Escher.gif (426 bytes)

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

Teredo Tape

Introduction: As engineer to the Telegraph Construction & Maintenance Company from 1864 to his retirement in 1894, Henry Clifford was concerned with a common problem encountered when laying cables in certain conditions—a variety of marine borers, popularly called teredo worms or shipworms, but actually a species of saltwater clam. These tiny marine creatures found their way through the cable armouring and dined on the jute serving and gutta percha insulation, exposing the conductor and causing earth faults.

Clifford devised a number of solutions, and his application of a spiral-wound Muntz metal tape came into general use in the late 1870s. The section below from Charles Bright’s 1898 book, Submarine Telegraphs, gives a comprehensive description of the problem, and of Clifford’s part in solving it.

Bracketed numbers in the text refer to footnotes in the original article, which here are all presented at the end of the page.

--Bill Burns



For those portions of any cable which are about to be deposited in waters where marine organisms exist, the core now very usually first receives a sheath of metal tape by way of protection.

Submarine Borers.—These enemies to cables consist mainly of what are known to naturalists as Teredo navalis [1] xylophaga, and Limnoria terebrans or Limnoria lignorum. [2] They are all various species of marine ravaging insects, or worms, provided with boring tools, which they make use of to pierce into the cable. Beyond idle curiosity, these creatures appear to have a penchant for jute and hemp, judging from the extent to which they have been found to devour it. In satisfying this taste, and in order to burrow a passage, these borers have a habit of also “scoring” the gutta- percha in the form of an elongated groove, though the actual meal appears to be, usually, in the main, made off the jute or hemp. [3] Vulcanised rubber seems to be too hard for their proboscis, or else the sulphur disagrees with them, for it is only gutta-percha cores that are so affected: hence rubber cores never require metallic taping as an anti-boring sheath. Though the core is, as a rule, only influenced quite superficially, cases have been known where these borers have penetrated down to the conductor, thus establishing more or less “dead earth,” electrically speaking. In any case, a tendency towards weak insulation is liable to be set up at this point, especially as the serving becomes conspicuous by its absence, and particularly when a loose place for the water to more thoroughly act on the iron wires, and on the gutta-percha, is established—eventually culminating, perhaps, in the water reaching the conductor. Such faults are naturally likely to occur where the sheathing wires have—due to chemical, or mechanical, action—become to any extent corroded or worn, leaving the serving open for comparatively rapid consumption.

All these ravagers are sometimes spoken of by the general name “teredo.” This is inaccurate.

The Teredo navalis is mainly to be met with in the Mediterranean Sea, and has probably been responsible for most of the experiences to which we allude. In higher latitudes, however, the “teredo” proper is scarcely known. For instance, it is the Limnoria terebrans which mainly affects our waters round about the English Channel and Irish Sea. This mollusc-like worm is found more particularly to confine its attentions to the jute, or hemp, serving. Thus, the cables under the control of the British Postal Telegraphs, since 1893, have been invariably brass-taped. The tape is applied outside the serving, [4] by way of evading these ravaging insects. [5]

Faults due to marine borers have been very usually found to occur at the mouth of a river, where, owing to various forms of food coming down the river, animal life is more prevalent. They also occur along more or less tropical coasts.

In any case the metallic riband is only applied to the core of that part of the cable likely to go into such insect-ridden waters, which more commonly means in depths under 250 fathoms, [6] though in warm climates the teredo has been known to exist at much greater depths—1,000 fathoms, or even more. [7] It is, indeed, almost entirely a question of bottom temperature which determines their presence, or otherwise; and this again is largely governed by currents. [8]

The Teredo navalis—a sort of soft-bodied snail—is but little bigger than a pin’s head.

Even when new, the sheathing wires of a closed cable are never likely to butt tightly together throughout the entire length owing to occasional irregularities. Thus, it can readily be understood how such a minute creature finds its way in; and with any species of open-sheathed cable, it becomes a comparatively simple matter.

The other borers are, however, a good deal longer—about ¼ inch, as a rule. These are said to exist ashore, and to have been traced in underground lines. [9]

Further particulars regarding marine animal and insect life—together with illustrations—will be found elsewhere: for full details, the reader is referred to an excellent early paper on the subject due to the late Mr George Preece. [10]

From what has transpired it will be readily understood that the metal taping here spoken of to ward off marine borers is only applied, as a rule, to the cores of the shore end, and intermediate types of a cable—and then only for certain waters in cases where such objectionable life is supposed to exist.

History of Metal Tape.—The first instance of a metal riband being suggested for a cable was that of the brothers Bright. [11] As early as 1852, a patent (No. 14,331) was taken out by E.B. and C.T. Bright, in the course of which was mentioned the covering of cables with hemp yarn, and a very thin metal spiral riband, or taping, between two fine flexible woven tapes. [12]

According to this device, the metal tape was applied outside the hemp serving. Theoretically speaking, this would be correct for warding off the ravages of insects, inasmuch as it is the yarn they have a special weakness for, and which they play the most havoc with.

However, in practice some difficulty is experienced in making the metal tape lay properly over such an uneven bed as that afforded by the serving of yarns; and, thus, if any boring insect should find its way anywhere, it would be liable to cause more trouble than ever.

It was, in fact, found to be essential that the metal tape should have an almost absolutely flat surface as a bed to lay on properly. [13] Moreover, an overlap at each turn is, in any case, necessary.

Mr Henry Clifford was the first to pay special attention to this particular point, and—whilst Engineer-in-chief to the Telegraph Construction Company—to introduce in 1878 a method of metal taping peculiarly suited for this purpose.

According to Mr Clifford’s plan, the metal tape is applied outside the core instead of outside the serving. [14] It is this which is now in extensive use in the present day. Clifford’s brass taping was first applied by the Telegraph Construction Company to a part of the Eastern Extension Company’s Penang-Malacca cable laid in 1879. Since then it has been adopted for at any rate all the shallow-water sections of the Eastern, Eastern Extension, and Eastern and South African cables which are laid in at all tropical waters, where the teredo et hoc genus omne are liable to flourish, and where gutta-percha is the insulating medium.

In selecting a suitable metal for this purpose, it was necessary to combine lightness with as much toughness and strength as possible, [15] the question of economy being also considered of course. Thus, on these grounds (especially on the latter), as well as on that of strong chemical action, it was at once found that copper was unsuited for the purpose.

Certain forms of brass were soon found—partly on account of the hardness introduced—to be particularly suited to attain the object in view. [16]

Though this metal riband is often spoken of as brass taping, it is nowadays more generally composed of that variety of brass which goes by the name of Muntz metal, being a compound of copper, zinc, and tin, containing rather more tin and zinc in it than ordinary brass, and correspondingly less copper.

This metal mixture is found to afford a remarkable degree of hardness and toughness to a comparatively small weight, besides which it introduces little or no chemical action. [17]

Recently phosphor-bronze [18] has been occasionally used for this purpose instead of ordinary brass, or Muntz metal. This is with the idea of still further combining the maximum strength with the minimum weight beyond that effected by Muntz metal. Again, phosphor-bronze is supposed to be more durable, though certainly more expensive at the outset.

Present-day Application.—According to Mr Clifford’s plan, first of all a fine (wet) cotton tape (about ½ inch broad) is applied spirally with about half overlap by way of bedding and preservative; then (2) the metal riband of same breadth and 4 mils, thick is laid on with half overlap; and finally (3) a waterproof preservative and protective cotton tape, rendered so by being previously soaked in stearine (mineral wax), ozokerite, or other resinous compound. [19]

These three tapings are each applied [20] alternately [21]—at one operation to save time—with opposite lays (the length of which depends on the width of the tape adopted), with a view to more thoroughly covering, upon each side, every overlap of metal tape, for its better preservation and waterproofing, as a complete armour against the ravaging insects. Inasmuch as in a brass-taped cable the tape acts practically as the main “earth,” it is undesirable that the inside tape should be compounded in any way, for fear of masking faults. [22]

By the Clifford system all these tapes are applied at one operation in a manner which will be described later.

The Silvertown Company use wet cotton threads, or yarns, as a bedding for the metal tape in place of the calico tape. This, in the author’s opinion, has certain advantages. It forms a softer bed, and avoids the ridges that are a very usual accompaniment with any tape. The difference in this respect mainly rests in the fact that the cotton threads, tending naturally to fill up the space on the surface of the core, do not require to be applied with any overlap. It also gets over the danger of ravelling (or “rucking” up, and of the riband pressing into the core at uneven places as above. Again, threads retain the moisture better than tape. [23]

Fig. 53 shews a recent form [24] of the ordinary three-head taping machine, [25] capable of applying the three coverings in one operation after the manner of either of the above descriptions.

As will be seen from the illustration, each head is provided with a three-speed cone pulley, by which it is driven through the medium of belting from the main shaft, which can be seen running longitudinally along the base of the machine. The opposite lays of each taping are arranged for by the tape head cone pulleys being run in opposite directions, i.e., by the method of reverse belting from the main shaft pulleys, in the ordinary way. The taping heads are, as usual, each provided with a counter-weight on the opposite side of the mandrel to the taping bobbin, to balance the weight of the latter, and so do away with any vibration.

FIG. 53.—Taping Machine.

The machine has a powerful hauling-off drum (such as ensures perfectly regular motion) driven by cog gearing round the inner circumference of its flange. The pinion which drives it receives its motion from the main shaft through bevel gearing, which is enclosed in the box seen to the right of the illustration.

A set of spur change wheels is provided with the machine, and by means of them the lay of the various tapes can be adjusted to the required pitch, according to the size of core which is being taped; thus also allowing of considerable variation in the width of the tapes and the amount of overlap. The machine also has an indicator, which records the length of core covered. This, of course, varies materially with the width of tape, length of lay, etc., but three or four miles per working day may be taken as an average output.

The machine above described takes up very little floor space—only about 17 feet, indeed.

The various core tapings are sometimes applied at one operation, with the two layers of yarn constituting the inner serving. Where, however, the latter are applied at the same time as the iron wire sheathing and entire outer serving, the core tapes are usually laid on first as a separate process.

Each taping head holds about a mile of tape. The successive lengths of the metal riband (from fresh heads) are “brazed”—i.e., united with brass solder. The cotton tape ends are secured by a dab of adhesive compound, and in the case of threads each successive length is attached to the previous one by the two ends being “wipped” together.

Unfortunately Muntz metal, or any species of brass, tape is subject to corrosion, in some degree, by the action of sea-water. Various suggestions have been made to obviate this. The plan which appears to approach nearest to success is that of Mr Arthur Dearlove. This gentleman has proposed (Provisional Specification No. 20,645 of 1889) to coat the metallic strip with anti-sulphuric enamel, by passing the strip through a bath of the material. It is then dried by being drawn through a steam-jacketed cylinder previous to being wound on a reel, or “tape-head,” ready for use. The great feature about this particular enamel is that it is flexible and does not crack, especially when applied very thin—like a varnish Mr Dearlove has also suggested its application to the outer sheathing, with the same object in view. For neither purposes has it, however, been so far adopted in practice.

Provided it is properly applied, metal taping is supposed to be a fairly absolute protection against borers; indeed, no cases of trouble have been known to be traceable to this cause with a cable, the core of which is so sheathed. Moreover, it is to some extent effective in preserving the dielectric from decay (by oxidation, evaporation, and the result of alternating wet and dry conditions) in the instance of gutta-percha insulated cables laid in dry soils ashore. [26]

However, in the recent Amazon River cable, Messrs Siemens Brothers, to meet these conditions, made what was a new departure as regards submarine telegraphy. [27] The core of the shore-end type, instead of being brass taped, was covered by a lead tubing. [28] thus rendering it absolutely teredo-tight, as well as air-tight. The core was followed by a cotton taping, over which the lead covering was applied [29] and externally compounded. Outside this were two servings of jute of opposite lays, followed by the usual sheathings, etc.

For the land connections in the form of trench cables the same plan was adopted, with a view to more perfectly obviating the decay which gutta-percha is subject to in dry soil, as already explained. The trench cable consisted of core, cotton tape, leaden tube, compound, [30] and two servings of tarred jute, the proper length of this being previously let into each length of shore end without any joint or splice.

Alternative Suggestions.—Some authorities have objected to the adoption of any form of metal covering to the core, not only on the ground of additional expense, but also of electro-chemical action with the galvanised iron sheathing.

Thus, from time to time, various other means have been suggested for defeating the ravages of insects. To preserve subterranean wires from the attacks of ants, etc., one inventor (as early as 1854) thought of incorporating in the insulating material “some resinous or bitter substance of a poisonous nature.” An external coat of india-rubber and various mixtures with it have also formed the subject of a patent for the special purpose of protecting the gutta-percha core. [31]

Again, in 1877 provisional protection (Specification No. 1,416) was obtained by Mr H.A.C. Saunders and Professor Andrew Jamieson, F.R.S.E., for a method of saturating the inner serving with andiroba oil. This substance having proved successful in overcoming the ravages of white (”termite”) ants, it was thought that it would also defeat the teredo and other animalcula. Moreover, it was suggested that the iron wires should be so treated as a preservative against rust. Though apparently an excellent idea—it being the serving that forms the main seat of attack—nothing appears to have come of this plan in practice. [32]

Finally again, with the same object in view, the incorporation of powdered silica in the gutta-percha covering was suggested by Mr Willoughby Smith in 1875, and subsequently Mr Edward Stallibrass devised a method of blowing this material on to the outside of the core. Moreover, in 1878 the late Mr Frank Lambert secured a patent (No. 759) for applying a serving of silicated cotton—sometimes known as mineral, or “slag,” wool—to the core (in place of the ordinary jute serving), mixed with a suitable binding of agglutinating material, such as pitch, tar, various oils, or resins. As an alternative, a cotton tape was to be impregnated with silicate compound, as above, to be applied spirally round the core, as in the metal tape. [33] It was suggested that the silicate cotton compound should also act as a protection from moisture (amongst other things) for the outside tapes of a cable. This latter claim would, however, have materially vitiated the rest of the patent, probably, if it had been challenged. However, it is believed that none of these ideas have ever been put into practice on anything like an extensive scale; and the metal taping—the application of which we have described—is almost invariably the means now adopted to ward off boring insects.


1. This miserable little mollusc first made itself a reputation by eating up wooden ship hulks, until builders took to plating them. When the cable came, it took to it at once.

2, Amongst the foes to submarine cables there are also, of course, the saw and sword fish, which have a disagreeable habit, whilst hunting for dinner, of attacking a cable—mistaking it, maybe, for some submarine monster bigger than themselves—with their beak or snout. Saw-fishes vary in size from 10 to 150 lbs., and sword-fishes (sometimes confused with the former) are usually of dimensions such as are represented by a weight of about 100 lbs. These are mostly met with on the Brazilian and other tropical coasts, and especially near the mouth of rivers.

The shark and the whale must also be included in this category—indeed, several curious cases of warfare between a whale and a cable have been experienced. However, the attacks of these monsters cannot be rendered at all less effectual by any such metal taping round the core; though a sheathing of tough, close-sheathed iron wires is a natural precaution, in tropical shallow water, with this object.

Fish attacks and fish bites, as above, are dealt with in full elsewhere.

3. Injuries to cables by marine organisms were first noticed in the Levant. The late Mr R.S. Newall found, in the case of a hemp-covered cable which had been down there only a few months, that the hemp had been destroyed by a species of “teredo.”

Again, in 1859, Mr C.W. Siemens (afterwards Sir William Siemens) found on a similar cable, which had been laid rather less than a year, millions of small shell-fish or snails, accompanied by tiny worms, which had entirely demolished the unsheathed hemp covering. Moreover, small circular holes were found pierced here and there in the gutta-percha.

Since then similar havoc has been wrought on several cables, on a small scale, in all parts of the world; notably in the Western basin of the Mediterranean, the English Channel, Irish Sea, Atlantic Ocean, and more especially on the coast of Brazil, in the Persian Gulf, East Indian Archipelago, etc., caused by the different species of ravaging insects above alluded to.

4. It is partly for a similar reason, moreover, that hemp is always adopted for the inner (as well as the outer) serving of these cables, since the rest of them have been taken over by the Post Office from the Submarine Telegraph Company.

5. This also has sometimes an economic advantage in the case of multiple-core cables ( like most of the above), by saving the individual taping of each core.

6. Thus it is not unusual in selecting a route for a coast cable to avoid a line of depth under about 500 fathoms, if possible. This is partly with a view to the cable being free from fish attacks, teredo bores, and vegetable and animal life generally; as well as to avoid strong currents and a rocky bottom.

7. There is said to be no limit to the depth at which other harmless forms of animal life exist.

8. Thus in the shoal waters on the West Coast of Africa, no doubt, the teredo would be found hut for the existence of cold bottom currents.

9. Indeed Mr Willoughby Smith once advanced the idea that they took a passage in the serving of the cable from the factory, and eventually turn into powder later, like the mite in a cheese.

10. “Cable Borers,” by G. E. Preece, Jour. Soc. Tel. Eng., vol. iv., p. 363.

11.Previously the protection of underground lines from such ravages had been sometimes effected by embedding them in cement.

12. Later on the bituminous silicated compound, which Sir Charles Bright patented in 1862 for the outer sheathing of cables, was partly intended to serve the same purpose. This was by the mixing of powdered silica with the pitch and tar; thus breaking and rendering useless the boring-tool of the teredo and such-like. Both systems have now been adopted for many years in the case of all cables exposed to these attacks—in depths under 100 fathoms, at any rate. These combined remedies have been found entirely effectual.

13. In early days, when metal tapes were applied spirally outside the serving as above, a packing of flexible cotton tape used to be interpolated longitudinally between the hemp serving and the brass ribbon. Messrs Siemens have used tapings of this description on some cables laid by them in the East.

14. It was found necessary to give up all ideas of protecting the serving, owing to the difficulty of applying the tape to it in an efficient manner.

15. This is so, largely owing to the necessity of keeping the bulk down as much as possible. Moreover, if the weight of the cable were materially increased by such a taping, the modulus of tension would be seriously reduced; though this would, as a rule, only apply to the case of fairly deep water.

16. Besides the metallic hardness defeating the borers’ intentions, it is supposed that what chemical action occurs is offensive to the taste of the ravagers.

17. No doubt the notion of the special suitability of this metal for the purpose first originated with its application to the bottoms of wooden ships. Here it was discovered that Muntz metal ensured such a permanently smooth surface that barnacles were unable to cling to it for any time, besides being so hard as to defeat all efforts on the part of borers. Moreover, a thin sheath of Muntz metal was found to be not only cheaper than copper as previously used, but, owing to the absence of chemical action, lasted as long as the ship itself, instead of having to be periodically renewed.

18. Both phosphor and silicium bronze are in reality hard-drawn copper, with 3 per cent, of tin added for producing still further hardness. They are so styled on account of the materials which are employed, in some form, as a flux, in the process of manufacture.

19. The main object of compounding the tape thuswise is to check any tendency in the direction of a galvanic battery being set up, with consequent chemical action, between the iron wires and the brass—the wet serving forming the electrolyte by absorption of sea-water or due to its previous saturation. By compounding, the tape becomes more or less waterproof, and therefore should prevent any electrolytic action such as would constitute a battery.

Were the above action allowed to take place, iron being electro-positive to brass, the sheathing wires would be gradually eaten away, though less so than if the metal tape was formed of copper—more highly electro-negative than any other metal—where the greatest potential difference would exist. It may be added that Muntz metal is at a slight advantage over ordinary brass in this respect again. The above compounding of the outer tape appears to effect its object fairly well, as there has never been absolute evidence of chemical action between the metal riband and the sheathing wires in a cable with brass sheathing to its core—even after many years’ trial. In the case of faulty insulation at any point—and especially if the salt-water has entered—the presence of brass tape would probably tend to accentuate it; but, on the other hand, it would very likely render its localisation more easy.

20. The tapings—especially the metal tape—have to be laid on very tightly, and in the process adopted to accomplish this the core is liable to become stretched beyond what it would do otherwise. This forms, in fact, one of the objections raised in early days to this method of protection, though not now considered serious in practice.

21. By another arrangement, due to the same gentleman, the metallic and cotton tapes are stuck together before applying them to the core. This latter plan, though experimented with at first, has never been turned to practical account.

22. In order to secure a still more regular bed, the Telegraph Construction Company now very often dispense with the inner tape bedding and apply the metal tape direct on to the core, followed by the outside tape as a preservative protection to the metallic riband.

23. Water is played on whilst being applied to the core.

24. This machine has only recently been designed and set up for use by Messrs Washington and Co., Engineers, of Sowerby Bridge, Yorkshire. A full detailed description will be found of it in Electricity for 21st February 1896.

25. It need scarcely be remarked that where only two tapes are applied, a similar machine would be used, but with two taping “heads” instead of three.

26. In multiple-conductor cables, metal taping outside each core has been adopted by the Engineer of the Post Office (Mr W.H. Preece, C.B., F.R.S.) in the 1896 Bacton-Borkum cable, with a view to reducing to a minimum the effect of static induction from wire to wire, and also in order to neutralise—by thus introducing magnetic induction—the electro-static capacity of each circuit. Previously the multiple cables of this department had only a single brass tape outside the serving of the wormed cores. Substantial benefit in working efficiency is said to have been derived from the above modification.

27. Since the above lines were written, the author learns that Messrs Hooper adopted a lead tubing for protecting the india-rubber core of the “Cuba-Submarine” Company’s Cienfuegos-Batabano cable in 1895.

28. Probably lead could never be applied as a riband taping, though a provisional specification (No. 855 of 1877), drawn up in the name of Colonel T.G. Glover, speaks of a spiral of lead-foil (with the same object in view) covered with a hemp serving saturated in castor oil. In any case, a seamless tube is a much more complete, though more costly, protection. However—no matter how thinly it is capable of being applied—it can scarcely fail to be less flexible than the ordinary so-called brass tape.

For shore ends and in land cables the consideration of weight does not, of course, enter into the question practically; there is, therefore, no objection to lead on this account.

29. There used at one time to be some difficulty and risk in applying a lead tubing to gutta-percha core, on account of the heat required for the molten lead, though Mr John Chatterton had a patent for effecting this many years ago. Since then, however, means have been devised for applying it cold and direct from a hydraulic press, and this is carried out on a large scale—more especially to india-rubber electric light cables.

30. The compound here would tend to act as a preservative against the decay to which lead is subject in certain soils.

31. However, as has already been pointed out, any admixtures of rubber in close contact with gutta-percha are highly objectionable, even if they would properly adhere.

32. Possibly the above oil was found to act as a solvent of gutta-percha.

33. This clause was on much the same lines as Provisional Specification No. 3,938 of 1868, standing in the name of Mr Henry Clifford.

Last revised: 3 October, 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:

—Bill Burns, publisher and webmaster: