Chapter 4

What at first was as much a daring adventure as a business enterprise has now taken its place as a task no more out of the common than building a steamship, or rearing a cantilever bridge. Given its price, which will include too moderate a profit to betray any expectation of failure, and a responsible firm will contract to lay a cable across the Pacific itself. In the Atlantic lines the uniformly low temperature of the ocean floor (about 4° C.), and the great pressure of the superincumbent sea, co-operate in effecting an enormous enhancement both in the insulation and in the carrying capacity of the wire. As an example of recent work in ocean telegraphy let us glance at the cable laid in 1894, by the Commercial Cable Company of New York. It unites Cape Canso, on the northeastern coast of Nova Scotia, to Waterville, on the southwestern coast of Ireland. The central portion of this cable much resembles that of its predecessor in 1866. Its exterior armour of steel wires is much more elaborate. The first part of Fig. 59 shows the details of manufacture: the central copper core is covered with gutta-percha, then with jute, upon which the steel wires are spirally wound, followed by a strong outer covering. For the greatest depths at sea, type _A_ is employed for a total length of 1,420 miles; the diameter of this part of the cable is seven-eighths of an inch. As the water lessens in depth the sheathing increases in size until the diameter of the cable becomes one and one-sixteenth inches for 152 miles, as type _B_. The cable now undergoes a third enlargement, and then its fourth and last proportions are presented as it touches the shore, for a distance of one and three-quarter miles, where type _C_ has a diameter of two and one-half inches. The weights of material used in this cable are: copper wire, 495 tons; gutta-percha, 315 tons; jute yarn, 575 tons; steel wire, 3,000 tons; compound and tar, 1,075 tons; total, 5,460 tons. The telegraph-ship _Faraday_, specially designed for cable-laying, accomplished the work without mishap.

Electrical science owes much to the Atlantic cables, in particular to the first of them. At the very beginning it banished the idea that electricity as it passes through metallic conductors has anything like its velocity through free space. It was soon found, as Professor Mendenhall says, "that it is no more correct to assign a definite velocity to electricity than to a river. As the rate of flow of a river is determined by the character of its bed, its gradient, and other circumstances, so the velocity of an electric current is found to depend on the conditions under which the flow takes place."[2] Mile for mile the original Atlantic cable had twenty times the retarding effect of a good aerial line; the best recent cables reduce this figure by nearly one-half.

In an extreme form, this slowing down reminds us of the obstruction of light as it enters the atmosphere of the earth, of the further impediment which the rays encounter if they pass from the air into the sea. In the main the causes which hinder a pulse committed to a cable are two: induction, and the electrostatic capacity of the wire, that is, the capacity of the wire to take up a charge of its own, just as if it were the metal of a Leyden jar.

Let us first consider induction. As a current takes its way through the copper core it induces in its surroundings a second and opposing current. For this the remedy is one too costly to be applied. Were a cable manufactured in a double line, as in the best telephonic circuits, induction, with its retarding and quenching effects, would be neutralized. Here the steel wire armour which encircles the cable plays an unwelcome part. Induction is always proportioned to the conductivity of the mass in which it appears; as steel is an excellent conductor, the armour of an ocean cable, close as it is to the copper core, has induced in it a current much stronger, and therefore more retarding, than if the steel wire were absent.

A word now as to the second difficulty in working beneath the sea--that due to the absorbing power of the line itself. An Atlantic cable, like any other extended conductor, is virtually a long, cylindrical Leyden jar, the copper wire forming the inner coat, and its surroundings the outer coat. Before a signal can be received at the distant terminus the wire must first be charged. The effect is somewhat like transmitting a signal through water which fills a rubber tube; first of all the tube is distended, and its compression, or secondary effect, really transmits the impulse. A remedy for this is a condenser formed of alternate sheets of tin-foil and mica, _C_, connected with the battery, _B_, so as to balance the electric charge of the cable wire (Fig. 60). In the first Atlantic line an impulse demanded one-seventh of a second for its journey. This was reduced when Mr. Whitehouse made the capital discovery that the speed of a signal is increased threefold when the wire is alternately connected with the zinc and copper poles of the battery. Sir William Thomson ascertained that these successive pulses are most effective when of proportioned lengths. He accordingly devised an automatic transmitter which draws a duly perforated slip of paper under a metallic spring connected with the cable. To-day 250 to 300 letters are sent per minute instead of fifteen, as at first.

In many ways a deep-sea cable exaggerates in an instructive manner the phenomena of telegraphy over long aerial lines. The two ends of a cable may be in regions of widely diverse electrical potential, or pressure, just as the readings of the barometer at these two places may differ much. If a copper wire were allowed to offer itself as a gateless conductor it would equalize these variations of potential with serious injury to itself. Accordingly the rule is adopted of working the cable not directly, as if it were a land line, but indirectly through condensers. As the throb sent through such apparatus is but momentary, the cable is in no risk from the strong currents which would course through it if it were permitted to be an open channel.

A serious error in working the first cables was in supposing that they required strong currents as in land lines of considerable length. The very reverse is the fact. Mr. Charles Bright, in _Submarine Telegraphs_, says:

"Mr. Latimer Clark had the conductor of the 1865 and 1866 lines joined together at the Newfoundland end, thus forming an unbroken length of 3,700 miles in circuit. He then placed some sulphuric acid in a very small silver thimble, with a fragment of zinc weighing a grain or two. By this primitive agency he succeeded in conveying signals through twice the breadth of the Atlantic Ocean in little more than a second of time after making contact. The deflections were not of a dubious character, but full and strong, from which it was manifest than an even smaller battery would suffice to produce somewhat similar effects."

At first in operating the Atlantic cable a mirror galvanometer was employed as a receiver. The principle of this receiver has often been illustrated by a mischievous boy as, with a slight and almost imperceptible motion of his hand, he has used a bit of looking-glass to dart a ray of reflected sunlight across a wide street or a large room. On the same plan, the extremely minute motion of a galvanometer, as it receives the successive pulsations of a message, is magnified by a weightless lever of light so that the words are easily read by an operator (Fig. 61). This beautiful invention comes from the hands of Sir William Thomson [now Lord Kelvin], who, more than any other electrician, has made ocean telegraphy an established success.

In another receiver, also of his design, the siphon recorder, he began by taking advantage of the fact, observed long before by Bose, that a charge of electricity stimulates the flow of a liquid. In its original form the ink-well into which the siphon dipped was insulated and charged to a high voltage by an influence-machine; the ink, powerfully repelled, was spurted from the siphon point to a moving strip of paper beneath (Fig. 62). It was afterward found better to use a delicate mechanical shaker which throws out the ink in minute drops as the cable current gently sways the siphon back and forth (Fig. 63).

Minute as the current is which suffices for cable telegraphy, it is essential that the metallic circuit be not only unbroken, but unimpaired throughout. No part of his duty has more severely taxed the resources of the electrician than to discover the breaks and leaks in his ocean cables. One of his methods is to pour electricity as it were, into a broken wire, much as if it were a narrow tube, and estimate the length of the wire (and consequently the distance from shore to the defect or break) by the quantity of current required to fill it.

FOOTNOTES:

[1] Henry M. Field, "History of the Atlantic Telegraph." New York: Scribner, 1866.

[2] "A Century of Electricity." Boston, Houghton, Mifflin & Co., 1887.

BELL'S TELEPHONIC RESEARCHES

[From "Bell's Electric Speaking Telephones," by George B. Prescott, copyright by D Appleton & Co., New York, 1884]

In a lecture delivered before the Society of Telegraph Engineers, in London, October 31, 1877, Prof. A. G. Bell gave a history of his researches in telephony, together with the experiments that he was led to undertake in his endeavours to produce a practical system of multiple telegraphy, and to realize also the transmission of articulate speech. After the usual introduction, Professor Bell said in part:

It is to-night my pleasure, as well as duty, to give you some account of the telephonic researches in which I have been so long engaged. Many years ago my attention was directed to the mechanism of speech by my father, Alexander Melville Bell, of Edinburgh, who has made a life-long study of the subject. Many of those present may recollect the invention by my father of a means of representing, in a wonderfully accurate manner, the positions of the vocal organs in forming sounds. Together we carried on quite a number of experiments, seeking to discover the correct mechanism of English and foreign elements of speech, and I remember especially an investigation in which we were engaged concerning the musical relations of vowel sounds. When vocal sounds are whispered, each vowel seems to possess a particular pitch of its own, and by whispering certain vowels in succession a musical scale can be distinctly perceived. Our aim was to determine the natural pitch of each vowel; but unexpected difficulties made their appearance, for many of the vowels seemed to possess a double pitch--one due, probably, to the resonance of the air in the mouth, and the other to the resonance of the air contained in the cavity behind the tongue, comprehending the pharynx and larynx.

I hit upon an expedient for determining the pitch, which, at that time, I thought to be original with myself. It consisted in vibrating a tuning fork in front of the mouth while the positions of the vocal organs for the various vowels were silently taken. It was found that each vowel position caused the reinforcement of some particular fork or forks.

I wrote an account of these researches to Mr. Alex. J. Ellis, of London. In reply, he informed me that the experiments related had already been performed by Helmholtz, and in a much more perfect manner than I had done. Indeed, he said that Helmholtz had not only analyzed the vowel sounds into their constituent musical elements, but had actually performed the synthesis of them.

He had succeeded in producing, artificially, certain of the vowel sounds by causing tuning forks of different pitch to vibrate simultaneously by means of an electric current. Mr. Ellis was kind enough to grant me an interview for the purpose of explaining the apparatus employed by Helmholtz in producing these extraordinary effects, and I spent the greater part of a delightful day with him in investigating the subject. At that time, however, I was too slightly acquainted with the laws of electricity fully to understand the explanations given; but the interview had the effect of arousing my interest in the subjects of sound and electricity, and I did not rest until I had obtained possession of a copy of Helmholtz's great work "The Theory of Tone," and had attempted, in a crude and imperfect manner, it is true, to reproduce his results. While reflecting upon the possibilities of the production of sound by electrical means, it struck me that the principle of vibrating a tuning fork by the intermittent attraction of an electro-magnet might be applied to the electrical production of music.

I imagined to myself a series of tuning forks of different pitches, arranged to vibrate automatically in the manner shown by Helmholtz--each fork interrupting, at every vibration, a voltaic current--and the thought occurred, Why should not the depression of a key like that of a piano direct the interrupted current from any one of these forks, through a telegraph wire, to a series of electro-magnets operating the strings of a piano or other musical instrument, in which case a person might play the tuning fork piano in one place and the music be audible from the electro-magnetic piano in a distant city.

The more I reflected upon this arrangement the more feasible did it seem to me; indeed, I saw no reason why the depression of a number of keys at the tuning fork end of the circuit should not be followed by the audible production of a full chord from the piano in the distant city, each tuning fork affecting at the receiving end that string of the piano with which it was in unison. At this time the interest which I felt in electricity led me to study the various systems of telegraphy in use in this country and in America. I was much struck with the simplicity of the Morse alphabet, and with the fact that it could be read by sound. Instead of having the dots and dashes recorded on paper, the operators were in the habit of observing the duration of the click of the instruments, and in this way were enabled to distinguish by ear the various signals.

It struck me that in a similar manner the duration of a musical note might be made to represent the dot or dash of the telegraph code, so that a person might operate one of the keys of the tuning fork piano referred to above, and the duration of the sound proceeding from the corresponding string of the distant piano be observed by an operator stationed there. It seemed to me that in this way a number of distinct telegraph messages might be sent simultaneously from the tuning fork piano to the other end of the circuit by operators, each manipulating a different key of the instrument. These messages would be read by operators stationed at the distant piano, each receiving operator listening for signals for a certain definite pitch, and ignoring all others. In this way could be accomplished the simultaneous transmission of a number of telegraphic messages along a single wire, the number being limited only by the delicacy of the listener's ear. The idea of increasing the carrying power of a telegraph wire in this way took complete possession of my mind, and it was this practical end that I had in view when I commenced my researches in electric telephony.

In the progress of science it is universally found that complexity leads to simplicity, and in narrating the history of scientific research it is often advisable to begin at the end.

In glancing back over my own researches, I find it necessary to designate, by distinct names, a variety of electrical currents by means of which sounds can be produced, and I shall direct your attention to several distinct species of what may be termed telephonic currents of electricity. In order that the peculiarities of these currents may be clearly understood, I shall project upon the screen a graphical illustration of the different varieties.

The graphical method of representing electrical currents shown in Fig. 1 is the best means I have been able to devise of studying, in an accurate manner, the effects produced by various forms of telephonic apparatus, and it has led me to the conception of that peculiar species of telephonic current, here designated as _undulatory_, which has rendered feasible the artificial production of articulate speech by electrical means.

A horizontal line (_g g'_) is taken as the zero of current, and impulses of positive electricity are represented above the zero line, and negative impulses below it, or _vice versa_.

The vertical thickness of any electrical impulse (_b_ or _d_), measured from the zero line, indicates the intensity of the electrical current at the point observed; and the horizontal extension of the electric line (_b_ or _d_) indicates the duration of the impulse.

Nine varieties of telephonic currents may be distinguished, but it will only be necessary to show you six of these. The three primary varieties designated as intermittent, pulsatory and undulatory, are represented in lines 1, 2 and 3.

Sub-varieties of these can be distinguished as direct or reversed currents, according as the electrical impulses are all of one kind or are alternately positive and negative. Direct currents may still further be distinguished as positive or negative, according as the impulses are of one kind or of the other.

An intermittent current is characterized by the alternate presence and absence of electricity upon the circuit.

A pulsatory current results from sudden or instantaneous changes in the intensity of a continuous current; and

An undulatory current is a current of electricity, the intensity of which varies in a manner proportional to the velocity of the motion of a particle of air during the production of a sound: thus the curve representing graphically the undulatory current for a simple musical note is the curve expressive of a simple pendulous vibration--that is, a sinusoidal curve.

And here I may remark, that, although the conception of the undulatory current of electricity is entirely original with myself, methods of producing sound by means of intermittent and pulsatory currents have long been known. For instance, it was long since discovered that an electro-magnet gives forth a decided sound when it is suddenly magnetized or demagnetized. When the circuit upon which it is placed is rapidly made and broken, a succession of explosive noises proceeds from the magnet. These sounds produce upon the ear the effect of a musical note when the current is interrupted a sufficient number of times per second....

For several years my attention was almost exclusively directed to the production of an instrument for making and breaking a voltaic circuit with extreme rapidity, to take the place of the transmitting tuning fork used in Helmholtz's researches. Without going into details, I shall merely say that the great defects of this plan of multiple telegraphy were found to consist, first, in the fact that the receiving operators were required to possess a good musical ear in order to discriminate the signals; and secondly, that the signals could only pass in one direction along the line (so that two wires would be necessary in order to complete communication in both directions). The first objection was got over by employing the device which I term a "vibratory circuit breaker," whereby musical signals can be automatically recorded....

I have formerly stated that Helmholtz was enabled to produce vowel sounds artificially by combining musical tones of different pitches and intensities. His apparatus is shown in Fig. 2. Tuning forks of different pitch are placed between the poles of electro-magnets (_a1_, _a2_, &c.), and are kept in continuous vibration by the action of an intermittent current from the fork _b_. Resonators, 1, 2, 3, etc., are arranged so as to reinforce the sounds in a greater or less degree, according as the exterior orifices are enlarged or contracted.

Thus it will be seen that upon Helmholtz's plan the tuning forks themselves produce tones of uniform intensity, the loudness being varied by an external reinforcement; but it struck me that the same results would be obtained, and in a much more perfect manner, by causing the tuning forks themselves to vibrate with different degrees of amplitude. I therefore devised the apparatus shown in Fig. 3, which was my first form of articulating telephone. In this figure a harp of steel rods is employed, attached to the poles of a permanent magnet, N. S. When any one of the rods is thrown into vibration an undulatory current is produced in the coils of the electro-magnet E, and the electro-magnet E' attracts the rods of the harp H' with a varying force, throwing into vibration that rod which is in unison with that vibrating at the other end of the circuit. Not only so, but the amplitude of vibration in the one will determine the amplitude of vibration in the other, for the intensity of the induced current is determined by the amplitude of the inducing vibration, and the amplitude of the vibration at the receiving end depends upon the intensity of the attractive impulses. When we sing into a piano, certain of the strings of the instrument are set in vibration sympathetically by the action of the voice with different degrees of amplitude, and a sound, which is an approximation to the vowel uttered, is produced from the piano. Theory shows that, had the piano a very much larger number of strings to the octave, the vowel sounds would be perfectly reproduced. My idea of the action of the apparatus, shown in Fig. 3, was this: Utter a sound in the neighbourhood of the harp H, and certain of the rods would be thrown into vibration with different amplitudes. At the other end of the circuit the corresponding rods of the harp H would vibrate with their proper relations of force, and the _timbre_ [characteristic quality] of the sound would be reproduced. The expense of constructing such an apparatus as that shown in figure 3 deterred me from making the attempt, and I sought to simplify the apparatus before venturing to have it made.