Escher.gif (426 bytes)

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

Sir James Anderson on Submarine Cables, 1872

Introduction: On 18 June 1872 Sir James Anderson, former Captain of the Great Eastern and at that time General Manager of the Eastern Telegraph Company, presented a paper to the Statistical Society on “Statistics of Telegraphy”. This was published as a book in July of the same year. A section is included on submarine cables, their design and construction, and reasons for their failure. That section of the book is reproduced here.

The Submarine Cables section of the book begins with drawings of thirty examples of cable, described as “Types of Cables which had been tested at the time the Joint Committee was appointed by the Lords of the Committee of Privy Council to enquire into the construction of Submarine Cables.” Three of the diagrams are referenced in the text, and those cables are shown in the appropriate place below. All the cables are shown in sequence at the end of the page.

--Bill Burns

Statistics of Telegraphy
Sir James Anderson


THE accompanying table [not shown] contains as accurate a statement as I can obtain of all the submarine cables laid up to this date, and in the following remarks I state, what, in my opinion, this experience of twenty years has established.

This is by no means a new subject of investigation, but in the present day I am certain it will be instructive to many amongst the thousands who are now interested in this class of property, to have their attention briefly called to all that has been done to make submarine cables a sound property.

Eleven years ago there was a Joint Committee appointed by the “Lords of the Committee of Privy Council for Trade and Atlantic Telegraphy, to enquire into the construction of submarine cables, together with other evidence.”

The report is dated April, 1861, and is signed by Douglas Galton, C. Wheatstone, W. Fairbairn, Geo. P. Bidder, Edwin Clark, Cromwell F. Varley, Latimer Clark, and George Saward, and they state that they had the benefit of the advice of the late Mr. Robert Stephenson.

They examined 40 witnesses, all eminent in their day, and numbering amongst them most of the names which are yet conspicuous in the engineering, manufacture, and submerging of this class of property.

Fifty cables had been laid at the date of this investigation, all upon the same general principle.

Eight thousand miles had been lost, all belonging to four undertakings, viz.: The Atlantic, Red Sea and India, Sardinia-Malta, Corfu-Malta, and Singapore to Batavia Cables.

They state that the loss of all these cables was “attributable to defined causes which might have been guarded against,” and they “believed there were no difficulties to be encountered which skill and prudence would not overcome.”

The Committee considered it unreasonable to expect more rapid progress than had then been made, the first few cables laid in shallow water across the Channel were comparatively easy to recover and repair; they had been manufactured without much necessity for extreme care, and had been accepted as successful precedents; further investigation was considered unnecessary, and bold attempts were made to lay cables of a similar type under entirely different conditions, but “they considered it doubtful whether the transmission of messages for even so short a period as three weeks through a cable across the whole width of the Atlantic was not a result worth all the expenditure which had been incurred.”

Attention is called in the report to the “remarkable fact that in almost all cases small cables had been found liable to mishaps, while the heavier the cable had been the greater had been its durability.”

At the date of this report, the types of cable illustrated on the preceding pages [see below] had been experimented upon, upwards of 1,300 tests had been made by Messrs. Siemens, Forde, and Gisborne for H.M. Government, with the object of discovering the best form of cable, and many hundreds of tests besides had been made by Messrs. Glass Elliot, Messrs. Newall, Messrs. Siemens, Messrs. Silver, and many others with the same object. At this date 1859 to 1861, there were ample data for investigation, and there were many eminent and practical men of experience in this class of enterprise; we are not therefore disappointed in the result of the enquiry. The report is full and complete, and established principles which up to the present time have uniformly guaranteed success, while the neglect of them has as uniformly resulted in partial loss or failure.


The loss of cables was found to be attributable to the following causes:—

A. First, and the most important of all, from imperfect manufacture, resulting without doubt prior to this date from inexperience of the materials for insulating the copper wire, and from ignorance of the fact discovered by Professor Thomson about 1856, viz., that some kinds of copper wire were no better than iron for the purpose of conductivity, and that it required carefully selected copper to give the desired standard, which may be represented by a copper wire one-tenth of an inch in diameter, being equal to an iron wire one-third of an inch in diameter, for electrical purposes.

All cables manufactured previous to this date had no advantage from this discovery.

There appear to have been mechanical difficulties in keeping the copper conductor in the centre of the insulating medium, so that the copper was sometimes found to be almost visible under the light film of gutta percha which covered it. The electric current soon weakened this film, stronger currents were used to overcome the weakness of the signals, and the cable was soon destroyed. Experience about this time had established that

A cable from the commencement of its manufacture to the time of its being laid should be tested under water and under pressure, and kept as much as possible under all the conditions in which it was meant to continue.

Sir S. Canning taught that the “great secret was to keep a cable quiet from the time it was made until it was laid,” and no one disputed the fact that every time a cable is coiled or uncoiled it sustains more or less injury.

B. Attempts to lay cables from sailing ships towed by steamers was another source of failure. The ships had not enough steerage way when met by strong head winds, and too much slack was paid out. It was difficult under such circumstances to steer a straight course, and sailing ships possessed no power of being readily stopped when a fault or accident occurred.

C. Many accidents happened from inexperience in the method of paying out cables; at the present day the wonder is that they should have succeeded so well with the rude methods and inexperience which then existed, and not that there should have been many failures and much recrimination. Reading the history of these first attempts to place a network of cables at the bottom of the ocean fifteen and twenty years ago, is a good deal like reading the old stories of the early voyages of discovery. There are difficulties and disasters peculiar to every attempt, and the grand result is that one way or another they were overcome, or else they suggested such modifications—that their recurrence was avoided, and an accident to a well-manufactured cable no longer constitutes a loss.

We read of the vessel paying out the Toulon-Algiers cable being run into by the French ship sent to assist her, and the cable, although buoyed, was lost.

Another attempt failed “from a fracture due to the occurrence of a storm.”

They were five days in laying the Corsican cable, a distance of only 70 miles. “They used to anchor at night holding on to the cable waiting for daybreak.”

The first attempt to lay the Sardinia-Bona cable failed from the cable breaking while trying to recover it by “heaving it in with the windlass.”

In the second attempt they ran short of cable; the vessel sent to guide led them out of their course. When day broke the ship which was leading was dressed with flags ready to land the cable with startling éclat, but they were steaming in the wrong direction, and there was not cable enough on board to allow for the error which had been committed. “The ship held on to it for four or five days, sent another steamer to bring assistance; rough weather came on, and the cable broke in 400 fathoms.” The third attempt failed, “owing to imperfect manufacture.”

The first Atlantic cable failed principally on account of imperfect manufacture, in a great measure arising from undue haste and urgency, but largely owing to insufficient experience.

The cable was not tested under water for fear of rusting the small steel wires of the external covering, and small wires have never since been used; large wires, the larger the better, is now a principle.

The copper was not all good.

It had often been coiled and uncoiled, and had been exposed to the strong heat of the sun and to many changes of temperature.

Any of these conditions would now-a-days be regarded as enough to condemn the most carefully manufactured cable.

The Red Sea and Indian cables are said to have been imperfectly manufactured and laid too taut, but they were not tested under water from the time of manufacture until they were placed at the bottom of the sea, and this one grand omission, largely due to inexperience, is enough without the recriminatory points to condemn to loss and failure any cable whatever.

The cables laid from Cagliari to Malta and Malta to Corfu are said to have failed from imperfect manufacture. One experienced gentleman in his evidence said these cables were “such as nobody should have laid in deep water.” It is sufficient at present to know that they failed from neglect or inexperience, and that they, amongst other failures, have established the principles which have since ensured success.

D. The want of constant supervision by engineers, exclusively in the interests of the purchasers of the cable, has been a great cause of defective cables.

There may often be minute defects in the core itself, or a slightly defective splice which may reduce the electrical condition of a comparatively short length; this may easily be raised above the average standard required by the contract, by the next length being more carefully manufactured.

These minute defects must, however, kill the cable in more or less time, and the principle is established that

Every inch should be tested in course of manufacture, and rejected if there is any irregularity of condition to cause suspicion.

There should be constant supervision, and a record of all the tests kept for the purchasers of the cable, from the commencement of the contract to its final completion, and continued ever afterwards by the purchasers.


The principle sources of injury to cables are: 1st, moving water, either currents or tides, chafing the cable upon rocks or shingle. Experience has given many costly lessons of the effect of moving water.

Ten years ago it was generally believed that water had very little motion below fifty fathoms, and one hundred fathoms was considered a point of great safety. We now know that there are exceptional localities where there is motion in the water at a depth of five hundred fathoms. The Falmouth cable was chafed and destroyed at this depth from this cause.

The Channel Islands cable was also destroyed from the same cause.

The first cable ever manufactured with due regard to the principle of careful supervision, testing under water, and being retained quietly in that condition until it was laid, was the Malta and Alexandria Cable laid in 1861.

This cable was submerged in too shallow water, for many miles in less depth than twenty fathoms; the result was the frequent recurrence of fracture from being rolled about by the surf, and yet this cable was only finally abandoned last year; not because it could not be kept in repair, but because it cost too much to keep it in order.

These and many other examples have established the principle that

No cable should be laid without first obtaining an accurate survey of the approach to the coast and landing places with accurate soundings over the intended route, and as much knowledge as possible of the nature of the bottom.

Currents and anchorage should be avoided, and where that is impossible the heaviest cable that can be laid should be provided.

Heavy cables should be laid out to depths of four hundred fathoms where there are tide-ways.

Where a current exists a position should be sought for as far removed from it as possible.

A great cause of injury to cables is the corrosion of the external wires caused by moving water or marine vegetation, &c., and this has established the general practice of covering the external wires with tarred yarn saturated with a mixture of pitch and silica. There is still great room for improvement upon the present method of protecting the external covering of cables, and I commend it to the further careful study of telegraph engineers as a subject of vital importance.

Another enemy of submarine cables is the toredo of all kinds; there is one kind which has proved destructive by boring through the core, but that has only occurred in shallow water; there is another kind which destroys the hemp in a few months, and is then satisfied to fix itself upon the gutta-percha and remain there. Cables have been recovered from depths of twelve hundred fathoms with all the hemp eaten away, and the core pitted with these marine animals. The recovery is then only possible by the strength of the external wires.

All the experience we have points to the value of protection, first, of the core, then of the external covering, and if those responsible for the safety and maintenance of submarine cables could be allowed to dictate the most desirable conditions of safety they would select, besides the strongest possible cable to be manufactured and laid with extreme care, a depth of water about five hundred fathoms, and a bottom of sand or mud; but as this cannot always be secured, nothing should be omitted in the direction of strength, and quality.

Lightning is still another source of injury to cables, this is however, so readily guarded against that we no longer hear of injury from this cause; it is said to have destroyed three cables. Mr. Siemens produced before the Committee a piece of the core of the Corfu cable injured by lightning; the land line had been struck, and from the absence of any lightning guards, the cable was damaged.

Mr. Preece described the Jersey cable to have been destroyed by lightning.

Mr. Fleeming Jenkin had seen a fault eighteen inches long due to this cause, and it is asserted that the same cause destroyed the Toulon-Algiers cable, which was connected to the land lines without lightning guards.


The most frequent injury arises from the wire with which the cable is covered, being too brittle or parting at the scarf joints, and at once becoming little poignards, liable to pierce the core during the process of laying. The necessity for laying the cable at a moderate speed and with great care, prolongs the voyage across a broad ocean for many days and nights, and it is not surprising that these broken wires should at times pierce the core and necessitate the instant hauling back of the cable, no matter what the depth of water, or what the condition of weather may be, and this establishes the principle:—

That all cables should be made with due regard to the depth of water in which they are to be laid, and strong enough to admit of being recovered in case of accident, which may as probably occur during a tempest as during a calm.

But accidents from this cause seldom or never occur (I do not know of a single instance) when the external wires are covered with yarn and bituminous compound; this covering has therefore the double value of protecting the external wires, and adding greatly to the safety while laying.

There are, besides, accidents liable to occur at sea which no human foresight can guard against; over a period of ten or twelve days, more or less bad weather is almost certain to occur, and should at all events be provided for by a margin of strength..

What are called foul-flakes and kinks, and accidents to machinery and to the men have occurred, and may occur again, requiring the ship to be suddenly stopped, and great strain to be thrown upon the cable, and it is sometimes necessary to cut and buoy, and leave it for several days.

That accidents need not occur often, and might not occur at all, at times, is not sufficient argument to justify a cable being made unequal to an emergency.


We are every now and then startled by the announcement that light cables are to be preferred to the present iron-clad type, and the object of this investigation has been to discover what data there are to justify any preference to one form of cable over another.

I have said already that the Committee called attention to the remarkable fact that, in almost all cases, small cables had been found liable to mishaps, while the heavier the cable the greater had been its durability.

Mr. Newall, in his evidence, said that the hemp covered cable which he attempted to lay in 1859, between Candia and Egypt, had the hemp eaten off by the toredo in a very short time, and it was too weak to recover for repairing.

The same firm laid an unprotected core from Varna to the Crimea, and it lasted until the winter set in; it is frequently said that it was cut by order of the French Commander-in-chief, but there is no proof of this, and I am not disposed to believe it. Mr. Woodhouse, the engineer, who laid this core, said in his evidence “he should not advise anybody to lay so light a cable across the Atlantic, because so small a strain would break it. If it is once safe  at the bottom perhaps it may rest.

Mr. Newall said he thought it folly to lay anything excepting unprotected core. Consistently with this conviction he laid in 1869 several lines of unprotected India-rubber core, connecting the Grecian Islands with the main land; they were protected only near the shore.

The sea is quiet and tideless in those parts; no better spot could be wished for the experiment, yet all of them gave out within two years.

The Red Sea Cable, covered externally with light wires, and unprotected with bituminous compound, was so rusted in a short time that it could not be lifted for repairs.

Notwithstanding Mr. Newall's partiality for light cables, he suggests at the close of his evidence what I assume he would consider the most perfect form of cable. He would cover the copper with india-rubber, protect this core with steel wires, vulcanised, the whole then passed through heat; thus insulating all the wires, he would make the cable in one length, and have no joints.

Mr. Lionel Gisborne considered a hemp-covered cable “perfectly useless for laying in water; it has both the liability to stretch and to shrink.”

Mr. Fleeming Jenkin, in his report to the International Exhibition of 1862, says:—

“So long as the iron wires lasted, the cables frequently continued to work in spite of faults, but sooner or later the iron wires of all these light cables rusted away in parts; so soon as this took place they one and all broke up into short sections; this fact has been observed in depths of one hundred fathoms” the reasons were not obvious to Mr. Jenkin, but he says, “meanwhile the use of large iron wire seems a sure guarantee against this danger, for as yet no cable covered with wire of the large gauges has ever parted in the manner described.

The difficulty is to find a permanent material which shall retain its strength and continue to afford protection after the cable is laid.

Every word of this can be written at the present moment, that is, ten years later, with exactly the same significance.  All cables which have been manufactured and laid upon the principles which were established in 1859, are yet in good working order, and every divergence from these principles has been at best a costly experiment or utter failure.

It is urged as a strong reason in favour of unprotected core (light cable) that there are many miles of cables now in existence from which the outer covering has fallen off by decay or otherwise; but I am not of that opinion, and it can only be an opinion. In many cases, perhaps in all, the outer covering may have lost much of its strength, but it is more likely to have the merit of keeping the core protected and undisturbed, owing to its weight and accumulation of deposit upon it, than to have fallen off and left the core unprotected.

I am of opinion that whenever the outer covering falls off, the life of the cable will be very short; and I am prepared to expect that in many of the cables now laid all the shallow water parts will have to be renewed from time to time.

There is no instance yet of a well-manufactured heavy cable breaking or giving out in deep water, where currents have been avoided, after it has been carefully laid free from defects, but there may be much due to the external covering keeping it quiet, there has assuredly been a great deal due to the external covering in the successful submerging, and there is no experience whatever to justify the assumption that an unprotected core would last, even if laid.

It has been urged that an iron-covered cable suspended from one point to another gradually becomes weaker, that rust and marine growth or deposit accumulate and break the cable with their weight; but I do not know of any instance in support of this assumption, nor is it at all certain that a simple unprotected core would exist for any length of time, or be in any way better adapted for the supposed conditions.

Mr. Latimer Clark in his evidence says:— “You want a certain degree of weight to enable your cable to sink steadily to the bottom, especially when it has to fall into hollows and cavities, and not lay loosely across elevations.”

Again, it is urged that experiments with light cables have been tried in factories or sheds, and the result proves that there are many advantages in their favour; but I am of opinion that no experiments which can be made on shore will sufficiently resemble the exigencies which may occur over a period of several days and nights at sea in storms and darkness, and still less will they prove their fitness for the unknown conditions which may exist at great ocean depths.

I desire to write with great respect for the opinions of the talented men who urge the adoption of light cables; it is my special duty to weigh well and without prejudice all they have to advance, but I think a careful investigation into the experience and practice of the last twenty years establishes conclusively that all light cables have been short-lived, and that all heavy cables have continued working, often under most adverse conditions.

Fig. 12

It is my own opinion, and I am authorised to say, that it is also the opinion of my friend Captain Halpin, who has laid all the cables from Suez to Australia, besides the French Atlantic Cable (eleven thousand miles) and has also recovered and repaired cables from a great variety of depths—that a cable should be as heavy as it can be laid with safety and admit of being recovered in case of accident. Multiply every precaution which shall increase the strength and keep that strength intact as long as possible.

The best form of light cable I have seen is the copper-covered core invented by Mr. Siemens (fig. 12). I should have anticipated that if any light cable could have been successful, this one would have met all the conditions, excepting that of extreme cheapness, but it has not been so uniformly successful as the heavy iron-clad cables.

The very light cable invented by Mr. Varley (fig. 30) admits of being laid by having the strain taken off the core by the two hempen strands, the core itself being the third strand of the cable. As a light cable to be manufactured in a great hurry and laid to meet some emergency, it has a good deal of merit, but for a deep sea cable, I am of opinion, that it would be found too incomplete and unfinished, and that difficulties would be experienced in laying, which are not at once foreseen, and there would be no durability even if successfully laid.

Fig. 30

Every day of my experience in watching over the permanence of the ten thousand miles of cable under my care, confirms me in the opinion that too great caution and vigilance cannot be exercised in making and laying a thread which is to be removed from all human vision for ever, and designed to earn dividends by continuing a perfect conductor of electricity.

Upwards of thirty thousand miles of cable have been laid since the report of the Committee was printed eleven years ago, and much experience has been gained of the exigencies incidental to submerging, buoying, grappling, and repairing; but no fact has resulted from all that experience which has established that any one precaution recommended in the report has been superfluous, whereas much has occurred, which I will not particularise, proving that any attempt to disregard any single precaution has resulted in great pecuniary loss or utter failure.

We have many reasons to confirm the belief that a submarine cable, manufactured and laid with strict attention to all known principles, may be regarded as a substantial property, likely to last for any length of time; for there is no evidence whatever upon record which shows any decay of the insulating medium or copper conductor of a well-manufactured cable, i.e., “there is no decay inherent in the nature of a cable, all deterioration is external;” nor is there any experience whatever to establish that this insulated copper wire will enjoy any durability if unprotected with an external covering.

A light cable or unprotected core must therefore be regarded at best as an experiment, with the chances against the successful laying, and still more against its existing as a permanent property.

Fig. 28

I have written enough to illustrate that the present submarine cable (fig. 28) is not a haphazard idea, but one which has grown out of many failures and thousands of experiments; all the principles of manufacture and laying down have been established by great anxiety and reflection on the part of the able men who gave their energies to this kind of enterprise prior to 1865. We who have come upon the stage since that date, have only discovered that we may not neglect one of all the known principles, but if possible elaborate every one of them, and even then the duty and responsibility of laying and maintaining this class of property, has enough of risks and anxieties to make one heartily dislike any experiment which can only be advocated for the sake of cheapness in the first cost. I believe this economy would be at the expense of security, and that the cable of the future will be even heavier, more perfect, and more costly than the cable of the present day.

Types of Cables which had been tested at the time the Joint
Committee was appointed by the Lords of the Committee of Privy
Council to enquire into the construction of Submarine Cables

Silver & Co's





Hall & Wells'



De Bergues'

L. Clark's



L. Clark's



Red Sea 1860

Atlantic 1858

Candia-Alexandria section

Atlantic Cable 1865 - 1866

Type of Present Successful Iron-Covered Cable

Red Sea, 1871, Shore End


Anderson, Sir James: Statistics of Telegraphy. London, Waterlow & Sons, 1872.

Bodleian Library, Oxford/Google Books

Copyright © 2007 FTL Design

Last revised: 30 November, 2008

Return to Atlantic Cable main page

Research Material Needed

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

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

If you have any cable-related items that you could photograph, copy, scan, loan, or sell, please email me: [email protected]

—Bill Burns, publisher and webmaster: