mr. Wollaston; whilst the idea of covering the core with iron wire was entertained by Mr. Brett and his brother in 1846; and Mr. Smith, senior (known for his machinery for the manufacture of wire ropes), shortly afterwards sent him prepared specimens.

He was an advocate for heavy cables in shallow waters, as being better able to resist injury, on anchoring-grounds; and he thought, that the large annual expenditure in maintaining the light cables, from Yarmouth to the Hague, in comparison with that incurred in the cases of the Dover and Calais and the Ostend heavy cables, where there had only been one instance of failure in five years, was conclusive on this point. But that which would suit one set of circumstances was not applicable to all cases; and from experience, he was decidedly in favour of light cables for great depths.

The two cables he had laid down in the Mediterranean weighed 8 tons per statute mile, or nearly 9 tons per nautical mile. The length laid between Spezzia and (Corsica was 90 miles. At the time he started upon that undertaking, he was not aware of the depths to be traversed. The Sardinian Government placed their finest vessels at his disposal, and the Ministers accompanied him from Genoa to Spezzia, to inaugurate the undertaking. The question was put to him, by an able Government Engineer at Genoa, whether he intended to take the straight line from Spezzia. He replied that that was his intention, He was then told, that in some places depths of 400 or 500 fathoms would be encountered, and, to a certainty, the cable would be lost; whereas if he made a circuit of 10 or 12 miles, the depths would not be more than 100 or 150 fathoms, and the cable would be safe. He replied, that that reasoning would be most judicious, if his work ended there, but as it was intended to carry on the line to Bona, in Algiers, far greater depths were expected between the island of Sardinia and the coast of Africa, than those mentioned. The officer who had made these predictions consented to accompany the expedition, and rendered valuable service. A Government vessel preceded that from which the cable was delivered, to take soundings; but the great depths were not made known until after they were passed, for fear, as it was said, the workmen should be rendered nervous. An accident occurred after they had passed upwards of 400 fathoms, when they were paying out in 230 fathoms. The cable ran out with great violence, and by the extraordinary means used to arrest it, the strain upon it was so great, that the cable was injured. The insulation was, of course, destroyed at this point, and also at some distance back. After several efforts, and some delay, the cable was drawn, inch by inch, from the water, until the injured parts were recovered. These were then cut out, fresh splices were made, and means were adopted to prevent similar accidents occurring in future. This occupied about 30 hours, and during the whole of this time, the vessel was anchored by the sea end of the cable.

The Captain, Marquis Ricci, who had hitherto considered it impossible to lay such a cable in great depths, then said, there appeared to be such elements of strength in that form of cable, that he now believed it would be possible to unite England and America, as Mr. Brett had proposed. Four days afterwards the heavy cable between Corsica and Sardinia was laid in a brisk gale, at the rate of 6 miles an hour. That was the most successful run he had ever experienced.

When laying down the cable in 1856, another accident took place, from a sudden run of the cable, caused, he believed, by there not being a sufficient number of ‘turns’ (only three coils) round the drum. The cable having parted, it was decided to return to laud, to drag for the cable, and to under-run it, which they did for 18 miles. The end was then spliced to the cable on board, and five ‘turns’ having been taken round the drum, the laying proceeded perfectly, even in depths of 1,640 fathoms, until the cable fell short some miles from land. With a view to avoid the heavy expense incurred by the hire of vessels, he had gone to sea with some parts of the machinery in sections, and these were not put together, until they arrived at the spot for commencing operations. This second cable weighed about 4 tons the statute mile.

One of the principal difficulties he had encountered in laying cables of great length, was the delivering of the cable in accordance with the progress of the vessel. It was of the utmost importance, that the way and speed of the ship should be ascertained, by the most accurate means, as currents, running 2 or 3 knots per hour, might be encountered, which would materially influence the ship’s course.

He had endeavoured to pay out the cable in accordance with the log, and had increased the strain, in order to economize length, but though this had been done for the last 60 miles, and though many on board were of opinion, that there would be 10 or 12 miles to spare, yet when within a short distance of the shore, the cable fell short. An order was instantly telegraphed through the cable to Greenwich, to manufacture the additional length of cable required; while the best means were adopted of holding on by the end, until some aid could be obtained. A vessel was despatched to Algiers for that purpose; but on the fifth morning,—almost the instant after receiving a telegraph message on board, from London, saying that the cable ordered five days previously would rapidly follow to the Mediterranean,—the cable parted in a storm, after having remained intact for five days, in a depth of 500 fathoms. Mr. Brett found it necessary to remain with the men at the breaks, and to keep them at their posts night and day, as when they became accustomed to handling the breaks, it was almost impossible to change them. The remarks in the second Paper, as to the necessity for an efficient and practical staff, in operations of this nature, could not be too strongly enforced.

Mr. Wollaston said, the first experimental line of submarine telegraph, from Dover to Calais, consisted of one, No. 14, copper wire, insulated with gutta-percha, both laid straight. The strain at times was such, that the wire was drawn down nearly to No. 16 gauge, and the gutta-percha was proportionately elongated. When the strain was removed, the gutta-percha had a tendency to return to its original length; the wire then forced its way through the gutta-percha in several places, in one instance from 2 to 3 inches of copper wire being exposed, so that the insulation was destroyed. That defect alone would have been sufficient to cause that line to be useless, within a few hours of its being laid down, He then suggested that, in a cable containing several wires, the core, including the conducting-wire, or wires, and the gutta-percha, should be twisted spirally, like the outer covering. That plan was adopted, and it was now universally followed.

It had been said, that when the lead weights were placed on that line, the ship had on each occasion been brought up. That was not quite correct. The speed of the vessel was only eased at such times; but it was never completely stopped. He considered, that the repairing of cables, in shallow depths, was perfectly easy and simple. In the case of the Dover and Calais line, he had taken up the cable, and joined it on board the vessel, whilst the communication was maintained between London and Paris.

Mr. Latimer Clark said, the remarks which he had to offer principally referred to some of the statements made by preceding speakers. It had been said, there was no doubt, that cables could be deposited “pretty exactly”[3] in the same way in water as in the air, and that “when a cable was payed out at an angle of 10°, there was a strain upon it of nearly sixty-six times the minimum tension,” or nearly sixty-six times the weight of a vertical section of the cable hanging from the deck of the ship to the bottom of the sea, where the cable had to be deposited. These opinions were, he ventured to say, not only erroneous, but were opposed both to practice and theory. The first of these assertions was illustrated by a model, by which it was shown, that when a chain hung vertically from a roller, on the roller being made to travel along the upper plane, the chain would be deposited in a straight line along the bottom, and it was argued, that nearly the same phenomena would take place in water as in air. If the course of any portion of a cable hanging perpendicularly from a vessel, and being deposited in water 2 miles deep, was followed, and if the vessel was supposed to move over the ground, and to pay out cable, at the same relative rate as in the model, then that piece of cable must in its descent move horizontally 2 miles, in its journey from the deck to the bottom of the ocean. Now, he contended, there was no force which would take the cable horizontally that distance. Or, if the case of a cable payed out at what was considered to be the proper angle (45°), was taken, even then any section of cable near the ship must travel forward seven-tenths of a mile, in order that the cable should be laid straight upon the ground. But there were other reasons against that theory. In the case of the Atlantic cable, it was stated, that 6 or 7 knots of cable were frequently payed out per hour, at an angle of 10°, whilst the vessel only moved forward 4 knots. That was inconsistent with the idea of a catenary, in which the strain was sixty-six times as great as the vertical weight of the cable. Again, when it was said that an undue strain was put upon the cable, a contradiction was implied; for, on the one hand, it had been assumed, that there was in water, as in air, a horizontal tension at the bottom of the ocean, and therefore the cable could be deposited straight, yet, on the other, blame had been imputed for the application of a strain of 30 cwt., in order to enable the cable to be payed out with horizontal tension. The reasoning would also lead to the inference, that in depositing the Atlantic cable at an angle of 10°, a strain must have been created near the ship equal to sixty-six times the weight of a vertical section of the cable of similar height, which, in the present instance, would give a strain of about 99 tons, whereas the breaking weight of the cable was known to be only 4 or 5 tons. He felt much indebted to the Authors of the first Paper, for their mathematical investigation of the subject, and agreed, generally, in their advocacy of light cables. Under the particular conditions and circumstances of the Atlantic undertaking, a cable of low specific gravity, though not necessarily small in size, was decidedly preferable. He had also no doubt of the correctness of the argument, that when a cable was payed out with uniform velocity, in such great depths as were likely to be encountered in the Atlantic, the cable must tend to descend in an oblique direction, though in a straight line, and be deposited in folds at the bottom. It was this tendency of the cable to run down through the water in the direction of its length, which caused the great strain, more especially felt with heavy cables, and which could only be resisted by the breaks, and the friction of the water upon the cable. But even a heavy cable might be safely deposited by the use of floats, or resisters, to which he attached great importance. He thought, that with the present Atlantic cable, they would constitute an excellent safeguard, and that, without such precaution, it would not be prudent to attempt to deposit such a cable, in a depth of water of 2 miles, and with 2½ or 3 miles of cable hanging between the deck of the ship and the bottom of the ocean. The kind of float which he advocated was simply a flat board, 3 or 4 feet square, with a long rope attached to the centre, by a bridle, like a kite. A number of these might be prepared, in the roughest manner, and when the cable was being payed out at the stern of the vessel, men might be stationed on a small gangway outside, to attach these floats, by forming an open knot round the running cable, and having lowered the board to the surface of the water, it might be drawn tight, and allowed to run. These would go down with the cable like a parachute, and would offer a resistance increasing in proportion to the velocity of the cable. It had already been stated, that a float 1 foot square, with the vessel going at the rate of 1 knot per hour, would offer a resistance of 31 lbs. It followed, from well-known laws, that a float 4 feet square, at 5 miles an hour, would give a resistance of about 12 cwt.; and if the velocity was increased to 8 miles an hour, the resistance would be increased from 12 cwt. to 32 cwt. He thought, therefore, that these were the most useful breaks that could be applied, to check the undue velocity of the cable. It was objected, that in the event of there being any necessity for hauling back the cable, these floats would increase the resistance; but he must say, while he doubted the possibility of hauling up the present Atlantic cable, yet, if it was ever attempted, he believed the floats would, by their buoyancy, rattier assist in the operation. The action of the floats was such, that a resistance was offered to the cable at different points, section by section, as they went down, instead of the whole strain being thrown on the point of suspension near the ship. If, with a cable travelling through the water at a rate of 5 miles an hour, one of these floats gave a resistance equal to 12 cwt., a small number would be sufficient to lighten, materially, the strain of the cable near the ship; and if there was any sudden increase in the velocity, the resistance would also increase, as the square of that velocity, and would thus have the effect of preventing the cable going out in jerks, with the rise and fall of the vessel, which it would do without their aid. If any argument was necessary, to show the superiority of a light over a heavy cable, it was furnished by the simple statement, appended to the Report of the Atlantic telegraph Company, that immediately after the fracture of the cable, soundings were taken in a depth of 2,000 fathoms. The fact of laying the sounding-line to the bottom, and bringing it up again, although it excited no surprise, was in reality much more difficult than that of simply laying a cable; and it showed the comparative safety with which a light cable might be deposited, or even be hauled up again, although with a heavy cable that operation would be attended with tremendous risk, even if it was possible. The Authors of the first Paper calculated, that the Atlantic cable, if payed out perpendicularly, and unchecked, would run out at the rate of 16 miles an hour; although, if laid horizontally on the water, it would only sink at the rate of about 2 miles an hour, thus showing the advantage of surface resistance. It had also been stated that, in the case of the Mediterranean cable, on one occasion, the cable overpowered the break, and ran out for five minutes, at the rate of 24 miles an hour.

[3] Vide ante, page 300.

Before the Atlantic Telegraph Company had decided on their plans and appointed their Engineer, Mr. L. Clark had been consulted, and had made various recommendations, which he thought, even now, might be usefully remembered, as while some had been adopted, others had not. He had advised the use of only one, large, conducting-wire. The first idea was to combine several small, separate wires in one small coating of gutta-percha. He thought the induction of one wire upon another would be a barrier to the use of more than one wire; and, although his views on this point were in opposition to those of the electrician of the Company, he felt confident, that the larger the conductor, the more easily and rapidly the cable would be worked. He also recommended, that the gutta-percha covering of the conductor should be as large as possible, so as to lessen the induction; and that the cable should be tested under water, before it was laid, as experience had proved, that it might have suffered mechanical injury, which could only be detected in that manner. He had likewise pointed out the importance of freedom from alloy in the copper conducting-wire, and Professor Thomson had since paid attention to that subject, and had published many valuable experiments. He had also suggested the employment of a lighter form of cable, made partly of hemp; and lastly, and more recently, the use of floats. Although he felt strongly that a light cable was more suited to deep water than a heavy one, yet he was not to be understood as giving his adhesion to the form of cable proposed by Mr. Allan. He agreed in the remark which had been made, as to the impropriety of employing iron wire within the cable; for, whilst iron wire possessed only one-fifth of the conducting power of copper, it added equally to the induction, and consequent retardation, by increasing the area which was subject to the induction. On these grounds he did not approve of that cable, although he strongly approved of a cable with a conducting-wire of a large size. There was nothing in the laws of conduction of electricity through wires, either in submarine cables, or on over-ground lines, which was difficult to understand. With regard to over-ground wires, the principle was identical with that which regulated the flow of water through pipes. For example, in sending a current of electricity from London to Edinburgh, the current possessed greater tension near London than near Edinburgh, and a certain portion of the current was lost at each post, like water from a leaky pipe; but as long as a little of the current reached its destination, it was sufficient to act upon the instrument; and as the battery-power was increased, so the power of the current was increased. In the case of the Atlantic cable, the first endeavour should be to make the instruments as delicate as possible, so that the smallest amount of electricity would move them. When this was accomplished, the next point to be ascertained was, the minimum current that would work the instrument at the distant end; and then the conducting-wire must be made sufficiently capacious to convey that quantity of electric current to its destination. In that instance, the length was so great, that a long time was required to supply the electricity in sufficient quantity to work the instruments. The natural remedy was to make both the wire and the conducting medium larger. So far Mr. Allan was right, but he should use copper instead of iron. Mr. Clark had paid great attention to the Atlantic undertaking, and he thought, if the breaks were not too heavy, and there was not too much material to be alternately stopped and thrown into accelerated motion, by every rise and fall of the vessel, and floats, or resisters were employed, there could be no insurmountable difficulty in laying that cable, and that the electric communication might be accomplished.

Mr. E. Highton stated, in reference to the durability of submarine cables, that the cable belonging to the British Telegraph Company was laid down about four years ago, and had remained perfect ever since. The only cost to the Company for maintenance, or repairs, was a small gratuity to the Coast-Guard, on each side of the Channel, who were instructed to give notice, whenever either end of the cable was exposed on the beach. He remarked, that simple cables with a single conductor had cost large sums of money for maintenance and repairs, while under the compound system, with two exceptions, but little, or no expense had been incurred. This should be borne in mind in designing submarine cables in future.

The plan of paying out a cable from the stern of the vessel was objectionable; for the stern, in rough weather, would evidently rise and fall considerably. The effect of this would be to bring great strain on the cable, unless it was allowed to pay itself out, to such an extent as to prevent any alteration in the catenarian curve, when much waste would ensue, a loss being sustained on the passage of every wave. The preferable part of the vessel for the cable to leave the ship appeared to be the centre, or the centre of gravity, by which all waste from the pitching, or rolling of the vessel would be avoided. A conical hole, with the apex of the cone tending upwards, should be made for the purpose, and then the cable would not be chafed, by rubbing, on its departure, on any part of the vessel. The almost certainty of a storm occurring, during the operation of laying down the Atlantic cable, rendered it the more desirable, that every precaution should be taken which could be devised to insure success. If it was not possible to alter a vessel, so as to allow the cable to be payed out at the centre, a ship should be specially built for the purpose.

The small number of words at present capable of being sent through the Atlantic cable,—the number being, according to the Company’s Report, only four per minute,—had induced Mr. Highton to devise a code system, for use in long lengths of telegraphs. He exhibited an instrument, which was capable of transmitting through a wire eight hundred million times two million preconcerted messages, the maximum period for the occupation of the wire not exceeding ten, or twelve seconds, if sent at the rate at which the Queen’s Speech was recently transmitted from London. He also explained, one of three instruments used in transmitting the American President’s last Message, from Liverpool to London, which consisted of upwards of sixteen thousand words, at the rate of three thousand five hundred words an hour. The desirableness of magnifying the effects of electricity arriving at a distant station, especially in the case of leaky wires, had led to the invention of an instrument for the purpose.

It had been found, that by exposure to the light and heat of the sun, the action of the mycellium of a fungus, and under other conditions, gutta-percha was rendered unfit for the insulation required for the transmission of messages by means of electricity. Several specimens of gutta-percha, in a decayed state, were exhibited, and also a piece of copper wire, 5 feet in length, covered with gutta-percha, which was strained until it broke, when the gutta-percha, owing to its partial elasticity, contracted, and left 7 inches of copper wire uncovered. A newly-made tube of gutta-percha, under a strain of 276 lbs., stretched from 14 inches to 24 inches before breaking; but a similar tube, which had been exposed, for about five years, in the atmosphere, to the light and heat of the sun, was so brittle as to be easily broken by the hand. He further contended, that when a spiral wire enveloped a soft core, the integrity, or electrical condition of the inner core was liable to be injured and be destroyed, in consequence of the straining and stretching, due, partly, to the weight of the protecting covering itself.

Mr. Allan said, it appeared to him, that the real subject-matter of the Papers,—the submerging of telegraphic cables,— had been lost sight of in the discussion. Nor had the mechanical construction of the ropes, hitherto used, met with the attention it deserved, seeing how important that point was in the successful issue of such undertakings. Attention had been very pointedly directed to the fact, that when a spiral wire enveloped a soft core, it must of necessity yield to tension, and that any strain upon it, consequent on its own weight, must, more or less, destroy the electrical integrity of the inner core. To this it had been replied, that in a compound cable with four conducting-wires, the core being spiral, it and the outer covering would yield alike. It was then explained, that a single-wire cable, like that used for the Atlantic, had been referred to; when the further curious remark was made, that on almost every occasion when the Atlantic cable had been tested, the outer protecting-wires gave way first. This statement naturally led to the deduction, that if the materials placed for protection gave way before the object to be protected, would it not be better to dispense with such protection altogether; especially when it was considered, that it was the weight of this so-called protection, which was one of the chief causes of the strain, tending to destroy the wires, and in consequence of the straining and stretching, to injure the insulation, and the integrity of the electrical conductor? Whilst the advantages of ropes of light specific gravity seemed to be generally admitted, it was yet remarked, that there had already been much experience with light cables, all of which had failed, including the first from Dover to Calais, the first from Holyhead to Dublin, and one from Portpatrick to Donaghadee. It should, however, have been explained, that there was not the slightest identity in the principles of their construction, and that they were early experiments in submarine telegraphy. The specific gravity of the Dover and Calais line was such, that it had to be weighted to submerge it; the cable used between Holyhead and Dublin was, no doubt, light, per mile, but from the small displacement of water, its specific gravity was about the same as that of the Atlantic cable. That from Portpatrick to Donaghadee was broken before half the distance was traversed. It should also have been stated, that upwards of fifty per cent. of all the submarine cables hitherto attempted to be laid had failed, chiefly in the act of submerging, or very shortly afterwards; and how far those at present in existence were in good working order might be reasonably doubted. Taking all these facts into account, he thought it would be apparent, that there was much yet to be done in this important branch of mechanical science.

Having been present at the laying of the first cable between Holyhead and Dublin, which gave way the next morning, Mr. Allan came to the conclusion, that the mechanical construction of the rope was at variance with maintaining intact its electrical condition as a conductor. To remedy this defect, he had proposed a cable which should be light both in weight and in specific gravity, and at the same time have the greatest strength in the centre. The weight being only half-a-ton, or less, per mile, a length of 2,000 or 3,000 miles might be conveyed in one vessel; and the specific gravity being low, all breaks, the source probably of four-fifths of the failures hitherto met with, might be dispensed with, when paying out. Such a cable could, from its lightness, be easily handled, be run out free over a pulley, like a hempen rope, be laid out horizontally on the water, and be payed out at a rate of eight, or ten miles per hour, instead of four, or five miles per hour. Thus a straighter course could be maintained, there would be less waste, and as the whole operation could be completed in about half the time usually occupied, there would be less liability of accident from foul weather coming on. The form of rope here alluded to was not liable to stretch, so that the insulating medium not being strained during the process of submerging, the electrical integrity of the conductor would be preserved; whilst the removal of the outer metallic covering would relieve the conductor, to a certain extent, of some of the Leyden-jar effects experienced in other ropes. The conductor itself being larger, must, it was believed, be more efficient for longer distances. The philosophy of the argument used in reference to the Atlantic cable, that the greater the distance to be traversed the smaller should be the conductor, was not one that could be credited.

Mr. Lionel Gisborne agreed in the opinion, that it was desirable that cables should be as light as possible, compatible with their sinking; but, at the same time, they must have strength as well as lightness, and it was as to where that strength should be applied, that there was a difference of opinion. It was said, that when the strength was entirely in the centre, a larger and a stronger road for the current was obtained, and one that could be kept in order with the least trouble and expense. In this view he could not concur. Assuming the light cable to be of the weight of half a ton per mile, as stated, of that probably 8 cwt. would be in the centre, and 2 cwt. on the outside. Now, it was to be feared, that when that cable was coiled in large masses in the hold of a ship, and was exposed to high temperatures, as it undoubtedly would be in a steamer, the heavy portion in the centre would be liable to sink into the surrounding soft material, and so to destroy the telegraphic communication. In the case of the Atlantic cable, coiled in the hold of the ‘Niagara,’ it was doubted whether the copper wire would not force its way through the yielding gutta-percha. Some experiments, made to ascertain how far that was the case, showed that the uncovered gutta-percha was very ductile, when exposed to a high temperature. Another objection to the particular form of light cable alluded to was, that when any strain was put upon the cable, by the necessary machinery to give sufficient retarding force in paying out, that strain would have to be borne by the soft gutta-percha, which would soon be squeezed through by the heavy conductor, and then the insulation would be destroyed. He contended, that the strength of the cable should be on the outside, and that it should be sufficiently strong, to allow of the requisite retarding force being applied, so as to enable the cable to be laid with success.

He regretted that so little attention appeared to be given to the paying-out machinery, the object of which was to prevent the cable running out too fast. He believed that, at no time, could there be more active weight of cable over the ship, than the perpendicular depth at which it was being laid; and that the strain upon the cable might be measured, by the difference between the velocity at which the cable was being payed out and the speed of the vessel. As long as the exact difference between those two speeds could be registered, the strain upon the cable, assuming it to be a constant weight, could be correctly ascertained. If, in a cable calculated for a depth of 2,000 fathoms, that difference was half a mile per hour, then the paying-out machinery must be such as to keep a retarding force upon the cable equal to half a mile per hour. But as the velocity of the vessel was variable, owing to the wind, currents, and other causes, the only constant quantity was the speed of the cable. Therefore, the measure of this difference in the paying-out machinery should be by means of a spring, or a pulley, carrying compensating weights, which by its rise and fall would indicate the increased, or lessened speed at which the vessel was going; the difference would then be correctly measured, and the strain upon the cable be regulated, so as to make it as uniform as possible. He thought that system was preferable to manual labour. He proposed, that between the hold of the vessel and the stern, the cable should be passed under a hanging sheave, to which a number of chains were attached, forming picking-up weights. When the perpendicular lift of the sheave was small, the weight upon the cable in the sheave would be small, whilst, at the same time, as the strain upon the cable increased, from whatever cause, the compensating weight of the chains pulled up would be an exact measure of the increased strain upon the cable, and would regulate the speed of paying out, so as to keep up a constant ratio between the speed of the cable and that of the vessel. If the vessel progressed at a uniform velocity, and the depth was uniform, the strain would be constant. If there was 10 per cent. of slack, and the cable required a holding force of 3 tons to suspend it over the side of the ship, 1 cwt. less than that on the drum would cause the cable to go out at a greater velocity than the speed of the ship; and, as long as the depth was uniform, no accident could occur to the cable. But as, in practice, neither the speed of the vessel nor the bottom of the ocean could be uniform, (although he thought the differences were not so sudden as some supposed,) what was wanted was a break, with a permanent retarding force upon the cable, and an apparatus which would indicate the various changes in the increasing and decreasing speed of the cable, and keep the ratio between the speed of the cable and that of the vessel uniform. This compensating weight, acting on a moveable pulley, effected that; and by having the upright groove marked with inches, or feet, or a simple graduated rod projecting up through the deck, the breaksman would always know the exact measure of the retarding force necessary to be applied to the break, which regulated the paying out of the cable. The hanging chains might be replaced by a spring; but a spring was liable to wear out. There were also other means, such as a carriage running upon the deck, and acting on a spring; but that might run off the rails, and the machinery get out of order. Suspended to the end of the lower lever, acting on the friction-break, was a short chain, and attached to the moveable pulley, was a small open box in a line with this chain, so that, should a sudden strain come upon the cable, from the pitching of the vessel, or any other cause, the moveable pulley would at once run up to the deck, and the chain on the lower lever, which was below the deck, being lifted by the box attached to the moveable pulley, the friction-band was at once relieved, and the drum could revolve more freely, and thus save the cable from any injurious strain. The permanent weight and the hanging chains acting as a retarding force, should together always be less than the breaking weight of the cable. The hanging pulley would give out, or take in, 45 feet of slack, which was nearly equal to the greatest probable rise, or fall of the stern of the vessel; and, as the apparatus was simple and self-acting, there was every reason to hope, that great security would be afforded, when paying out a submarine cable. Not the least important aim attained would be, that the breaksman had the means of knowing when the strain was in excess of, or under, that required to keep up the ratio between the speed of the cable and that of the ship, and also time to regulate the action of the break.

He doubted whether it would be practicable to attach to the cable the floats, or resisters, previously adverted to, in the manner described, whilst the cable was running out at 6 or 7 knots an hour.

Mr. C.W. Siemens had paid considerable attention to this subject. When assisting at the laying of the Mediterranean cable from Cagliari to Bona, his brother, Mr. Werner Siemens, had devised an apparatus similar to that just described, to regulate the strain on the cable, as it was payed out. The results were very favourable. It not only enabled the breaksman to regulate the strain upon the cable, with great nicety, by the deflection of the weighted lever, which rested with its pulley upon the cable, between the break-wheel and the stern-pulley, but it overcame, to a great extent, the bad effects arising from the pitching of the vessel. When the vessel pitched, the weight rose, and allowed more cable to run out, so that the pulleys of the break travelled at a more uniform velocity.

With reference to the best form for a submarine cable, he considered that there were several questions involved, which required to be balanced against each other. It had been proved satisfactorily, by the mathematical investigations in the first Paper, that a cable of light specific gravity was best suited for laying in great depths. But the cable was composed of several materials. There was the conductor, which, when of copper, had a specific gravity of 11; the gutta-percha insulator, nearly equal in weight to seawater; and the iron external covering, having a specific gravity of 7. Taking two cables of the same specific gravity: one might have little strength in the covering and a large central core; while the other might have a small core and great strength in the external covering. In analysing what produced weight in the cable, there came first the conductor, which, for electrical reasons, ought to be of the best conducting material, such as copper. The conductor constituted the weight to be carried, and should, therefore, be as light as possible, consistent with the highest conducting power; and, to insure its continuity, it should be relieved from strain by the external coating. He thought that the newly-discovered metal, aluminium, might be used, with advantage, in deep-sea cables, as it was nearly equal in conducting power, and was only one-third the weight of copper. If the proposal to substitute an iron conductor should ever be adopted, it would be found, that the retardation by lateral induction, which was the great impediment to the successful working of long submarine communications, would be much increased, and would eventually become practically insurmountable. After the conductor was determined upon, then came the consideration of the insulator, the thickness of which ought to be increased with the length of the cable, in order to keep down the retardation by lateral induction. The insulated conductor, if composed of copper and gutta-percha, was always specifically heavier than water,—and it was the outer covering which must give strength to the fabric. If the outer coating was of soft material, such as caoutchouc or gutta-percha, there was no strength to resist the action of the break, and the coating would be torn away from the wire within it. Therefore the outer coating should be of hard material, and of great strength, so as to resist the longitudinal strains during the process of submerging, but it should add as little as possible to the weight. He thought that no material fulfilled these conditions so well as soft steel. A thin steel wire covering would produce a cable of the least weight, and capable of suspension in the greatest depth. Nor would it be more expensive than the iron coating, if power of suspension was taken as the basis of the calculation.

Mr. John W. Brett stated, that before the present Atlantic cable was decided upon, from twenty to thirty specimens of different forms of light cables had been tested for strength. Some of these had an outer covering of hemp only, others of hempen strands containing threads of steel wire, one was the present cable, and there was one having a coating of steel wire. These tests were made on three several occasions, at the works of Messrs. Brown, Lenox, and Co., by the usual hydraulic machine for testing chain-cables, which admitted a length of 6 yards at a time. He had been always strongly opposed to cables having a light covering, and he thought they had only to be tested in that way, to lead to their immediate abandonment. The form adopted was, with one exception, the strongest. A small cable, having three conductors, with a steel-wire covering, and weighing little more than 2 tons per mile, bore a strain of 24 tons: these conditions were far superior to any cable hitherto tested; but in addition to its great cost, it was ascertained, that it would have been impossible to obtain the necessary quantity of wire in less than from one to two years. The chief consideration in submarine cables was to protect the conducting-wire, or core, from outward pressure, or other external injury, and from leakage of water. It was proved by these tests that, in the specimens covered with a spiral lay of iron wires, when subjected to the breaking strain, the core, the conductor, and the gutta-percha were perfect; whilst in those covered with hemp, or hemp and steel wire, the gutta-percha was so cut and injured, as nearly to destroy the insulation. In one instance the core, of gutta-percha and copper, was stretched nearly 11 inches without separating, or the insulation being injured; and in all cases, the outer protecting wires broke before the copper-wire conductor was affected. In the Atlantic cable, the repeated coverings of gutta-percha, in place of a single coating, and the adoption of seven twisted copper wires for the conductor, were important improvements; for it was improbable that flaws should occur in each of the seven wires at the same place. When paying out this cable last year, the ‘Niagara’ on two occasions hove to, in depths of 1,900 to 2,000 fathoms, until the cable hung vertically over the stern, without the least injury resulting.

With regard to the failures in the attempts to lay a heavy cable between Sardinia and Africa, in the years 1855 and 1856, they were due, in both instances, to want of length. In the first instance, after having payed out 60 miles successfully, from a sailing-vessel, it was found that the quantity used was so much in excess of the distance run, that it would be impossible to reach the land. He then decided to cut the cable, although perfect, in order to save the remaining length of 84 miles which was on board. In the other case, after a heavy cable had been successfully submerged in depths of 1,640 fathoms, it unfortunately fell short about 12 miles from the land. They then held on for several days and nights, by the electric cable, and contended with a rough sea, corresponding, from the end of the cable on board, with Paris and London, until the cable parted at the bottom, at a depth of 502 fathoms. It was all-important, in enterprises of this kind, to ascertain, with accuracy, the relative speed of the ship and of the paying out of the cable. In the case last alluded to, the lead was hove every quarter of an hour, and the length of cable payed out was ascertained from the number of revolutions of the drum. By these means the breaksmen regulated the tension. He believed, that the greatest speed practicable, in ordinary cases, would be found to be from 5 to 6 knots per hour. The idea had occurred to him of using floats, or resisters, constructed of iron, and in form something like a parachute, so that they might be laid one within the other, and thus save space. He preferred, however, not to use them, and without such aids depths of 1,640 fathoms had been successfully accomplished. He thought, however, that it might be prudent, in an operation like the laying of the Atlantic cable, where a distance of upwards of 1,600 miles had to be traversed, to employ a good system of buoys at every 300 or 400 miles.

With respect to the durability of electric cables after submersion, he produced part of a kedge-anchor he had purchased about twelve years since, which had been recovered from the bottom of the sea. The oxide of the iron had attracted a solid covering of concrete, not only over the iron, but entirely covering the knot of rope around it, which the oxide had penetrated. The wood- and iron-work of a pistol fished up off Tilbury Fort, were entirely covered with concrete from the same cause; the brass portions only being exempt. These illustrations satisfied him, that the Atlantic cable, when once submerged in the still depths of the ocean, would, before it was destroyed, form for itself a protecting covering. And though gutta-percha, when exposed to the atmosphere, or laid under-ground, might be subject to decay, yet, when submerged in the ocean, it was found to be perfectly preserved, as was proved in the case of the Dover and Calais cable.

Mr. Gravatt illustrated his views on the diagram-board, and expressed his regret at having been unable to be present at the reading of either of the Papers. With regard to a cable falling through water, he remarked, that for a cable whose diameter was d feet, and specific gravity s, the angle formed with the surface of the water being a, and the terminal velocity in a vertical direction v feet per second, the following formula v = √64 d (s-1) sec a nearly, was easily obtained. For a sphere whose diameter was d feet, specific gravity s, and terminal velocity in a vertical direction v, the formula was v = √81 d (s - 1) nearly. From a tolerably careful measurement of a specimen of the proposed Atlantic cable, it appeared that, d = .05 and s = 3.5; whence the terminal velocity v was 2.83 √sec a nearly. For a sphere, where d = .05 and s = 3.5, v = 3.16 nearly. These results were only premised, as calling attention to the scale of the operations, which was of great consequence when only practical truth was required.

It could easily be proved, that if a chain lay upon any inclined surface, whether straight, or curved (curved in plan as well as in elevation, a curve of double curvature), it would be exactly balanced, that was, just prevented from slipping, by the weight of another portion of the same chain equal in length to the vertical height of the inclined surface; and this would be the case, whether both portions of the chain were in air, or both immersed in water, or other fluid. If, therefore, from a vessel moving with any uniform velocity, a cable was continuously delivered into the water at an angle a, and with a vertical velocity vt, then, provided it was restrained at the water surface in a tangential direction, by a force equal to the weight in water of a portion of the cable, whose length was equal to the depth of the water, every particle of that inclined cable must fall in a vertical direction, and the form of the cable would be a straight line, inclined to the surface of the water at an angle a.

If the bottom of the sea was horizontal, then what might be called the waste of cable would be equal to the difference between the hypothenuse and the base; or if m was the number of miles a vessel run, then the waste would be equal to m (sec a - 1), a quantity, in practice, certainly not more than would be required to cover the irregularities of the bottom, which could not be expected to be really either horizontal, or even.

Now, supposing vt to be somewhere about 3 feet per second, as it would be with the Atlantic cable, it would naturally be seen, that no deviation from the conditions here supposed, which could practically happen, would materially alter the form, or condition of the cable, the more especially as if, from any pitching of the vessel, the cable should be made to deviate from a straight line; then, as vt = √ sec a x a constant, all portions of the cable which were less inclined than the mean angle would be falling, so to speak, too fast, and all portions more inclined would be falling too slow; and by the connexion between them, in the very long time (about one hour in water 2 miles deep) that each elementary portion took to arrive at its resting-place, the less-inclined fast portions would become more inclined, and slower, and the more-inclined slow portions would become less-inclined, and faster; until the mean angle and mean velocity were obtained, or nearly so. The greatest strain on the cable would be independent of the velocity of the vessel, and for the same cable depended only on the depth of the sea. The effect of a horizontal cross current would, it was obvious, be merely to curve the supposed inclined surface, on which, by virtue of its terminal velocity, the cable might be supposed to lie, in a horizontal direction;—it would slightly increase the ‘waste,’ but could not affect the strain. If the strain on the cable was less than that due to the depth of the sea, the cable itself would, so to speak, slip down the supposed inclined plane, and occasion a waste greater in proportion as the strain was less than it ought to be. If, on the other hand, the strain was a little greater than that due to the depth of the sea, a portion of the cable at the foot of the incline would assume a form approaching a catenarian arc; and if the strain was needlessly increased, the whole cable would assume a form resembling a catenarian arc, but then no known material would be sufficiently strong for such purpose. The necessity for, or even the fear of, enormous catenarian strains was therefore groundless.

The idea of breaking the fall, or lessening the strain on the cable, by means of floats, might be thus illustrated. Suppose, from a vessel moving uniformly with any velocity, a number of spheres, or beads, of any size, or specific gravity, was delivered into the water one after another, each with the due vertical terminal velocity vt; then the beads would each fall vertically to the bottom with uniform velocity, and would always form a straight line, which might be connected, or be supposed to be connected, by means of an unresisting thread; if each portion of this thread was equal in length to the hypothenuse between the beads, then it would not be exposed to any strain, nor would it produce any strain on the vessel. The beads would all be found arranged on the bottom of the sea, at the same horizontal distances apart they were dropped in, and the thread, which before formed the hypothenuse, would lie upon its base. A cable of such, or similar form, however deficient in strength, could therefore be laid in any depth of water whatever. He understood that the Atlantic cable weighed in water 15 cwt. per mile, and was calculated to break with a strain of 4¼ tons. By the foregoing reasoning, if the deepest part of the sea was taken at 2 miles, it would there require a strain of 1½ ton, or a little more than one-third of the breaking weight. Nor should it ever be exposed to any much greater strain, provided only the break-apparatus was properly contrived, and efficiently worked. The nearer the break-apparatus could be made to approach in its action, that exceedingly delicate and simple apparatus the fisherman’s fly-rod and winch, the more perfect it would be. When the vessel was progressing at the rate of 6 miles per hour, then, as the cable would fall about 2 miles in that time, the tangent of a would be one-third and the secant 1.054, so that ‘the waste’ would not be more than 5½ per cent. When it was remembered, that the length of even so large an arc as one of 90° was only about 11 per cent. longer than its chord, it would be seen that the waste, occasioned by any currents likely to be met with, must be very moderate.

Steel wire was, however, much better adapted for deep-sea operations than iron wire, inasmuch as the modulus of tension for steel was 6,700 fathoms, whilst for iron wire it was only 4,000 fathoms, and for bar iron only 3,000 fathoms.

Mr. Allan explained, that he had distinctly stated, that one of the greatest and most obvious advantages of ropes of light specific gravity, constructed in the manner he advocated, was that all friction-breaks, and other retarding machinery, could be entirely dispensed with; so that the liability of accident from that source, whether in the actual process of paying out, or as affecting the electrical condition of the cable, would no longer exist.

Mr. Rennie called attention to the effects which the heavy waves of the Atlantic would have upon a vessel, such as the ‘Niagara,’ from the stern of which a telegraph-cable was payed out, as compared with paying it out through a hole in the middle of the vessel. This he illustrated by a model, showing the ‘Niagara,’ which was 345 feet in length, when passing over the crest of a huge wave, rising 10 feet above the mean horizontal level and falling 10 feet below it, or together 20 feet in height, and from 600 to 700 feet in length. In that case the stern, from which the cable was payed out, would, according to his views, rise, or vibrate, 75 feet, thus causing sudden and violent tension of the cable; whilst if payed out through the middle, or centre of oscillation of the vessel, the rise would only be 20 feet. About two years back he had had a conversation with Mr. Brett, relative to the accident which had occurred, while paying out the first Mediterranean cable, which, after having been successfully laid to within a few miles of its termination, broke from the vibration of the stern during bad weather. Mr. Rennie then suggested, that the next cable should be payed out through the middle of the vessel, and this principle he had advocated ever since. He exhibited a model of a ship’s hull, similar to the ‘Monarch,’ but flat-bottomed, which he considered more suitable to the purpose of paying out marine cables. The opening in the bottom was made trumpet-mouthed and smooth, so that whether the vessel was lying over on one side, or the other, the cable could pass out, without meeting with any obstruction. If this mode of paying out the cable, through a hole in the centre of the vessel, combined with a self-governing break, such as had been proposed, was adopted, the danger of the cable breaking would, in future, be greatly diminished. He was aware that there was a difference of opinion as to the height of waves. According to some authorities, the waves in the Atlantic never exceeded 12 feet in height, whilst others asserted they were as high as 50 feet, and some said even more than that. He had assumed 20 feet as an average height; but in future operations, he thought that this subject was worthy of consideration by those engaged. The waves of the Mediterranean were, however, very inferior in magnitude to those of the Atlantic ocean.

Mr. Rennie produced a coloured print by J. Daniell, the artist, representing a vessel passing over the summit of a wave during a storm in the Atlantic, which, from his own experience, he did not think was exaggerated.

Mr. Cyrus Field had crossed the North Atlantic many times in the winter, and had never seen such waves as those referred to. On one occasion, in the month of January, when in the ‘Persia,’ he had asked the opinion of the commander, Captain Judkins, as to the extent to which the stern of the vessel rose and fell in a heavy sea, such as was then running; when he received the reply, that it did not, at that time, exceed 3 feet rise and 3 feet fall. He believed the rise and fall, from a dead level, of such a vessel as the ‘Persia,’ or the ‘Niagara,’ in crossing the Atlantic, never reached 10 feet.

In discussing a question of this kind, it certainly did appear strange, that those who advocated a particular form of light cable, should discard, altogether, the electrical portion of the subject; for it appeared to him, that one of the first things to be considered, in regard to a submarine cable, was its conducting power. If the best kind of pipe to conduct water into London, or elsewhere, had to be determined, he thought it would be considered necessary to make it in such a way, that the fluid would flow freely through it. Now, he did not hesitate to say, that with the light cable referred to, it would be almost, if not quite, impossible to pass an electrical current through it from Europe to America. This opinion was confirmed by Professor Faraday, who had stated, in a previous discussion on this subject at the Institution, that “the larger the wire, the more electricity was required to charge it, and the greater was the retardation of that electric impulse, which should be occupied in sending the charge forward.”[4]]

[4] Vide Minutes of Proceedings Inst. C.E., vol. xvi., p. 221.

Sir Charles Fox said, it appeared to him, after much thought upon the subject, that the main difficulty to be encountered in laying down a telegraph-cable, in deep water, was the probable existence of deep-sea currents. If it was certain that no such currents existed, then, he thought, a cable of a little greater density than water might be adopted, in which case the operation of sinking would be slow, and could be conducted with the greatest certainty, and with the smallest expenditure of cable. But if, as it sank, it came in contact with an under current of considerable power, the cable would, in all probability, be broken; therefore, he deemed it of the first importance, to make a careful investigation as to the existence, or otherwise, of deep-sea currents, as on this, he thought, rested the greatest source of responsibility. He felt sure, from his own observations, repeatedly taken, during two voyages across the Atlantic, of the beautiful straightness of the wake from the steamer to the horizon, that the currents on the surface were of very small velocity. If a cable had to be laid where deep-sea currents were known to exist, it would be desirable to have it as small as possible, and, at the same time, of great strength and of high specific gravity, regardless almost of the question of expense, so as to insure its passing through currents quickly, thereby presenting only a short length of cable, in the act of sinking, to the action of the force of such currents; and he thought a covering of steel would best meet these requirements. With regard to the conducting medium, that which had been adopted in the Atlantic cable was perhaps as good as could be devised. It had been believed, that the induction might be somewhat reduced, by drawing the seven wires together, through a circular die, so as to close them up, like a ‘patent axle;’ but as the induction would not be much lessened, whilst the wires would by this operation become harder, and consequently more liable to break, he considered it an undesirable step to take. The use of copper (not iron) wire was essential, and the mode of insulating was as good as could be desired. He thought the present amount of knowledge was not sufficient to enable any one to say, whether deep-sea currents might be disregarded, so as to justify the use of cables covered, not with iron, but with longitudinal fibres of hemp (which, weight for weight, was double the strength of iron), with a close serving of hard yarn, well soaked in pitch and tar. These were matters which could only be determined by further research. If no such currents existed, he should consider by far the greater part of the difficulty to have vanished.

Mr. Macintosh considered the rise and fall of the stern of the paying-out vessel, caused by the undulations of the sea in rough weather, to be the great source of the disasters that had hitherto befallen submarine cables. In the attempt to lay the Atlantic cable last year, 380 miles had been successfully payed out, during fine weather, 100 miles of which were in water 2,000 fathoms deep; hut when a breeze sprung up, the ship pitched, and from the cumbersome nature of the paying-out machinery, and the rigidity of the break apparatus, the inertia was so great as to cause the cable to snap. If the sea had remained smooth, and free from undulation, it was more than probable, that the whole cable might have been laid from Valentia to Newfoundland. When a vessel was progressing regularly at the rate of 6 knots an hour, the motive force which caused the drums to revolve remained uniform; but if a sea struck the ship on the quarter, or otherwise, it was essential that the apparatus for paying out should be as light as possible. He, therefore, proposed that the break should only exert a uniform, constant retarding force, and that the additional strain due to these causes should be compensated, by suspending the delivery-pulley, by vulcanised India-rubber springs, in such a way that there should be always 50 feet or 60 feet of cable in reserve, without resorting to the break. He proposed having a spar under the quarter of the ship to which the delivery-pulley should be attached, so that when the ship rose the weight of this spar should hold down the pulley, and keep it constant with the water, although inconstant with the ship. This apparatus was designed to supersede the necessity of releasing the breaks, for however sensitive the machinery might be, upon the vessel suddenly rising, or falling, 20 or 30 feet, the strain was sufficient to endanger the cable. If a strain of a ton, or more, was required to retard the cable, the friction breaks should be fixed at that amount; and as there was always a certain amount of cable in reserve, the strain would remain the same. A vessel steered very wildly in a heavy sea. Such a vessel as the ‘Niagara’ could not get away from the following wave; and when struck by a heavy sea, great strain was brought upon the cable, tending to break it. An attempt had been made to recover, by under-running from the west coast of Ireland, the 380 miles of the Atlantic cable which had been lost. So long as the weather continued fine, the operation was successful, but after about 53 miles had been recovered, the cable broke, from the same causes that happened during the laying. He had since arranged with the Company to pick up the remainder of that cable in a depth of water of 140 fathoms, with his apparatus; and when the end was secured, which formed no part of his contract, he saw no difficulty in the undertaking.

With regard to the construction of the cable, and the insulation of the conducting-wires, he suggested that after the centre core had been insulated in the usual way with gutta-percha, great strength might be obtained by embedding fibres of flax, hemp, or cotton, in an outer covering of gutta-percha, or other insulating material. This could be done with great pressure, between rollers, which would consolidate and close up every pore. This covering might be subjected to treatment, which would enable it to resist tropical heats, and afford quite sufficient protection from any blow, or rough usage.

Mr. A. C. Hobbs remarked, that as, unfortunately, the water was very inconstant, he could not see how the arrangement just described could compensate for the rise and fall of the water, even though the spar might remain on the surface.

Mr. Macintosh explained, that the spar floated on the surface of the water, and was bridled to the ship in such a manner, as to allow of its rising and falling with the water; consequently as the ship rose the beam lowered. The delivery-pulley being fast to the spar, was, therefore, constant with the water, but inconstant with the ship.

Mr. Alfred Varley said, the principal issue raised in the first Paper, was as to the relative merits of light and heavy cables. Throughout the discussion, there had been very few remarks upon light cables, for the simple reason, that although there had been a considerable amount of experience with heavy cables, yet there had been scarcely any with light ones. The light submarine telegraph, consisting of a simple gutta-percha-covered wire, between Varna and the Crimea, had been alluded to in so cursory a manner, that the full practical value of that experiment had not been elicited. It had even been spoken of as a failure, although it had answered the purposes for which it was laid down, and was the longest cable, with one exception, that had yet been successfully submerged. He thought, therefore, that a more minute detail of this cable would be interesting to the Institution. The cable consisted, throughout the greater portion of its length, simply of one No. 16 copper wire, served with gutta-percha a little less thick than the core of the present Atlantic cable, and wholly unprotected. The shore ends had an iron sheathing, extending to a distance of 10 miles from the Varna shore, and of 6 miles from the Crimean coast. Although the cable was submerged under considerable difficulties, “during a storm,” and the length of the cable, he learnt from the report of Colonel Biddulph, R.A., the head of the telegraph department, was 340 miles, yet it was laid so straight, that the direct distance was only exceeded by 3¾ miles.[5] This reflected, undoubtedly, the greatest credit upon those who directed the expedition, and gave, at the same time, abundant proofs of the facility with which light cables could be submerged. Having been in the Crimea during a portion of the time that this telegraph was at work, in charge of the Field Telegraph, and subsequently in charge of the other link of this cable, between Varna and Constantinople, Mr. Varley could state that its insulation was very perfect; and it remained in that condition for nearly twelve months, during the period of the Russian war, notwithstanding the many violent storms to which it was exposed in the Black Sea, until during a storm of more than usual severity, it was broken on the 5th December, 1855. Trials were then made to ascertain where it parted. From the Crimean station, the usual amount of induced charge was observed as when it was sound, whilst from the Varna end no induced charge was perceived. Colonel Biddulph’s testings, by the resistance it offered from the Crimean station, indicated the fracture to have occurred at 230 miles from the Crimea, but this could only be considered as an approximation. The Varna shore end was then under-run for a distance of 6 miles, when circumstances unfortunately prevented the operation being continued. But as no induced charge could be perceived from the Varna end, there was little doubt of the fracture having occurred very close to Varna. As this unprotected cable had remained uninjured for nearly twelve months, there could not have been any disturbance at the bottom of the sea, otherwise it must have been cut through, by continual friction. No portions of this cable had been picked up; but portions of a similar one, which had been submerged for some time in the Black Sea, had been recovered, and in these the gutta-percha was quite sound, and was covered with a thin white deposit. The Varna and Crimea cable having evidently been broken by the violence of the storm, led to the conclusion, in his mind, either that the shore ends were not carried far enough out, or that the depth of the Black Sea, 70 fathoms, was not sufficient for an unprotected cable. He was well aware, that a cable adapted for a depth of 70 fathoms might not be suitable for a depth of 2,000 fathoms. Nor did he wish to appear as an advocate of any particular class of cable; because his experience had not been sufficient to warrant him in forming an opinion on the subject.

[5] Vide also “The Atlantic Telegraph, &c.” 8vo. Pamphlet. London, 1857, pp. 52 and 53.

Fully concurring in the remark, that the whole gist of the matter lay in obtaining an efficient conductor, he should be glad to say a few words on the electrical portion of the subject. He believed erroneous conclusions had been arrived at, by the projectors of the Atlantic Telegraph, which were the more serious, as coming from those who had extraordinary facilities for conducting experiments, they must necessarily carry weight with them. In a work published under their authority, he found the following passage:—“‘The law of the squares’ may possibly apply to the transmission of electricity freely along simple conducting-wires, but it certainly does not apply to the case of its transmission along submarine, or subterranean gutta-percha-covered wires, (the facility of transmission being estimated by the rate of speed,) because in this the case is not one of simple conduction, but of transmission, after the wire has been charged inductively to saturation as a Leyden jar.”[6]

[6] Vide “The Atlantic Telegraph,” &c., page 23.

This was the reason, probably, why it had been concluded, that a small wire was a better conductor than a large one, for submarine telegraphs, for if a jar had to be charged to saturation, the smaller the jar the more quickly would it be filled. But Mr. Varley submitted, that the conditions between a submarine circuit and a Leyden jar were not precisely the same. At the same time he did not wish to be understood as believing in the law of squares, for he knew of no electrical phenomena that followed that law. In a Leyden jar, the inner and outer coatings were perfectly insulated one from the other; but in a submarine wire they were not so. If the instrument was disconnected at the further end and the wire was sealed up, then they would be almost precisely alike; but, in practice, the wire was always open, through the instrument, to the earth, and the resistance opposed to the passage of the current, by the very long length of wire, was the only insulation between the inner and outer coatings. Therefore, if the wire offered no resistance to the passage of the current, there would be no insulation, and consequently no induction to cause retardation; if it offered a little resistance, then there would be a little induction; and precisely in proportion to the resistance offered by the wire, provided always the insulating medium was of the same thickness, would the induction manifest itself. He believed the error, in supposing that a large wire conducted more slowly than a small one, arose in this way. It was true, that when the area of the wire was increased, the inductive surface was also increased, at the same time, but not in the same ratio; for if the diameter of the wire was doubled, although the area was increased four times, and the resistance was reduced to one-fourth of what it originally was, yet the surface, and consequently the inductive force, had only been doubled: hence the relative proportion of induction would be only half of what it was before. If four cables were worked through as one, it was evident, that there would only be one-fourth of the resistance, and that a battery of one-fourth the energy would pass the same quantity of current through the four combined, as could be done through one with four times the power. The signals would not, however, be obtained more quickly through the four than through one alone, for although there was only one-fourth of the resistance, the surface, and consequently the amount of lateral induction, had been increased four times. But if they were merged into one, it was evident that the resistance would be the same, whilst the surface would be reduced to one-half. Hence the relative amount of inductive force being reduced to one-half by doubling the diameter of the wire, he believed signals would be obtained twice as quickly through a wire of double the diameter, and coated to the same depth with gutta-percha, as through one of half this size, and coated with gutta-percha to an equal extent.

Mr. C.E. Browning described a proposed plan for regaining the sunken end of a submarine cable, in case it should break while being payed out. Several floating-buoys were to be carried on board the paying-out ship, and fixed at intervals to the cable; the greatest tension being on that part between the water and the stern of the vessel, it would most probably break at that point, and to regain the sunken end it would only be necessary to return to the last buoy, and pick it up. A ship should follow to release the buoys, that they might be used again. He described the method of fixing the buoys to the cable, and considered there would be no difficulty in disengaging them, when they had performed their duty.

Mr. G.P. Bidder, V.P., said, the great experiment of the Atlantic cable had the best wishes of the profession, as all engineers felt most deeply interested in it. He confessed, however, for his own part, that the experiment had disappointed him. He did not think, that sufficient attention had been paid, either to the mechanical appliances, or to the construction of the cable itself, which the vast importance of the undertaking demanded. Neither had any very authentic records been made public, of what took place immediately previous to the fracture of the cable. It had been stated, that it would occupy two years to manufacture sufficient steel wire to cover the Atlantic cable. He thought it was a strong assertion, that the resources of this country could not supply that amount of steel wire in less than two years. Neither did he believe the statement, as to the amount of additional strength that the steel wire would give. It had been asserted, that whilst the present cable, with an iron wire covering, was equal to a strain of upwards of 4 tons, with steel wire it would be equal to a strain of 12 tons, being no less than 8 tons additional, as due to the substitution of steel wire for iron wire. Now, the sectional area of the wire in that cable was not above 1/10th of an inch; and, although he believed steel wire was equal to a strain of 60 tons per square inch, he was of opinion, that it would not give 80 tons additional strength over iron wire. He must also be allowed to say, that he was wholly at a loss to conceive why the Atlantic cable had never been submerged in its entirety, and tested in that state, to ascertain its condition. When the cable was submerged to a depth of 2,000 fathoms, it would have a pressure upon it of 3 or 4 tons to the inch, and what might be the effect of that pressure upon the gutta-percha core, in that iron case, he would not predicate; but he thought the experiment ought to have been tried, before any attempt was made to lay the cable. The failure of last year was stated to have occurred, when the cable was momentarily left in the charge of a workman; but it was much to be deplored, that the whole fate of so great an enterprise, got up at so much cost, and upon the success of which so much depended, remained, even for an instant, in the care of unscientific hands. Nor did it appear to have been necessary, for he knew there was an experienced member of the scientific staff on board the ‘Agamemnon’ at the time, where his services could not then possibly have been required. He considered, therefore, that neither the shareholders, nor the public, had been done justice to in that undertaking; an omission which he hoped would be corrected, when the next great experiment was made.

The Institution was under very great obligations to the Authors of the first Paper under discussion, who were the first to bring anything like science to bear upon the subject, to treat it in a logical manner, and to submit it to mathematical and mechanical analysis. At the same time, there were so many novel elements involved in its consideration, that it was hardly possible to do more than attempt to approximate to general principles, without fixing the exact amount of strain that the cable would be subjected to.

In the commencement of the discussion it had been said, that it was impossible there could be any ‘waste,’ if the cable, like the chain in the model, was payed out in a vertical direction; and this position had been illustrated by a model, in which a chain was suspended, and was payed out from a cylinder, showing exact agreement between the length payed out and the distance traversed. Now it was impossible for the chain to be payed out faster than the cylinder travelled: therefore, the length of the chain and the bottom of the ocean must be exactly measured by the number of revolutions made by the roller. If, however, the chain had been coiled upon a larger, or a smaller part of this cylinder, either the whole operation would have been brought to a stand-still, or more chain would have been payed out, than was due to the distance traversed by the cylinder. Moreover, a chain passing over a pulley controlled by a break, could not run on at a continually-accelerating velocity, if the strain upon the break was less than the weight of the suspended portion of the chain; and, ultimately, the velocity of the chain would overtake any speed at which the vessel carrying the apparatus might proceed, and even exceed it. It had been said, that although that was the case in the atmosphere, the circumstances would not be modified, to any material extent, in passing through water. Now, a bar of iron, one inch in diameter, and 2 miles in length, would weigh 13 tons; and when this was suspended vertically in the water, the resistance to that bar through the water would amount to 50 tons, at 5½ miles per hour. How far the observations referred to could be relied on, when such elements as these were neglected, the Members might judge for themselves. He wished to invite particular attention to the facts stated, that when a bar of iron, or a chain, or a rope, of one inch sectional area, was suspended vertically from a vessel, in water 2 miles deep, then the weight would amount to 13 tons, and the resistance due to that depth, if the vessel was moving at the rate of 5½ miles per hour, to 50 tons. When the vessel got into motion, that bar, or rope, would immediately assume an inclined position, and the inclination of the bar would adjust itself to the strain, and, in the case cited, might be at a slope of 4 to 1. If, however, the specific gravity was reduced, and the position of the bar was made to approximate nearer to the horizontal, then the strain from the vessel would be much reduced; for, whatever might be the inclination of the rope going through the water, the resistance would be as the square of the sine of the angle. The Atlantic cable being about 5/8ths of an inch in diameter, and 50 tons of resistance being due to one inch diameter, the resistance to it, at the depth stated, would be about 30 tons; and, as the breaking weight of that cable was about 4 tons, it was clear that the cable must be inclined, so as to reduce the strain, probably, to about 2 tons. Therefore, as 2 tons were to 30 tons, so was the square of the sine of the angle to radius; from which it resulted, that the least length at which the cable should be payed out was, from 8 to 9 miles from the vessel to where it touched the bottom of the ocean. Whenever the speed of the vessel was reduced, then the cable would have a tendency to assume a vertical position; and this would bring into play the direct weight of the cable, more or less, in addition to the above resistance, so that the strain on the cable would be increased. The results of his consideration of the subject had led to the belief, that a much lighter cable than that now proposed, would be found most advantageous for long distances at great depths.

Mr. J. S. Atkinson thought the value of the advice just given could hardly be over estimated; that in all new undertakings, especially when of such magnitude as the laying of the Atlantic cable, and on the success of which such important interests depended, there was a great necessity for actual experiments, on a large scale, in order to fix the data on which to proceed. Under ordinary circumstances, engineers would allow sufficient margin for unforeseen difficulties; but where there had been no previous experience, it was difficult to estimate the proper margin to be allowed.

If a submarine cable was allowed to pass free over a pulley into still water, gravity being the only force causing it to sink, the breaking strain would be that due to its own weight, less the buoyancy due to an equal column of water. This quantity would not be affected by the speed of the pulley; but the friction would differ with varying velocities, as would also the strains caused by the passage of knots, or splices, and the blows against the pulley sides, or flanches. Again, the heave and roll of the ship would vary the velocity, and thus also the strain. Permanent strains might arise from the bridging over of any ocean valleys, having abrupt sides. During also the change from one velocity to another, caused by the rope entering side, or cross currents, much extra strain would have to be borne; and likewise when meeting a current; for it must be remembered, that the cable, in the process of laying, was not moved horizontally by the ship passing under it. He believed, that experiment alone could determine the values of these several strains. No cable could be constructed strong enough in itself to bear them; they must be met by loss of cable and by strength combined. He did not think the Atlantic cable was strong enough. Its own weight in water was, he understood, about 16 cwt. per mile, which, at the greatest depth, would give, say, 37 cwt. The force of a current on that cable, when nearly vertical, would, he estimated, at 3 miles an hour, amount to 2½ tons; and allowing only 1 ton for the pull of a head current, at 1 mile per hour, before the velocity could be changed to meet it, there would be a strain of nearly 5½ tons, or 1 ton more than, as he understood, the cable was calculated to bear. Increased strength could only be gained by varying the specific gravity; and, he believed, it would be found dangerous to make deep-sea cables of a specific gravity of more than twice that of water. He thought the chain-weights proposed for the paying-out machinery, described in the course of the discussion, would be too sluggish in their action, to save the cable from any sudden strain; and he suggested, that an air-cylinder with a light piston, so connected with the break as to free it instantaneously, if necessary, would be better; and that the break should be automatic in its action, and be so regulated as to prevent any accelerating velocity in the cable.

Mr. Cyrus Field said, as the discussion had principally turned upon the proceedings of the Atlantic Telegraph Company, he begged permission to make a few remarks, in reply to some points which had been started by the various speakers.

First among the objections, were those which had been made to the form and specific gravity of the cable. And here he would remark, that he had yet to learn, that a single mile of the proposed light cable, so prominently noticed in the first Paper, and so frequently alluded to in the discussion, had ever been constructed, or practically tried in any way. It had for many years been perseveringly advocated by literary talent of the highest order, and yet, no engineer, who had had any experience in laying cables in deep water, had been found willing to lend his name to it; no electrician had come forward, who would undertake to work through it across the Atlantic; nor had any capitalist at present been discovered, whose faith was strong enough, to induce him to risk his money upon this much-neglected invention. The present Atlantic cable was a practical invention,—the result of reasoning and experiment,—and was entitled to confidence, until a cable was devised, which could be shown, either by actual results, or by the united testimony of competent authorities, to be more suitable for submersion in deep water. It had been successfully laid over by far the most difficult portion of the route between Ireland and America, and the accident of last year was in no way attributable to its form, or structure. In confirmation of this opinion, he read the following Report, which was signed by all the Engineers engaged in the undertaking last year:—

“London, August 24th, 1857.

“Gentlemen,

“We, the undersigned, who have been engaged as Engineers during the construction, shipment, and submersion of the Atlantic cable, beg to submit to you our judgment as to its fitness for the object for which it was intended, in regard to material, dimensions, and structure.

“We consider it to be so eminently suited for laying in deep water, that we cannot suggest any improvement in its form and construction, and do not recommend that any change whatever should be made in the manufacture of any future cable for crossing the Atlantic.

“The outer covering being composed of strands of iron wire, instead of single wires, as in former cables, gives such flexibility to the rope, as almost to prevent the possibility of kinks occurring in the paying out, if the cable is properly coiled; while the strength, with an equal weight of metal, is also increased.

“In case of a broken wire passing into the paying-out machine, no damage is done; while with the solid wire, it sometimes unravels, and occasions serious inconvenience.

“The conducting-wire being also formed of a strand of seven copper wires, the chance of continuity being lost in the circuit is reduced to a minimum, from the improbability of any flaw, or weakness, occurring in more than one of the seven wires at the same place.

“We are, Gentlemen,

“Yours faithfully,

“Charles T. Bright, Engineer to the
Atlantic Telegraph Company.
”S. Canning, Assistant Engineer.
”F.C. Webb, Assistant Engineer.
“Henry Woodhouse.”

“To the Directors of the
“Atlantic Telegraph Company.”

The opinions expressed in that document, and signed by every Engineer who had successfully laid a great length of cable in deep water, except Messrs. R.S. Newall & Co., coincided, in every respect, with the opinions set forth in a communication signed by every commanding officer attached to the telegraphic squadron, of which the following was a copy:—

“London, August 21st, 1857.

“Gentlemen,

“We, the undersigned, commanding officers of the several ships composing the Atlantic Telegraph Squadron, have great pleasure in expressing our opinions, with reference to the Atlantic telegraph.

“We are of opinion, drawn from our several observations and experience, that no obstacles of a nautical, or physical character exist in the way of the enterprise; that the efficiency of the form of cable adopted by the Company is in every way adapted to its mission.

“With regard to the machinery, we are of opinion, that the form of controlling-power adopted, and the mode of lubricating and adjusting the breaks admit of very great improvement.

“We are thoroughly convinced, from the soundings made by Lieut. Berryman, U.S.N., on the plateau between Newfoundland and Ireland, subsequently confirmed by Lieut. Dayman, R.N., and the investigations of Lieut. Maury, U.S. Observatory, and from our own nautical experience, that no under, or surface current exists between these points to interfere with the successful laying of your cable.

“We all agree in thinking, that no form of submarine telegraph cable could be devised, more suitable in every respect to the object intended to be accomplished; that its lightness, toughness, and flexibility adapt it in every way for the purpose of being laid between Newfoundland and Ireland, and we are unwilling to recommend its modification, or alteration in any way. We are also of opinion, that no natural obstacles exist to prevent its being successfully laid between those points, and our views as to the future prospects of your enterprise are sanguine.

“We have the honour to be, Gentlemen,

“Yours faithfully,

“J.F.B. Wainwright, Capt. H.M.S. ‘Leopard.’
“ Joshua R. Sands, Capt. U.S.F. ‘Susquehanna.’
“Wm. L. Hudson, Com. ‘Niagara.’
“C. Noddall, Com. H.M.S. ‘Agamemnon.’

“To the Board of Directors of the
Atlantic Telegraph Company.”
 
Commander Dayman, of H.M.S. ‘Cyclops,’ only hesitated to append his signature, as his name was mentioned in the Report; but he stated distinctly, that to be his only reason, and that he entirely concurred in the opinions expressed therein.

Reference had been made to the Black Sea cable, and one gentleman who advocated light cables appeared to indicate that such a plan, with some modification, would answer for the Atlantic Ocean. Now he believed he was correct in stating, that its usefulness was of less than one year’s duration. With reference to that cable, Captain A.B. Becher, R.N., had expressed himself thus—“the distance in nautical miles from Varna to Balaclava is by our chart 257; but 1 have reason to believe, that for the sake of having a bed something well under 100 fathoms from the surface, the line which was adopted was about 287 miles. The direct distance we know now runs over a depth of 870 fathoms, but when the cable was laid I believe this was not known."

But if this, or a similar form, had been considered a success in the comparatively shallow waters of the Black Sea, there had recently been an excellent opportunity of giving practical evidence of the fact, in an important link of telegraphic communication just successfully completed in the Mediterranean, where the depths approached more nearly to those of the Atlantic Ocean. It was singular too, that the contractors, Messrs. R.S. Newall and Co., who laid the Black Sea line, were also the contractors for the cable of the Mediterranean Extension Company, the entire risk of which they undertook. Now the description of cable they adopted, like practical business men, holding on to principles which were known and tried, was, in all but the external wires, precisely identical with the Atlantic cable; the only difference being, that the external covering was of eighteen solid wires, instead of eighteen strands of wire. The effect of using these solid wires was merely to diminish the flexibility of the cable, without adding to its strength; whilst its specific gravity considerably exceeded that of the Atlantic cable. Yet this cable, notwithstanding all the theoretical arguments in favour of a lighter structure, had been successfully submerged in depths exceeding 1,900 fathoms, and was now in regular work.

With reference to the strange statements that had been made, as to the height of the waves in the Atlantic Ocean, which were said to rise to 50 and 75 feet, he might state, that he had frequently crossed the Atlantic Ocean, including nineteen times solely on the business of the telegraph, and in all weathers. He had taken great pains to make inquiry, on this subject, from nautical men, and he was enabled to state, that the greatest wave of which he had obtained any authentic record, in any part of the world, measured 32 feet in all—that was 16 feet above and 16 feet below the dead water level. Commander Dayman, R.N., had recently furnished him with the following details of some experiments on the height of waves, made by him off the Cape of Good Hope:—

Determination of the Height and Velocity of the Waves off the Cape of Good Hope, 1847

Date, 1847 No. of
Obser-
vations
Speed
of
Ship
Height
of
Wave
Length
of
Wave
Time of
passing
from Spar
to Ship
Speed
of Sea
per
Hour
  REMARKS
 
April 21
23
24
25
26
May 2
3
 
..
8
6
9
..
6
7
Knots.
7.2
6.0
6.0
5.0
6.0
7.0
7.8
Feet.
22
20
20
..
..
22
17
Fathoms
55
43
50
37
33
57
35
Seconds.
10.0
8.0
10.0
7.8
7.4
10.4
8.9
Knots
27.0
24.5
24.0
22.1
22.1
26.2
22.0
 
}
}
}
}
}
}
}
 
 
Before the wind, with
a heavy following
sea
 
Sea irregular, and on
the port quarter

(Signed) Joseph Dayman,
Commander, R.N.

Commander Wilkes, U.S.N., said, on the same subject, in his ‘Antarctic Voyage’ that when off Cape Horn, “the ‘Porpoise’ was directly ahead of the ‘Sea Gull,’ and but two waves apart; the rate of sailing was about eight knots an hour, both vessels apparently very steady. In heaving the log, I found, that the ship, in drawing in the line, was, when on the top of the next wave astern, distant by line 380 feet, equal to one-sixteenth of a mile, and the schooner being on the next wave, was twice the distance, or one-eighth of a mile. The time occupied for a wave to pass from the schooner to the brig was thirteen seconds, taking the mean of many trials. This was about 26½ miles in an hour, for their apparent progressive motion. In order to get their height, I took the opportunity when the schooner was in the trough of the sea, and my eye on board the ‘Porpoise’ in the horizon, to observe when it cut the mast. This gave me 32 feet. The waves ran higher and more regular on this occasion, than I have seen them at any other time during the cruize.”

With regard to the censure that had been passed upon the Atlantic Telegraph Company, because the entire completed cable was not tested under water, before it was coiled on board ship, he stated, that the whole of the inner portion, or core of the cable, was from the first carefully tested under water, and was again subjected to a severe test, while out of water, at the manufacturers’. It was further tested a third time, when completed, and for the fourth time when on board ship. As to testing the whole cable under water at once, when completed, it was simply, in his opinion, impossible to do so, in a practically useful way. The cable must be coiled in some shape to effect this, and unless tanks were provided at an immense cost, and hydraulic pressure of enormous force was made use of, there would not be a chance of the water penetrating beyond a few feet into the coil, and the only result then obtained, would be to oxidise the external strands, and weaken the cable, without gaining any further knowledge of its electrical condition than that already possessed. He had been informed that the Mediterranean extension cable was not so tested, and in fact that the only cables which had been tested in that way, prior to submersion, were those between England and Holland, which it was reported were the worst working telegraph cables of any in existence.

It had been intimated, that the affairs of the Atlantic Telegraph Company had been conducted with secresy, and that there had been a desire to avoid giving information, as to its scientific arrangements. He could not understand upon what foundation that statement rested. From the very commencement of the lines in Newfoundland, in connexion with the Atlantic Telegraph Company, down to the present time, the policy which had been adopted by all engaged, was one of perfect openness and straightforwardness, inviting observation and benefiting on many occasions by the remarks of friends, as well as by the censures of ill-natured opponents. The sole object of all associated in the undertaking was to achieve success, and thus to demonstrate the practicability of one of the greatest engineering, electrical, and nautical problems of the age.

Mr. Paul R. Hodge suggested that experiments should be tried, with a sufficient amount of cable, in the deepest part of the ocean to be traversed, so as to ascertain what the cable was capable of sustaining, and that a buoy should be attached to it, with a spring, or springs, between the bottom of the buoy and the slack of the cable, so as to yield to the play of the waves, without adding to the strain on the cable. For this purpose, he suggested the use of india-rubber, as it was capable of being extended, or compressed, to a greater range than any other material.

Mr. Cyrus Field stated, that an experiment was to be tried in the month of April, or May, previous to starting on the actual expedition.

Mr. Hodge proceeded. The amount of slack between the buoy and the ship should be equal to the maximum rise of the waves, and india-rubber springs should be suspended at the bottom of the buoy, with a sufficient amount of slack between the holdfast of the cable on the bottom of the buoy, and the fastening on the lower end of the spring. Measures should also be taken to provide against storms, which suddenly arose in the Atlantic, even during summer. He had seen several storms in the Atlantic, in the middle of summer, although they were of short duration. He agreed with the statement, that the waves were equal to 32 feet, or 33 feet in height. On the approach of a storm he considered that the best thing would be to attach the cable to a buoy, or buoys, with one, or more, flexible springs, such as he had alluded to, and then to let it go. There should be an amount of slack between the vessel and the top of the buoys, where the cable would be fastened; the cable should then be passed to the under side of the buoys, at the upper end of the springs, and thence to the lower end of the springs, where it should be again fastened; but between these points there would be an amount of slack more than enough to compensate for the play of the springs. He felt sure that this was the only way to secure the cable in case of storms.

Mr. Longridge said, that in the commencement of the Paper which Mr. Brooks and himself had laid before the Institution, they had expressed surprise, at some of the opinions reported to have been advanced on this subject, at the meeting of the British Association, in Dublin, in 1857. He might express equal astonishment at the observations made in the commencement of this discussion, but which having been already disposed of by succeeding speakers, did not require any further attention. There was, however, one remark which he would venture to make; that was, that in his opinion, it would tend to promote the interests of the Institution, if gentlemen of high station and character would, before discussing a Paper, examine it to see what it contained. If that course had been taken in the present instance, he thought some of the opinions which had been enunciated would never have been uttered, and that some time and trouble would have been spared to all parties.

The first point he would touch upon, was that of the law of a body sinking in a resisting medium. The Authors had given the equations of motion for a sphere and a cylinder, like a cable, laid horizontally. They had shown, that in a short time after letting the body drop, it acquired nearly its terminal velocity. Taking a sphere of the specific gravity and diameter of the Atlantic cable, in six-tenths of a second it acquired within one-three thousandth part of its ultimate velocity, the depth descended to acquire that velocity being about 16 inches. Therefore, for all practical purposes, the line passing through a series of balls, dropped at equal intervals, from a ship moving uniformly forward, would be a straight line.

The next question was the difference between such a series of balls and a cable. It was obvious, that with those balls the resistance of the water was in a vertical direction, but if the body was continuous and was placed in a sloping direction, it was evident another force came into play. This was represented in Fig. 12. A line of cable A E was supposed to be lying in the water. If any particle of it at A was considered, it would not move vertically, but in some unknown direction, such as A B. To find this direction, there were the weight of the body acting vertically, and the resistance of the water in the direction B A. The latter must be resolved into A C, at right angles, and B C parallel, to A E, the latter acting by friction. They had given the equations of motion under those conditions, from which the angle between A B and the vertical, which gave the direction of motion, and also the velocity, could be ascertained. Now that problem met the case which would occur when a cable broke in the act of paying out. The cable would be in a straight line from the ship. Every portion would be subject to the same forces as the particle at A, and would run back in a direction parallel to A B, so that the whole length A C (Fig. 2), would deposit itself in wavy folds, between C and D. The same would occur, supposing the cable was left to run out without tension. The point A would descend in the line A D, and the whole distance A C, which ought to be laid from C B, would be laid between C and D. They had given the formula by which the angle might be calculated, as also the waste of cable. The Tables, page 225, contained calculations for two cables, one being the Atlantic cable, and the other a cable of a specific gravity of 1.5. The first line contained the velocity of the paying-out vessel; the second the inclination of the cable to the horizon; the third the angle of motion of each particle when payed out without tension; the fourth the velocity of the cable in feet per second; and the last the waste of cable. When running out vertically, the velocity of the Atlantic cable would be 24.2 feet per second, and the waste, of course, 100 per cent. The next point necessary to be ascertained, before the tension could be calculated, was the form of the curve assumed by the sinking cable. That was a problem of some difficulty. Supposing a cable to lie in the direction A B (Fig. 3), in order that A might not run back, some force to the right hand would be required to bring it to the point C, and it might be supposed that that force would lead to a curve, represented by A’ a. It might also be, that a cable could not be laid straight, and free from folds, without tension at the bottom. If so, that tension must give rise to another curve, B’ b, at the bottom. The only way to ascertain the exact form of the cable was to consider the forces acting upon it. Now, it was shown in the Appendix, that in order that any cable might curve from the position A C (Fig. 13) and lie straight along the bottom, the length of the line A C must be equal to the length of the line B K (Fig. 3), and the direction of motion of any particle C must bisect the angle made by the cable and the horizon at C. Every point must move in a direction bisecting the angle made by the horizon and the tangent of the curve at that point. Taking any portion of the curve A C (Fig. 13), the forces would be—first, the weight of this portion of the cable acting vertically; secondly, the horizontal tension (if any) at the bottom; and thirdly, the tension at C, in the direction of the tangent of the curve. These forces would give the common catenary. But there would also be the resistance of the water, and that not acting in any one direction, but at every point, in a direction bisecting a varying and unknown angle. The resistance must be resolved into two forces, one at right angles and the other parallel, to the curve, and acting by friction. These were the forces in the fundamental equations, which came out rather complicated; but they had been integrated by Mr. Brooks, whose absence Mr. Longridge very much regretted, for to him was due a very great share of whatever merit the Paper might possess. From these equations it appeared, that the curve was very nearly a straight line, and was always concave to the horizon. It had an asymptote of which the position might be found, although it was of no practical value. By making the tension at the bottom equal to zero, the equation became the equation to a straight line; consequently, in the case of a cable dropped from the ship sailing uniformly, it would descend in a straight line to the bottom, so long as it was laid without tension at the bottom. In the case of the Atlantic cable in a depth of 2,000 fathoms, and the paying-out vessel moving 6 feet per second, or 4 miles per hour, the angle with the horizon would be about 28°. This did not exactly agree with what had been stated as the observed angle, but the difference could be readily explained. It was not very easy, under the circumstances, to observe an angle of that description, and moreover, in practice, the cable was not payed out from the surface of the water, but from the ship’s stern, a point considerably above it; consequently, the cable would take a catenarian form, till it came to the point of touching the water, and therefore the angle at the water would be less than that given by the equation. In addition to this, the resistance of the water did not act above the surface, and that would also give a small inflection, which would quite account for the difference between the observed angle, and that found by calculation.

The next question was as to tension. The equation for tension could not be exactly integrated, but the supposition upon which the integration was affected was accurately true, when the tension at the bottom was equal to zero. The equation showed, that so long as the cable was free to run out of the vessel, no greater tension could come upon it, than what would be expressed by the weight in water of a length of the cable equal to the depth of the water in which it was being payed out. Whatever the angle might be it mattered not. The tension would be slightly diminished by the friction of the cable against the water.

The next point was to find the diminution of the tension arising from letting the cable run out quicker, than the rate at which the ship was moving. That problem was also solved, and it was found, that unless the ship was moving at a high velocity, very little tension was saved by letting the cable rim out quicker than the ship was going. The Authors had calculated the amount of such reduction of tension, for the Atlantic cable and for a light cable of the specific gravity of 1.5, and had given the results in two Tables, page 255, as well as exhibited them in the form of curves (Figs. 4 and 5). Taking the velocity of the paying-out vessel at 6 feet per second, if the rate of the cable to that of the ship was as 1 to 1; that was to say, if the cable and the ship were moving at the same rate, the tension would be about 3,842 lbs. for the Atlantic cable in a depth of 2,000 fathoms. Supposing, then, the cable to run out one-fourth quicker, or as 5 to 4, the diminution of the tension would be very small—not more than about 6libs. Even if the cable was to run out at double the speed of the vessel, or at 8 miles of cable to 4 miles of the ship’s rate, the tension would be still 3,227 lbs., thus giving only a decrease of about 615 lbs., and wasting half the cable. It would be different when the velocity of the ship became high. If the speed of the ship was 10 miles an hour, by increasing the speed of the cable one-half above that of the ship the tension would be reduced by about 2,000 lbs.; but such high velocities were wholly impracticable. Consequently, it was evident, that letting the cable run out slack, was not the way in which the question of tension should be met. Whilst upon the question of tension, Mr. Longridge might refer to the subject of floats, or resistors. It had been said, that the Authors had suggested the use of floats. They did not suggest them, but said they had been suggested. Floats, or resisters, in the form of square, or round boards attached to the cable by a cord, had been strongly advocated during the course of the discussion. There seemed to be some misapprehension on the subject; because if they examined the forces that acted upon a cable, it would be found that the only force acting upon the resisters was the resistance of the water. That was in a direction bisecting the angle between the cable and the horizon, and that force must be resolved into two—one parallel to the cable and the other at right angles to it. The parallel one was that which acted by friction, and that had been taken into account in the equations. It was only that part of the resistance of the water that could affect the tension. The other portion of the float’s action would have the effect of altering the angle at which the rope would sink, but it had been shown, that the tension was quite independent of the angle, except in so far as friction was concerned. The only way in which he thought such a plan could be usefully applied, was to attach the floats along the cable at intervals—no matter what their form—they would be better if put across at right angles—but by attaching them at intervals, the specific gravity of the cable might be diminished, by their buoyancy, and the resistance by friction increased.

He now came to an important point in the Paper which had been somewhat overlooked, although some reference had been made to the subject it treated of. That was the action of currents. That question had been carefully examined in the Appendix, and he would not therefore burden them, by entering into the demonstrations, but would simply state the results arrived at. First, after the complete entrance of the cable into the current, there was no waste of cable due to its action. Secondly, the action of the current caused no extra tension upon the cable. That was a curious result, but he was satisfied of its correctness. The Authors had first considered the action of a stream of water upon a flexible line stretched across it, and kept from running out at both ends. That was well known to be powerful. The mathematical investigation showed the curve to be a catenary, differing however from the common catenary in this, that the tension was constant throughout. In the case of a stream of water, where the force at every point was resolvable into a force at right angles to the curve, and another parallel, the tension was constant. In laying out a cable with slack, there could be no tension at the bottom; and since any tension must follow the above law, the tension at the top, arising from the action of the current, must be zero also.

The next question was that of stopping the paying out, or attempting to haul the cable in. Supposing A B (Fig. 15) to represent the normal position of the cable, then, if the paying out was stopped suddenly, the cable would rise from the bottom, and assume a catenarian form. The only difficulty was to find where the vertex of the catenary would be; but that was attainable, and had been worked out. It depended upon the original angle of the line A B, which varied at different speeds of the paying-out vessel. The Tables, pages 237 and 238, showed the resulting strains. At a speed of 8 feet per second, in a depth of 2,000 fathoms, the consequence of stopping would be to bring a strain of 7 tons upon the Atlantic cable. If the velocity was 10 feet per second, the strain would be at least 16 tons. Practically it would be more, as it was impossible to stop the paying-out vessel at the same moment as the cable stopped. The ship moving whilst the catenary was being formed, the strain would be increased. The same formula gave the strains in any attempt to pick up a cable. In the case of the Atlantic cable at 2,000 fathoms, and with an angle of 58° with the horizon, there would be a strain of 4¼ tons upon the cable—its ultimate strength—consequently if the angle became less than 58°, the cable must break. This showed the difficulty and danger of attempting to haul up a cable in deep water.

The next point was the pitching of the vessel. It had been said, that the Authors of the Paper did not appear to attach much importance to that point; but it was omitted to be stated, that they had put in a proviso—that the paying-out apparatus should be light, and the breaks work freely. If those conditions were not attended to, the pitching of the vessel became of serious consequence. Since the Paper had been before the Institution, Mr. Brooks had worked out the problem, to ascertain what extraordinary strain would be brought upon it, in consequence of the inertia of the rotatory part of the machinery. If there was no pitching, but simply a uniform motion over the wave, the accelerating force would be a minimum, when the vessel was at the top of the wave, and a maximum when she was at the lowest point in the trough of the sea,—just as she was beginning to fall.

A great deal had been said about the height of the waves. Dr. Scoresby had stated, that the highest waves, measured by him in the Atlantic, were 36 feet high from the trough to the crest, the length of the wave was about 790 feet, and the time in passing the ship was 16½ seconds. He had assumed these data in estimating the accelerating force, and found that, under the above circumstances, for each cwt. of weight in the rotatory machinery, the variation in the strain would be about 5¼ lbs. Consequently, if the rotatory machinery weighed 90 cwt., that would give a variation of strain of 475 lbs. He had also calculated the case of a sudden pitch of the vessel. Supposing the vessel to pitch 6 feet high in five seconds, he found, for every cwt. of the rotatory machinery, a variation of the strain of 9½ lbs.; so that when the rotatory machinery weighed 90 cwt., there would be nearly 8 cwt. of variation of strain.

The next point was the paying-out apparatus After what he had said, they would be prepared to admit the great importance of having that apparatus as light as it could be made. This was one of the difficulties in dealing with a heavy cable. With the Atlantic cable, the strain in a depth of 2,000 fathoms would be 35 cwt., and they must have strong paying-out apparatus, and powerful breaks, and he saw very great danger in the use of those, under the circumstances in which they must be employed. They required the most vigilant attention to prevent accident. It was not only the extraordinary strain, but where one break had such a mass of matter to restrain, it was liable to get jammed. The screw might be pinched too tight, and a momentary jam take place; and an enormous strain would be brought on the cable, arising both from the inertia and the catenarian strain, which would be sufficient to break any cable that was ever made. In the apparatus on board H.M.S. ‘Agamemnon,’ (Plate 8, Fig. 14) he thought there was danger in the sheaves being geared together. They must all revolve at the same time. In the Engineer’s Report it was stated, that an accumulation of tar, &c., in the grooves had thrown the cable off the sheaves, and any unequal accumulation must either cause the slipping of the cable round the sheaves, or else the apparatus would stop. He thought that by the adoption of gearing an additional and unnecessary danger was incurred. With a break attached to each sheave, there would be less chance of the break seizing, and other serious difficulties would be avoided. With regard to the question of self-acting breaks, he remarked that self-acting machinery was excellent in certain cases, but he would not like to depend on it, in an operation like the laying of the Atlantic telegraph. He was satisfied where such great interests were at stake, and the slightest accident would involve the loss of the whole property, it was better to obtain the services of efficient men, who should pay personal and unremitting attention to the breaks, and not to depend upon self-acting machinery.

Whilst on the subject of paying-out apparatus, he might remark, that he had submitted a sketch of the plan proposed during the discussion, embodying the “picking-up chains,” as they were called, to Mr. Brooks, who had sent him the results of some careful calculations on the subject, which he would read to the meeting. Mr. Brooks said:—

“If the chains are hung in a short thick bundle, their vertical motion will be but small, the moveable pulley nearly at rest, and consequently the apparatus nearly useless. If the chains are in a thin bundle, so that a considerable rise will add but slightly to their weight, or if they are completely suspended so as to be a constant weight, then Formula 21 (see MS. Appendix) will show the effect of its inertia.

“It is also plain, that when the ship begins to rise, and more cable is required, the chains will have a tendency to remain at rest in absolute space, and consequently the effect of the apparatus is a tendency to double the waste of the cable, while the vessel is rising. In reality this tendency of the chains to remain at rest, in space, as the vessel rises, is sufficient reason for not using the apparatus. The chains have almost precisely the same effect as a heavy weight hanging on the cable, and attached to the vessel’s stern.

“If springs were used instead of chains, the inertia would be avoided, but even then the pressure would be propagated to the break, just as though nothing were used; but the springs would give out a little of the cable, as the vessel rose, owing to the increase of tension due to the inertia of the wheels.

“A constant suspended weight is beneficial, but only slightly so.”

After some further remarks, Mr. Brooks concluded thus:—“Upon the whole then, the theoretical disadvantages are greater than the advantages; if to these are added the practical disadvantages, viz., the irregular oscillation of the chains, the danger of the cable slipping off the moveable pulley, and the extra pulleys required, it is plain that the apparatus had better not be used.” Mr. Longridge had looked over Mr. Brooks’ investigations, and quite concurred in his conclusions, as expressed in the extracts just read.

The next point was that of catching the ends in case of fracture. There were certain difficulties, chiefly arising from the pitching of the vessels. In calm water it could be done without difficulty; and he thought means might be devised, by which it might be done in rough water, and that it was a subject well worthy of experiment.

He now came to the question of the construction of the cable. In the Paper, allusion was made to the stretching of the outer covering. When submerged at great depths, the inner core became compressed, and the strands of the outer covering tended to alter the form of their spirals. Whether that was of sufficient importance to become a practical point, he did not pretend to say. But he thought there was one point not sufficiently considered, in the tension of the heavy cable. It was stated that gutta-percha had great extensibility. That might be true, and yet, though it might not be torn asunder when stretched in those great depths, it was possible that its solidity might be affected, and the water might penetrate to the core, and decrease the insulation. Experiments had been made by Mr. Macintosh, to prove the extent to which water could be forced through gutta-percha. He first put upon the inner core some chemical substance, which upon contact with water changed colour. This was then covered with gutta-percha, and was subjected to hydraulic pressure, and the effects were seen in the discolouring of the chemical substance. He believed that in every case, the water penetrated to the core, under a pressure of about 2 tons per square inch. If that was the case, something of the same kind might occur at great depths, and it was desirable to ascertain the extent to which it might arise, and its effect in reducing the insulation. The investigation of this subject had induced the conclusion, that cables of light specific gravity could be much more easily and safely laid than heavy ones. He had never intended, nor did the Paper express the idea, that the construction of Mr. Allan’s cable was the best for all circumstances; nor had he ever expressed the opinion that light cables should be used under every circumstance. Cables must be adapted to the localities and the conditions under which they were to be laid; but his attention, and that of Mr. Brooks, had been directed, chiefly, to the circumstances of cables laid in water so deep, that when once deposited, they would not be subject to currents, or to agitation, arising from the waves on the surface. It was considered, in the case of the Atlantic cable, that if once laid at that great depth, it would be perfectly secure from any such disturbing causes, and that was also evidently the opinion of the promoters of that project, as evidenced by the following extract, from the publication issued by the Directors at the beginning of the last year:[7] “The reliance of the projectors and executors of the work is not however placed upon this natural process of cementation, neither is their trust in the durability of the wire. The fact is simply that the iron investment is only intended to serve the end of protecting the coated core from mechanical violence, and to confer upon the cable a convenient amount of proportionate weight during the process of submergence: it is designed for these purposes, and for nothing further. When the cable is once fairly laid in the still soft depths of the Atlantic, and on its diatom-strewn shelf, the rust may eat up the iron external coat, as soon as it pleases. Then the conducting strand will be as safe as if it were bound in a trench, and the gutta-percha covering will be as permanent and durable, in the absence of a high temperature and air, as the gum in the mummy, which lasts unchanged for thousands of years.” From this it was plain, that the outer iron casing was not intended as a permanent protection, and that it was very far from fulfilling the anticipations of its projectors, as regarded the effect of its weight during submergence. What he said was, that if they got a light cable to the bottom, it would answer its purpose equally well, without any outer metallic covering; and with a light cable, they had a much greater chance of getting it safely to the bottom. But he certainly should not advocate a light unprotected cable from Dover to Calais, from the causes which were in operation there, and which did not exist in deep waters. He did not wish to attach too much importance to the document from which he had quoted, as he thought it had been drawn up somewhat incautiously, and indulged rather too much in prophecy; but it was evidently not expected, when the above extract was written, that the weight of the Atlantic cable would have the effect in deep water which had been observed. It would appear, from the remarks which had been made, that the Directors were not yet divested of the opinion, that this was the best form of cable that could be adopted for the purpose. He was very sorry they held to that opinion, although it was supported by gentlemen of high standing; but Mr. Longridge still believed it to be erroneous. He thought some of the remarks upon Mr. Allan’s cable were not justified by the facts of the case. In mentioning that cable, he did so because it was the only one that had come under his notice, when the Paper was written, which fulfilled the conditions he considered to be essential. Since then he had seen other cables very much upon the same principle. It was assumed, that Mr. Allan insisted upon putting an iron core into his cable. He believed this was not the case. Iron was suggested because it was cheaper, and because a larger channel for the electricity to go through could be employed, at a moderate cost. It had been stated that the smaller the channel, the better the passage of electricity through it; but that theory had been satisfactorily disposed of. If a small conductor was the best, it could still be used, with any form of cable, and the smaller the channel the less the specific gravity of the cable, and the easier it would be laid.

[7] Vide “The Atlantic Telegraph, &c.” p. 46.

He quite concurred in the remark, that experiments were still wanted for an operation of this magnitude. For his own part, he would not like to undertake so heavy a responsibility, without further experiments, on some of the more important points, and he hoped the Paper which he, in conjunction with Mr. Brooks, had submitted would serve to point out the direction in which those experiments should be made, and as such he trusted it would prove of some value to the Members of the Institution.

Mr. Webb said, before making any observations on the mechanical questions which had been discussed, he wished to reply to two general assertions, having reference to the comparative difficulty of paying out and of repairing cables, and to the number of submarine cables stated to have failed during the process of laying, or shortly afterwards. It had been said, that it was perfectly easy to lift cables in shallow seas, and then to mend them. He thought such an unqualified remark would bear considerable modification. It was true, that the single operation which had been instanced, that of the Calais cable, might have been excessively easy. It consisted, simply, in joining a short piece of cable to one that was not severed, and at a distance of only half a mile, or a mile, from the beach. Moreover the position of the faulty part was suspected (if not known), from the occurrence that took place during the paying out. This was very different to repairing a cable broken at 10 or 15 miles, and in some cases 60 miles, from land, and where the distance of the fault from the shore could only be ascertained approximately, even by the electrical tests. The means for overcoming the various difficulties, by under-running, picking-up, grappling, buoying, splicing, &c., all of which, it must be remembered, had to be performed on the open sea, and out of sight and shelter of land, would not be regarded as so obviously easy even now, and still less when they were performed for the first time, in the instances described in his Paper. As far as his experience went, he would only further say, that the operation of paying out was mere pastime, as compared with that of repairing cables.

The second point to which he would allude was the statement that one-half the submarine cables hitherto essayed had failed during the process of submerging, or immediately afterwards. Having been engaged in laying 1,500 miles out of the 2,500 miles successfully submerged, he was in a position to state, that out of the forty-three submarine cables hitherto attempted to be laid, six only had failed during the process of laying, four had failed subsequently, and one of the Hague cables was at present under repair; leaving thirty-two in perfect working order. Of the ten total failures, three were strictly light cables, with no outer wires, being the only uncovered cables tried; and the two failures of heavy cables after submersion, arose from their not being sufficiently heavy. Several cables, including the Hague, the Holyhead, the Dover and Calais, and the Dover and Ostend, had, it was true, been broken by anchors, in consequence of the absence of sufficient external protection, but they had been immediately repaired, and were now in regular work. Of the six failures in submerging, two occurred with the Mediterranean cables, in the years 1855 and 1856, when 256 miles were lost, of the value of about £70,000; a third with the Newfoundland cable; a fourth with a light cable from Portpatrick to Donaghadee; a fifth with a heavy cable on the same route; and lastly, the Atlantic cable. Of these, 42 miles of the Newfoundland, and the whole of the heavy Portpatrick and Donaghadee cables, had been recovered, the latter by Mr. R.S. Newall in the steamer ‘Monarch;’ and during the present summer the raising of the Mediterranean cables was to be attempted, although the time which had elapsed since they were submerged would add considerably to the difficulty. It should be added that, with the exception of the Atlantic and the Portpatrick cables, no British civil engineer was employed, in a responsible position, in any of the undertakings. With reference to the Atlantic cable, it should be recollected, that it was a great step in advance of all preceding enterprises. He was quite willing to bear his share of the blame of the failure of that undertaking, if due allowance was made for the fact, that he was not on board the ‘Niagara’ when the cable was being payed out, and that he did not design, or approve of the break.

With regard to the mechanical and electrical questions involved in the consideration of this subject, he regretted that the Paper by Messrs. Longridge and Brooks had not been discussed in proportion to its merits. There was much valuable information in that Paper, and if mathematical investigation could ever be relied on in practice, it ought to be of considerable use in guiding those who had the superintendence of the paying out of cables in deep water. He was not able to follow all the algebraical investigations, but as far as he could judge, geometrically and practically, he felt convinced of their accuracy. It had been shown, that the loss, or waste, was not caused by the tendency of the cable to fall perpendicularly, in which case the loss would only be equal to the difference between the length of the cable leading to the ground and the distance run from the point where the cable touched the ground, nor yet by a direct retrograde movement in the direction of the cable leading to the bottom, but by a tendency to move in a direction between these two. Also, that to lay a cable straight, a tension must be exerted equal to what had been termed the minimum tension, or the weight of a piece of cable equal in length to the depth of the sea, where the cable was being payed out. That result was certainly contrary to what had generally been supposed, and seemed difficult, at first, to believe, but there was no practical proof that it was incorrect. According to this law, the required tension for the Atlantic cable, for a depth of 2,000 fathoms, would be 34 cwt.; but as this strain had never been continuously applied to the breaks for any length of time, that experiment did not disprove the rule. It had also been demonstrated, that where the strain on the break was decreased, even a little below the proper tension, when the vessel was proceeding slowly, the per centage of loss, or waste, would be considerable, and vice versa; that under those circumstances, a very little easing of the tension was obtained by allowing great loss; and that as the velocity of the ship was increased, the strain on the break might be much decreased, with a moderate loss, although at whatever velocity the vessel was proceeding, a strain of 34 cwt. would be required in the case of the Atlantic cable, to lay it absolutely straight. These laws, with the exception of the latter, the possibility of laying a cable absolutely tight on the ground, had been, more or less, suspected in practice, but they had never been previously worked out and recorded with such clearness and precision. It appeared from the Tables, page 255, that with a velocity of the ship of 2 feet per second, a decrease from the proper tension of about 50 lbs. would cause a loss of 50 per cent.; whilst if the velocity was increased to 8 feet per second, a decrease of about 300 lbs. would only cause the same loss. Again, if the ship was proceeding at the rate of 4 feet per second, a decrease in tension of 300 lbs. would cause a loss of cent. per cent.; whilst at 8 feet per second it would require a decrease of 1,300 lbs. to cause the same loss. Finally, that Paper showed, that a decrease in the specific gravity of the cable was the real remedy for great tension. No doubt, if the absolute weight in water of a cable could be decreased without diminishing its strength, or in a greater proportion than its strength was decreased, all other necessary conditions being complied with, the margin of strength, or the difference between the breaking strain, and the strain required to support a given length of the rope in water, would be augmented. A mere reduction in the specific gravity of the cable, even without a reduction in strength, did not, however, necessarily produce an increase in the margin of strength, or the modulus of tension. For, it was evident, that the specific gravity of the rope might be reduced, by adding a material which was heavier than water, although lighter than the rest of the rope; as, for instance, by adding a hemp serving to an iron rope, in which case the specific gravity would be decreased, whilst the absolute weight of the rope in water would be increased, the absolute strength also remaining the same. Thus, the modulus of tension would actually be reduced by a decrease in the specific gravity of the rope. No increase in the modulus of tension could, in fact, be obtained by a reduction of specific gravity caused by adding a material which, without adding strength, was as heavy, or heavier, than water. Still a reduction in the specific gravity, if attained by an increase of some part of the rope which was of the same specific gravity as water, although it could not increase the modulus of tension, would benefit the cable whilst paying out, as its terminal velocity would be reduced, without the modulus of tension being decreased. This had, apparently, been aimed at in the Atlantic cable, by reducing the section of the outer wires, and enlarging the gutta-percha, so as to make the gutta-percha and the copper conductor together of about the same specific gravity as water. He was of opinion, that no plan had hitherto been proposed, by which this system could be materially extended, that did not possess other serious disadvantages. Possibly, the outer covering of the Atlantic cable might be still further reduced in thickness, and thus the specific gravity be diminished, without any decrease in the modulus of tension, although the absolute strength of the rope might be less. But if the layer of wires around the core was smaller, they would become more unstable, and afford less lateral protection, by which the perfection of the insulation would be exposed to great risk, He thought the lowest limit had been reached, at which they would fulfil, properly, the beautiful mechanical properties of a wire covering, which he should shortly endeavour to explain. By the use of steel, however, the absolute strength of the cable might be augmented, without any increase in its specific gravity. He considered that no system had yet been devised superior to metallic sheathing, for affording lateral protection during the act of submerging, and for giving permanent strength in shallow seas; and that it would be imprudent to do away with that sheathing, unless some very good substitute could be found, even if the removal of all the longitudinal strains, not caused by weight, could be guaranteed; for the insulation would be exposed to the greatest risks. He thought that the plan which had been suggested, for placing the iron external wires on the inside, around the conductor, and within the gutta-percha, would increase the risk in a mechanical point of view, setting aside its electrical disadvantages, which had been so well pointed out by previous speakers. With regard to the light cable advocated by Mr. Longridge, having a specific gravity of 1.5, and the whole of the metallic portion placed in the centre, surrounded by the insulating medium, he thought that the strain required to lay such a cable—8 cwt.—would, in summer temperature, force the metallic core through the gutta-percha, whenever it made an angle over a casting, or roller; and even the weight of the cable, in the hold of a ship, would be exceedingly liable to flatten the rope and to expose the iron wire; and the slightest friction against a casting, or anything not revolving, under a tension of 8 cwt., would scrape the yarn and gutta-percha off for a mile, or two miles at a time. A better system would be to have a simple copper conducting-wire, surrounded with gutta-percha, and covered with hemp. Indeed the success-of the Varna cable went to show, that the danger to that material, when submerged, was not so great as had been imagined.

With regard to the objections to the spiral ‘lay’ of the outer wires, he explained that for any elongation to take place, the spiral must be straightened. This could only arise, taking a single wire, in two ways; either by its retaining its regular number of convolutions, and by its being flattened in that position, or by its unwinding itself from the rope. The first of these effects could only take place by each wire approaching to the centre of the rope, and thus occupying a smaller circle; but as the wires were already in contact, they could only assume this new position by crushing one another laterally. They in fact formed an arch, which, if hollow, would be in unstable equilibrium, but the core, without receiving any great pressure, acted as a kind of centering, keeping the whole system in shape, by preventing any one wire getting out of position, by being forced inwards; whilst the tension on each wire prevented it from getting out of position, by being forced outwards. Thus, as long as the wires were kept in their circular position, no elongation could take place, whilst no crush could come on the core, if the rope was well made, and the wires were in contact, as they ought to be. When one wire was much slacker than the rest, or the core was too small, the equilibrium might be disturbed, and one wire be forced out, in which case elongation would take place. In a rope composed of strands, where they did not form such a stable arch as the solid wires, the elongation was more likely to take place, and it was evident, the smaller the wires, and the greater their number, the less stable would the system become. The second effect could only happen when one end was free to revolve, whilst the rope was under tension; for if both ends were held tight, it was evident that one part could not become unwound without winding up the other. But even if the end was free to revolve, which was not the case in submarine cables, there was yet another force tending to resist unwinding. The wires being only bent into their places by the untwisted wire-rope process, and not twisted,—to unwind them torsion must take place in each wire, and the joint resistance of all of them to torsion, acting at a mechanical advantage, from the length of the spiral over the longitudinal strain, would almost entirely prevent this action. It would, therefore, be seen, that in a well-made, solid wire telegraph cable, only a trifling elongation, say about three-quarters per cent., could possibly take place; this would give a wholesome spring to the rope, without endangering, in the slightest degree, the conductor, or the gutta-percha, which would stretch 10 per cent. without injury. The outer covering consequently prevented any injurious strain from coming on the gutta-percha and copper wire, unless it was itself broken.

With regard to the mechanical contrivances for paying out a cable, it had been stated, that many cables had been ruptured by the breaks. He had previously shown, that there were only six cases of loss during submergence, and of these only one (the Atlantic) had, to his knowledge, been broken in paying out, and that was with a break totally different to any before used. On every heavy cable yet laid successfully, except the Dover and Calais, one, or two breaks, nearly similar in construction to the one represented in Plate 8, Fig. 13, and entirely so in principle, had been employed. That figure gave an exact representation of the break which was fitted on board the ‘Monarch,’ and it had been employed in paying out many hundred miles of cable; altogether about 2,500 miles of cable had been successfully laid with that kind of break, without any fracture. No doubt, if the varying strains brought on the cable by the movement of the ship could be distributed more uniformly, it would be a great advantage. But he did not think this could be satisfactorily accomplished by means of weights, as the vis inertia of the pulleys and the weights would interfere with that arrangement. The apparatus described, in the course of the discussion, would necessitate a weight of chains equal to double the breaking strain of the cable. In the case of the Atlantic cable this would amount to upwards of 8 tons. If this weight was drawn up to nearly the full height, and the strain was then greatly diminished by the break being taken off, the weight would suddenly descend, and the momentum caused by its accelerating velocity would be liable to bring a greater strain on the cable, which would at that moment have the break again applied, by the long lever resuming its original position. In the case of fracture to the cable, the whole weight would be precipitated to the bottom of the vessel. If it was applied to a cable of the strength of the Mediterranean cable, a weight of 80 tons would be required, and the fall of such a weight of metal would endanger the vessel. In his opinion, if ever a useful compensating apparatus should be introduced, it would be by means of springs, or buffers, or cylinders of steam acted on by compression, and not by weights, or springs of tension. The fracture of tension springs would be attended with great danger, whilst if a similar accident occurred with a compression apparatus, it would only leave the cable without a compensating arrangement.

The system of paying out cables from the centre of oscillation had been frequently suggested; and although it presented another advantage, not previously mentioned in the discussion, that of facilitating the steering of the vessel, it had not been deemed expedient to adopt it. In the first place, the cable would form a sharper angle round the bottom of the vessel than it did over the stern, so that the crush on the cable, tending to flatten it, and the friction, would be greatly increased; and as it would be almost impracticable to keep it in a sheave fixed at the bottom, this would take place whilst running on a dead casting, or a wrought-iron plate. Should there be a broken wire, which with the most carefully-devised ‘lead’ of the cable, with rollers and sheaves, would sometimes occur, there would be no means of freeing it, as when paying out over the stern; thus the breaking of a single outside wire might entail the loss of the cable, whilst the chance of their occurring would be greatly increased. Hut perhaps the most serious disadvantage was, that there would be no means of ascertaining what the sailors termed the ‘grow’ of the cable, or the horizontal and vertical directions it assumed after leaving the ship; by which, in the one case, valuable information was obtained, as to whether the vessel, or the cable, was being acted upon by currents, or by tides; and in the other, a rough and ready means of approximating to the strain on the cable. The cutting for buoying, and the change from one hold to another, and so altering the ‘lead’ on to the break, would also be attended with certain practical inconvenience.

In reference to the statement, that the best angle at which to pay out a cable was about 45°, he remarked, that this would necessitate the speed of the vessel being limited to about half-a-mile per hour; whilst it could not possibly ease the strain on the cable, but would rather increase it. He thought, that the table of catenarian strains, mentioned at the commencement of the discussion, could not possibly be applied to the case of submarine cables, in the way that had been attempted. He had positive proof to the contrary, as he had measured the angle of the Atlantic cable, when paying out in a depth of 2,000 fathoms, from the ‘Agamemnon,’ (and not, as had been concluded, on the Hague cable,) and had found it to be 10°. The strain, according to the Table, would, therefore, be sixty-five times the minimum tension, or about 100 tons, or nearly twenty-five times the breaking strain of the cable. In answer to another statement, that the loss, or ‘waste,’ could not possibly exceed 1 per cent., instead of from 30 to 50 per cent., or be arranged in wavy folds at the bottom of the sea, as had been described, Mr. Webb observed that, in the case of the Atlantic cable, 350 miles were payed out, although the distance traversed was only 250 miles, and if this supposition was not correct, it was difficult to imagine how the extra length was disposed of.

He thought it would be generally admitted, that the larger the sectional area of the conductor, with any given material, the less would be the resistance to the electric current; and that if that section was a circle, the retardation would also be decreased; for the retardation might be briefly stated to be increased in proportion to the surface under induction, and be decreased directly as the resistance was diminished. When, therefore, the resistance was decreased in a greater ratio than the surface under induction was increased, the retardation of the wave towards its maximum would be lessened. When the diameter of a circular conductor was increased, the surface under induction was also increased in direct proportion; but the conduction was increased as the square of the diameter. The opinion that no benefit was obtained by an increase of section had apparently been based on a fallacious experiment. Some trials with a conductor composed of three wires showed “that the retardation observed in a single wire was considerably augmented, in fact nearly doubled, by the addition of two other similar and parallel wires, using the three as one treble wire, possessing threefold lateral dimensions.”[8] But it was to be remarked, in reference to this experiment, that although the conducting area was trebled, yet as there were three separate circles, the surface under induction was also trebled, and hence no decrease in retardation had resulted.

[8] Vide Letter from W.W. in “The Engineer,” January 23, 1857.

In paying out the Atlantic cable, the reception of an electric current seemed to have been used as a test for insulation. This, he contended, was a dangerous system. As long as no resistance was added to the circuit, a partial fault might pass, without producing any other inconvenience than an additional adjustment of the relay, which might be set down as the effect of the ship’s movement. But when resistance was added, after a fault had passed, a much greater decrease in the strength of the received current, than that due to the resistance of the additional wire, if there was no fault, would occur. The current on arriving at the fault from the shore would divide itself in the inverse ratio of the resistance of the fault to that of the wire between the ship and the fault. As long as these resistances were equal, the loss at the vessel would he 50 per cent.; but if the resistance from the fault to the ship was increased to ten times the resistance of the fault, then ten-elevenths of the current would pass to earth at the fault, whilst only one-eleventh would reach the ship. After the addition of one length in the ‘Niagara,’ a stoppage in the communication occurred for several hours. It seemed, therefore, extremely probable, that a greater decrease in the received current occurred, than that due to the mere resistance of the additional length, which was in a perfect state, or else there would not have been so much difficulty in adjusting the batteries and relays, to work through an increase of resistance, seeing that that addition must have been expected. It was on these facts, and on Professor Morse’s statement, that he had based the observation, that a fault, although perhaps a partial one, had passed unnoticed. It had not been stated whether the cable was tested for insulation, with strong battery-power, the last thing before it parted, with a perfectly favourable result; and this alone could give positive evidence of the perfection of the insulation.

In conclusion, he expressed the hope, that the Atlantic undertaking might be carried to a successful issue during the present year; for he feared, that another failure might destroy public confidence in such enterprises, and so retard the much-desired extension of telegraphic communications.

Mr. Locke, M.P., President, said, the Institution was much indebted to the Authors of the two Papers, which had been discussed at such great length, and had excited so much interest and attention, that he thought it augured favourably for the success of the next attempt to lay the Atlantic cable. Both Papers had evidently been prepared with considerable care, and with great labour; and the two together might be said almost to exhaust the subject of which they treated. In the first Paper the opinions advanced were given with that clearness and precision always characteristic of sound mathematical investigation; whilst the second Paper was descriptive of actual operations, performed by the Author, and of the difficulties that were constantly encountered, in carrying out such undertakings. In fact, the one might be looked upon as a theoretical exposition of the subject,—the other as a more practical account of the modus operandi, the machinery, and the means by which submarine cables had been successfully laid. There had also been exhibited a series of specimens of different kinds of cables, contributed by Messrs. R.S. Newall and Co., and Messrs. Glass, Elliot, and Co., to whom the Institution was likewise indebted. It was not to be expected, in discussing a subject of so comparatively novel a character, and where many of the remarks had reference to an enterprise of great magnitude, that there would be perfect unanimity of opinion; but this he must say, considering the number of speakers, and the abstruse nature of the subject, there was as much agreement as could reasonably have been expected. This result showed, that even where private interests and feelings were concerned, conflicting opinions could still be urged with good temper and good taste. He thought he should not be doing justice to Mr. Longridge, if he did not express, what he was sure was the feeling of all around him, that not only in the Paper, but also by the mode in which he had handled the subject in the discussion, he had displayed the greatest skill and ability.

At the Meeting of April 27th 1858, the subject was again resumed, by the reading of some “Further Observations, explanatory of former remarks, on the Submerging of Telegraph Cables,” by Professor Airy. As, however, this communication had been published by the Author,  “without the previous consent of the Council,” contrary to the Regulations of the Institution (Section XIV. of the Bye-Laws), it could not be published in the Minutes of Proceedings. The following is a reprint of the Abstract which appeared at the time:—

Professor Airy commenced by remarking, that he trusted the importance of the subject, and the urgency of time, would be accepted as excuses for reverting to this matter out of due order; and the desire to arrive at accurate, instead of only approximate results, had induced further investigations and new records of their results. It had lately been stated to him, that the terminal velocity of the cable in falling through water, would not much exceed 3 feet per second; this led to more accurate investigation, in which no part of the resistance of the water was omitted; and the general result was:—that for the safe circumstances, when the angle of inclination of the cable, on leaving the ship, approached very near to the limiting, or critical angle, the views expressed by Mr. Gravatt were correct, the numbers given by Professor Airy, in a former Table, being considerably too large;—and that for dangerous circumstances, when the cable, on leaving the ship, was very little inclined to the horizon, the numbers in the former Table were very nearly correct. Thus, the inequality in the amount of tension of the cable, depending on the angle of its leaving the ship, was even greater than had been supposed; and the necessity for the severest caution in regulating the angle of delivery of the cable, was even more urgent than had been stated. In order to present the results in the clearest form, a new Table was prepared, on the assumption that the velocity of the ship was double the terminal velocity of the cable, when allowed to fall freely in the water. The unit of measure was, in all cases, the depth of the sea. The unit of tension was, in all cases, the weight of a piece of cable, whose length was equal to the depth of the sea, as weighed in sea-water. So that if, for example, the tension was expressed as 2.5, and the sea was 2 miles deep, the meaning was that the tension was the weight of 5 miles of cable, as weighed in water.

Angle made by
the Cable with the
Horizon on leaving
the Ship
Tension at
the point of
leaving the
Ship
Length of
suspended
Cable
Corresponding
horizontal
extent
Difference, or stray
length, necessary for
the slope of the
suspended Cable
2 ° 52 '
5   43  
8   34  
11   22  
14   9  
16   53  
19   32  
22   6  
24   30  
25   37  
26   34  
Limiting Angle
747.23
173.62
71.33
36.78
21.36
13.26
8.50
5.46
3.32
2.39
 
0.53
39.326
19.303
12.616
9.254
7.222
5.849
4.843
4.055
3.376
3.031
 
2.236
39.311
19.267
12.564
9.184
7.134
5.742
4.716
3.905
3.200
2.839
 
2.000
0.015
0.036
0.052
0.070
0.088
0.107
0.127
0.150
0.176
0.192
 
0.236

This Table showed the amount of tension in special circumstances, as well as the increase of tension, or the amount of danger, which might be introduced by mismanagement of the delivering apparatus. Thus, supposing the delivery to have been effected with a stray length of 0.236, and with a tension of only 0.53, if, by inattention to the mechanism, the delivery was impeded for a time, so that the stray length was diminished by 0.166, the stray length would be reduced to 0.070, implying that the tension was increased from 0.53 to 36.78; the only indication of this being, that the angle of inclination of the cable had been reduced from 26° 34’ to 11° 22'. A vigilant watch on the angle of inclination of the cable was, therefore, of the utmost importance for its safety; and the mechanical arrangements should faithfully exhibit to the superintendent, the slope of the cable on leaving the ship.

As to the general characteristics of the curve formed by the cable, he remarked, that in no case whatever was the convexity of the curve upwards. The speed of delivery being then supposed equal to the ship’s speed, and the terminal velocity of the cable, when falling freely in water, being divided by the ship’s velocity the angle, of which this quotient was the trigonometrical tangent gave the critical, or limiting angle. The lower part of the curve approached in form to the common catenary; but the inclination of the upper part to the horizon never exceeded the limiting angle. If the tension became great, the curve was nearly the lower part of a catenary of large dimensions. When the tension at the bottom was absolutely nothing, the form of the cable would be absolutely a straight line, lying at the limiting angle, with only a small tension at the point where the cable quitted the ship. When the speed of delivery was greater than the ship’s speed, he observed, that as the speed of delivery was augmented, the tension was diminished, until the proportion of the speed of delivery to the ship’s speed, became the same as the proportion of radius to the cosine of the limiting angle. There was then no sensible tension in any part of the cable, which descended at the limiting angle, and probably deposited itself in folds on the bottom. If the speed of delivery was further augmented, the cable would descend in a serpentine form.

In viewing the practical inferences, he observed, that when the cable was weak, or the sea deep, the cable should be delivered rather faster than the ship’s speed, by which means the cable could be safely deposited, as long as the ship continued in motion. If, however, the cable was strong, and the sea not deep, the velocity of the delivery need not much exceed the ship’s speed, but the angle of inclination should not be less than the limiting angle. In case of the ship’s motion being stopped, and the delivery of the cable being at the same time arrested, the tension of the cable would be much increased. If this occurred in deep water, the cable should be delivered out liberally, or the ship should be backed, until the cable assumed a position not much inclined to the vertical. By this precaution, little or no length of cable would be lost, as when the ship again advanced, without paying out cable for a short time, the depending cable would be quickly taken up into its former position, without any irregularity at the bottom. The minimum tension, when the ship was stationary, was taken as 1.00.

The suppositions adopted in the preceding investigation were, First,—that with velocities so small as that of the cable descending in water, the resistance was expressed by the sum of two terms; one depending simply on the velocity, and the other on the square of the velocity; and that when the velocity was very small, the former term alone need be considered, which had been the case in the present investigation. The results obtained corresponded almost exactly with those due to a rather greater velocity of the ship. Secondly,—the cable was subject not only to a motion transverse to its length, but also to a longitudinal motion, and to the consequent friction which accompanied it. No data being known for estimating this, it had been assumed as equal to the lateral resistance, at equal speeds. Allowing for these, it was believed that the results of the theory promulgated were perfectly accurate.

It had been stated, that the strength of the Atlantic cable and its weight in water were such, that it would safely carry a length of about four miles and a half, or, as expressed in terms of the depth of the sea, about 2.00. Consequently, the mode of delivery must be such, that the tension could never exceed the value 2.00. This tension occurred so near to the limiting angle, that it would not be safe to trust to the delivery at the limiting angle, with a velocity equal to the ship’s speed; but the delivery must be more rapid. The length of cable to be provided must therefore be greater than the run of the ship. From all these considerations, it appeared probable, that the practicability of laying down the Atlantic cable in the present year might depend, principally, upon the numerical value of the element, hitherto not ascertained, of the coefficient of friction, for a cable sliding end-ways through the water.

Professor Airy explained a series of diagrams, exhibiting the cable, under a variety of conditions, when being submerged. It was shown, that the general character of the curve was the hyperbolic, much resembling a catenary at the bottom, and going off towards an inclined asymptote at its upper extremity. It had been an object of interest with him, to ascertain with certainty, what was the position of the cable at the bottom; because, it would be remembered, that in some of the citations made in the Paper first communicated, it was stated, that in paying out a cable, under those conditions, there must necessarily be considerable loss of cable at the bottom. Such was not the case. But when a cable was delivered in a straight line from the ship, at a greater speed and an alleviated tension, it would be deposited in some irregular kind of wavy form of which he could give no account. When the cable was given out with an unnecessary degree of waste, the tension was not thereby diminished, and, at the same time, there was loss of cable on its way to the bottom, as well as at the bottom, when deposited.

He had only farther to remark, that the more important object he had in view, in presenting this communication, was to direct attention to the data that were wanted at the present time. Those he apprehended were the coefficients of resistance and the laws of resistance. The lateral resistance of a cable in moving through water was well known; but as far as his own knowledge went, he was not in a position to state the resistance to the longitudinal motion of the cable through the water. It had been correctly conceived by engineers, that if there was no resistance to the end-ways motion, the tension on the cable at the ship would be the weight in water of a piece of the cable equal in length to the depth of the water: but if there was resistance to the sliding motion of the cable, then, by sufficiently easing out the cable, the tension from the ship might be reduced to nothing, but of course with waste of cable; so that, in passing over a deep sea, the possibility of delivering a cable, with tension sufficiently small to insure its safety, might depend upon the amount of resistance to the longitudinal motion

Mr. Longridge agreed, that it was only by mathematical investigation, that a correct conclusion could be arrived at on this important question. In closing the discussion on this subject on a previous occasion, he had stated that the Paper by Mr. Brooks and himself was based upon such an examination, the formulae and calculations being given in extenso in an Appendix. It was desirable, that the mode of obtaining the results now recorded, should also be placed before the Members, in order that a comparison might be instituted. It might seem presumptuous in him to hint a doubt, as to the accuracy of the mathematics of the Astronomer-Royal, for whose attainments he had the most sincere respect; but from an anxious desire to arrive at the truth, he should wish that comparison to be made, as he could not but think, that the difference in the conclusions could only arise from the neglect, in one of them, of some of the conditions of the problem.

He would remark, in reference to an observation at the close of the communication just made, that in their Paper the coefficient of longitudinal friction, which afforded the only means of diminishing the tension, had been introduced as one of the forces affecting the result. These forces were: first, the tension at the ship; secondly, the weight of the cable; thirdly, the resistance of the water, which could be resolved into two forces, one at right angles, and the other parallel, to the cable, and acting by friction; and lastly, the horizontal tension at the bottom. The numerical value of the coefficient of longitudinal friction had been derived from the experiments of Colonel Beaufoy, and was probably somewhat less than what it would actually be in the case of the cable. But after all, the amount of ease to be obtained by the introduction of that coefficient was so small, unless the ship was progressing at a very high velocity, that, practically, it was not a desirable method of seeking relief from tension. It had been shown, that in the case of a cable running out quicker than the rate at which the vessel was progressing, the relief to the tension was inconsiderable, unless the ship was progressing at a very high velocity. If the speed of the ship was 10 miles an hour, and that of the cable two, or three times that velocity, then the tension might be reduced to nothing; but the conclusion had been arrived at, that, for all practical purposes, there was no benefit to be derived from letting the cable run out at a quicker rate than the ship was advancing.

He confessed, be did not quite understand, from the last column of figures given in the Table, whether what was stated to be the stray length, represented the waste when paying out at the angles given in the first column, being the difference between the length of cable suspended and the horizontal length. The tangent of the limiting angle was stated to be equal to the ratio between the terminal velocity of the cable, descending vertically, and the velocity of the ship, when they were respectively 3 and 6 feet per second, and the cable was being payed out at the same rate as the ship was advancing. But the tangent of 26° 34’ was just one-half, So that the table must have been calculated upon the supposition, that the vessel was progressing, and the cable was being payed out, both at the rate of 6 feet per second. The angle 26° 34’ did not differ, materially, from that given by Messrs. Longridge and Brooks; but it was not possible to imagine, that any circumstance could arise in practice, by which this angle could be reduced to 2° 52', unless the rate of paying out was below that of the velocity of the ship, or the paying out was altogether stopped, in either of which cases the result must be to break the cable. If the cable was being payed out at the angle of 26° 34', and the angle was then reduced, by diminishing the rate of paying out, it was clear the cable must be broken; and that was just what they had said would happen, in case the paying out of the cable was stopped. That was not what ought to be expected in practice; as the object was to lay the cable, and not to break it. They had desired to ascertain, if it was possible to lay the cable without tension at the bottom, and, at the same time, without waste, or slack; and a formula had been given, for calculating the amount of tension requisite to insure that result, under any varying circumstances. It was clear, that if there was no tension at the bottom, the cable must descend in a straight line. The catenarian form was impossible, except upon the hypothesis of tension at the bottom. Now, if the velocity of the cable exceeded that of the ship, there could not possibly be any tension at the bottom, and the cable must take a straight line; it could not be concave upwards. There were no forces in action to cause the cable to assume a wavy form in its descent. It must run down in a straight line, and be deposited in wavy folds at the bottom.

It had been stated, that the resistance of the water consisted of two parts, one of which was as the velocity simply, whilst the other was as the square of the velocity. It appeared to him, that the latter had been neglected in the present inquiry, but as he and Mr. Brooks thought it to be the more important of the two, and indeed believed that the assumption, that the resistance was a function of the square of the velocity, was very nearly correct. it had been adopted in their investigations.

Under all these circumstances, he was compelled to dissent from the views contained in this communication, which could scarcely be considered applicable to the submerging of telegraphic cables.

Mr. Bidder, V.P., remarked, that when this question was first discussed, he felt that the Astronomer-Royal had neglected certain elements in the consideration of the subject, and he had then ventured to comment upon those omissions. He still thought, that the data now brought forward were not such as could be practically relied on. He could not see how the circumstance of a body sinking in water, at the rate of 3 feet per second, could be amalgamated with any mathematical deduction from the tension due to the catenarian form, which the cable might be assumed to take in sinking through the water. No doubt, the mathematics were perfectly correct; but the value of the deductions depended upon the correctness of the premises upon which the investigation was based; and on that point he differed from what was now stated. The question to be decided was, the tension which was due to a cable payed out of a vessel at the rate of 6 feet per second, and which, if left to itself, would settle through the water at the rate of 3 feet per second. A certain regime must then be established, and he believed it was from the triangle which these two rates gave, that the angle of 26° 34’ was deduced. If the cable was disconnected, it would descend in a vertical direction. All the vessel had to do, was to pass the cable through the water, and to overcome the displacement. The quicker the vessel proceeded through the water the better, provided the cable could be payed out without kinks, and no more tension was required, than was sufficient to overcome the friction in the apparatus, from which the cable was payed out. It was impossible to amalgamate this hypothesis, with the existence of any catenarian curve. The Table showed, that at an angle of 22° 6', the tension was represented by 5.46, whilst at 24° 30’ it was only 3.32, so that increasing the angle by 2° 24’ lessened the tension by 2.14. Again, at 25° 37', which was a variation of only 1° 7', the tension was reduced 0.93; whilst at the angle of 26° 34’ the tension was represented by only 0.53. If this principle was carried out, it would lead to the result, whatever might be the terminal velocity of the cable, that as the limiting angle approached the vertical, the tension would appear to be 0, whereas it must incontestably be represented by 1, that was to say, the weight of a length of cable, weighed in water, equal to the depth of the sea. The inconsistencies he had pointed out, in the Table, were irreconcileable both with fact and experience. It had also been said, that there was no great danger when the cable was vertical, as the moment the vessel was set in motion, the cable became inclined again. Now, that could not take place without great resistance from displacement, and that displacement must be so large, in great depths, that it would inevitably break the cable. He would only add, that he knew of no resistance of a body in water, that was coincident with the simple velocity. Whether in friction, or displacement, his experience always had been, that the resistance was as the square of the velocity. He did not think there was much to fear from longitudinal friction. If the cable was payed out properly, all that had to be attended to was its subsiding, and the greater the velocity of the ship, so as to reduce the Horizontal angle to a minimum, the less would be the tension.

Professor Airy said, the whole question of submerging a cable was a problem of a most abstruse nature, far exceeding the complication of the motions of a planetary body through the heavens. The motions of any given point on the cable could only be expressed in terms of a certain curve, yet unknown, which must be assumed to travel on, while the point assumed different positions on the curve. The forces in action were,—first, a certain tension of the curve, with a certain inclination on that part of the curve which was above the point in question;—secondly, a different tension, with a different inclination on the part of the curve below it;— thirdly, the weight of that portion of the cable;—fourthly, the resistance to the lateral motion;—and fifthly, the resistance to the longitudinal motion;—all these must be expressed by means of a curve yet unknown. It had then to be ascertained, whether the forces produced in this manner, and depending upon the form of the curve, on the one hand, would produce these motions, depending upon the form of the curve, on the other hand. These results could not be arrived at intuitively, but must be worked out by mathematical investigations which were very abstruse. If he had been aware that the details of the investigation he had made would have been desired, he would have furnished them; he had intended to publish them in a periodical devoted to such matters; but he would take care that it should be placed, as early as possible, before the Institution. The possibility of some of the conclusions had been doubted. Now he might state, that, when the cable was given out at a proper speed, the combined resistances, transversely and longitudinally, to the cable, would allow it to slip down the incline plane, retaining its straight form to the bottom; but if it was payed out faster than was consistent with the retention of the straight form, then it must take the curve he had described. He would further observe, that the Table had been applied, in the discussion, to a very different case from that for which it was computed, and he contended that, to the circumstances described, it was strictly applicable, as would be seen from the calculations, the details of which he had promised to transmit.[9]

[9] These calculations have since been published in the ‘Philosophical Magazine,’ for July, 1858, pp. 1 to 18.—ED


Return to the Atlantic Cables index page

Last revised: 5 August, 2012

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: [email protected]

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