Part 70

When the electric telegraph came into use and it was found possible to use it for communication of intelligence to great distances, it is not surprising that the further problem of transmitting by electricity, not signals merely, but audible speech, should be suggested. Perhaps the first scientific person who avowed a belief in the possibility of doing this, and even indicated the direction in which the solution of the problem was to be sought, was a Frenchman of science, M. Charles Bourseul. In 1854, he pointed out that sounds are caused by vibrations, and reach the ear by like vibrations of the intervening medium, and, although he could not say what took place in the modifications of the organs of speech by which syllables are produced, he inferred that these syllables could reach the ear only by vibrations of the medium, and that if these vibrations could be reproduced the syllables would be reproduced. He suggests that a man might speak near a flexible disc, which the vibrations of his voice would throw into oscillatory movements that could be caused to make and break a battery circuit, and that, at a distance, the currents might be arranged to produce the like vibrations in another disc. The weak point of this scheme was the want of any suggestion as to the mode in which this last effect was to be produced. Even when this part of the problem was solved in a few years afterwards, as we shall presently see, it was musical—and not articulate—sound that could be transmitted by an arrangement, using make and break contacts. The reader, who has understood what has been said of electrical currents, and also the account of the compounded vibrations in articulate sounds introduced into our section on the phonograph, should have little difficulty in seeing this must necessarily be the case, for the contacts could only give the succession of the vibrations by currents of equal intensity, and could not, like the yielding wax of the phonograph cylinder, correspond with their relative intensities. M. Bourseul pointed out advantages which would arise from the transmission of speech by electricity, such as simplicity of apparatus and facility in use—for, unlike the telegraph, no skilled operators would be needed—to signal messages, or time spent in spelling out the words letter by letter. He says that he had made some experiments, which promised a favourable result, but demanded time and patience, and that he is certain that, in a more or less distant future, speech will be transmitted by electricity, so that what is spoken in Vienna may be heard in Paris. One cannot help thinking that if M. Bourseul had but pursued his experiments a little longer, he would not improbably have achieved the invention of the speaking telephone, for which the world had to wait twenty years longer. As it is, we cannot but admire his scientific foresight and his confidence in the ultimate realization of his idea.

But before this came to pass, an intermediate stage was reached in the apparatus contrived by M. Reiss, a schoolmaster of Friedrichsdorf, who, in 1860, solved the problem of electrically transmitting musical tones. So far as concerned the reproduction of the sounds, this telephone was founded upon a discovery, made in 1837, by an American physicist, named Page, which was this: At the moment a bar of iron is magnetized, by sending a current through a coil surrounding it, as shown in Fig. 265, a slight but sharp click is heard. The transmitting apparatus was, in principle, Mr. Scott’s phono-autograph (described in the section on the phonograph), which had been invented in 1855. The tracing style of this was replaced in Reiss’ apparatus by a small disc of platinum, connected by a very light spring of the same metal with a binding-screw for the battery connection. Nearly in contact with the little disc was a platinum point, so arranged that the slightest oscillation of the membrane would bring them into actual contact and thus close the circuit. Worthy of remark is the very primitive nature of the materials with which Reiss made his first experimental apparatus. The receptacle for the voice was simply a large bung hollowed out into a conical cavity, and the membrane was supplied by the skin of a German sausage, while the clicking bar of the receiver was a stout knitting needle, surrounded by a coil of covered copper wire and stuck into the bridge of a violin, which, by acting as a sounding board, made the clicks produced in the needle distinctly audible. M. Reiss finally produced his telephone in the form shown in Fig. 302_a_, where _I_ is the receiver; _B_, the voltaic battery; _I I_, the receiver; _c c_ is a coil of insulated wire, surrounding a slender iron rod, mounted on the supports, _f f_, which rest on the sounding board, _g g_. The transmitter consists of the hollow box, _A_, provided with a trumpet-mouthed opening in one side and having at the top a circular piece cut out, across which is stretched a membrane with the little disc of platinum, _n_, fixed in its centre. When a person applying his mouth to A sings into the box, the membrane is thrown into vibrations corresponding with the notes, and at each vibration a contact is made and a click is emitted from the distant sounding box. The tones are concentrated by covering this box with the perforated lid. It was afterwards found that a trumpet mouth fitted into the receiver was still more effective. Reiss tried to use his arrangement for transmitting speech, but without success, although occasionally a syllable could be very indistinctly heard. An instrument, with springs so nicely adjusted that slight vibrations did not separate the platinum from actual contact, but merely caused change of pressure, has indeed been made to convey articulate sounds, although the arrangement was not essentially different from that of M. Reiss. This mode of action is, however, a different thing, and we shall presently see that very effective speech transmitters have been constructed by applying it in a more refined way. This musical telephone could give the pitch of the sounds in the song but not their quality (_timbre_), and the receiver added to the main system of vibration other sets that belonged to itself, the result being a shrill and by no means pleasing tone, recalling that of a penny trumpet. Messrs. C. and L. Wray afterwards effected some considerable improvements in M. Reiss’s telephone, with the object of intensifying the effects and producing better tones.

A further step towards the speaking telephone may be illustrated by an earlier invention of Mr. Graham Bell, a native of Scotland, who had settled in the United States. Mr. Bell’s inventions, it may be mentioned, were by no means the results of fortunate accidents or of unsought and spontaneous flashes of conception, but rather the outcome of long, patient and systematic studies. His father, Mr. Alexander Melville Bell, of Edinburgh, had assiduously cultivated acoustic science, and had in conjunction with his son, undertaken special researches into the mechanism of the organs of speech, the elements of articulate speech in different languages, and the musical components of vocal sounds. When Graham afterwards pursued these studies in the light of the fuller investigation carried out by Helmholtz, he was naturally led to the application of electricity to acoustic transmission. After some experiments in the production of vowel sounds by combinations of electric tuning forks, he invented a telephone for reproducing musical sounds at a distance, which was a great improvement on that of Reiss, and involved another principle, which indeed is the same as that utilized in his more mature invention of the speaking telephone. As a like explanation of the action would apply in both cases, the reader will find his advantage in following the observations we have to make on the earlier instrument. This consisted of what was virtually two sets of electric tuning forks, each set being acted upon by one electro-magnet. Fig. 302_b_ will suffice to show the general form of the arrangement. A plate of steel is bent twice at right angles longitudinally, and is magnetized so that any transverse slice of it would constitute an ordinary horse shoe magnet. This is seen endways in Fig. 302_b_ at M, and N. and S. will indicate the north and south poles respectively. To each limb of this broad magnet is attached a plate of steel, T, cut into teeth, just in the same way as the steel plate in a common musical box or mechanical piano, except that the teeth are not pointed. These are tuned to give severally in pairs the notes of the musical scale when thrown into vibration. Between the prongs of the series of tuning forks thus formed is an electro-magnet, L, made of a bar of soft iron, I, wound longitudinally by a coil, one end of which makes an earth connection at E and the other is connected by the wire, W W´, to complete the circuit through the coil of the distant apparatus. It will be observed that the receiving and transmitting instruments are exactly alike. Now, suppose one of these teeth is struck or otherwise thrown into vibration, the result will be, since the free ends of the teeth are magnetic poles, that alternating electric currents will be generated in the coil of the electro-magnet (see page 509), and these will flow through the entire circuit, including the coil of the distant instrument, where the magnetism generated will alternately attract and repel the polar extremities of the teeth in the steel plate. It will be understood, of course, that the fellow prong of the fork will vibrate also, and will simultaneously approach to and recede from the soft iron core, so that being of opposite polarity, the effect on the electro-magnet will be doubled. The action on the distant electro-magnet will be a rapid series of reversals of the polarities of the core, and hundreds of times in every second the ends of the steel teeth will be alternately attracted and repelled. But not all of these will thereby be thrown into vibration—only the one pair which were tuned into unison with the former can and will respond to that particular series of impulses, and the consequence will be that the same note will be emitted by the receiving instrument. If two or more notes of the transmitter be simultaneously thrown into vibration, the same notes will be heard from the receiver, for each series of currents will flow along the wire independently, just as if the other did not exist, and each will produce its particular effect on the transmitter. In this way an air played on the one instrument is heard also from the other, with all its accents and combinations. But more than this, if a tune be played on a musical instrument near the sender, or if a song be sung, the air will be reproduced by the distant receiver. The reason of this is that the steel tongues take up, or are thrown into movement by, the vibrations that have the same periodicity. The manner in which a vibratory body responds to impulses of its own periodicity may be easily shown by exposing the wires of a piano and raising the dampers, when, if a note be sung near the instrument, it will be found that a number of the wires respond, namely, those that are capable of vibrating synchronously with the constituent vibrations of the voice, for neither a voice nor a sounding wire gives forth one simple system of vibrations, the audible effect being due to the superposition or composition of several diverse elementary systems. With the same arrangement another experiment may be made, as an illustration of a matter important for our subject. Let the different vowels be sung to the piano-wire on the same note or pitch, and in the responses to each a difference of the quality of the sound will be noticed, although the piano will not distinctly give back the vowel itself. It would, however, do so if a number of its wires were strung with certain definite relations in pitch to that of the fundamental note and in unison with the voice components of the vowel sound.

It has been said above that two systems of electrical currents of different periodicity would flow along one wire independently of each other, but it should be explained that this takes place by a composition of the currents, for it is evident that at any given instant the wire can only be in one of three conditions, viz.: (1) with no current flowing; (2) with a current in the positive direction; (3) with a current in the negative direction. Such must always be the case, and, therefore, it should be clearly understood how this is consistent with the superposition of currents of different periodicities, a matter which the diagram, Fig. 302_c_, is intended to illustrate. Suppose the _flow of time_ to be represented by the dotted lines from _a_ to _b_, the whole length of which we may call 1/100th of a second, and that the current passing through the wire is represented in intensity and direction by the plain lines; the intensity by distance above or below the dotted line; the direction being positive where the plain line is above, and negative when it is below the dotted straight line, and of course no current at all occurs at the instant when the change of direction takes place. The line A will thus represent alternating currents, rising and sinking in intensity, and changing from one direction to the other, going through 600 regularly recurring phases in one second of time. Similarly, B may represent another series of currents, having here a periodicity of 500 in one second of time. These are here supposed to have greater intensity than the former. If the two currents are sent through one wire their effects are superposed, so that the actual electrical state of the wire would be represented by the curve C, which is compounded from the two others, and where it will be observed the rise and fall of the current, its maxima and minima, no longer recur at regular intervals within the space of the 1/100th of the second, the whole of that period being taken up by a less regular series of changes, the cycle being repeated only 100 times in the second. The same diagram might serve to illustrate the motions of, say, a particle of air or the drum of the ear in acoustic vibration, the distances above and below the straight line being taken to represent the displacements from the position of rest on one side and the other. If the sounds of an organ or piano consisted of only these primary vibrations, B would roughly[8] represent the movements of the wires, the air and the drum of the ear, when the note _si_{3}_ was sounded alone; A when the note _re_{4}_ was more faintly sounded alone, and then C, if these notes were sounded together, would correspond with the movements of the drum of the ear. The movements it actually makes when we hear speech, or even a single musical note, are, however, a thousand-fold more complex, for no musical instrument gives out a note with a single set of vibrations, the fundamental one being always accompanied by other sets diversely related to it, according to the class of instrument. In some cases, fifteen or sixteen sets of vibrations have been distinguished along with the fundamental note, without exhausting the possible number. Of a like order of complexity will be the currents which the wire of a speaking telephone must convey, and the difference between the undulatory nature of the currents in Bell’s musical telephone and any produced by mere make and break contacts, as in Reiss’ arrangement, will be obvious, and recognized as an important step towards the solution of the problem of transmitting speech. When Mr. Bell invented his instrument, he was seeking for a method of simultaneously transmitting by one wire several messages by audible _signs_ merely; and by the method used in his musical telephone this is practicable, for all that would be required would be pairs of transmitters and receivers, each adjusted to one single particular note. Another point that should be noted is that in the Bell musical telephone no battery is used, for the currents are those generated by magneto-electric induction, and the circuit through the wires and coils are completed by earth connections.

Footnote 8:

The lines A and B in the diagram have not harmonic ordinates.

[Music]

In passing from the invention of the musical to that of the speaking telephone, Mr. Bell passed from the more complex to the more simple instrument, for of all apparatus by which communication can be carried on at a distance, the Bell speaking telephone is one of the simplest. He had only to make its vibrating disc of Scott’s phono-autograph into a magnetized body, capable of producing currents in an electro-magnet coil in the same way as did the vibrating plates in his musical telephone. The Bell speaking telephone was publicly exhibited for the first time at Philadelphia, in 1876, and was shown the same year to the British Association by Sir William Thomson, who pronounced it the wonder of wonders. For the first time in England, the instrument in a still simpler form was exhibited by Mr. Preece, at the Plymouth meeting of the British Association in 1877, and of nearly the same construction as is still often used, although, as we shall presently see, for battery telephones the transmitting apparatus is now made of larger dimensions, of a different shape and on a different principle. We shall describe the simple form in which transmitter and receiver are identical, each consisting externally of a small cylindrical wooden or ebonite box, and with a handle three or four inches in length of the same material. Fig. 302_d_ is a section of the instrument where N S is a cylindrical steel magnet, on one end of which is wound the small coil B, made of fine silk covered copper wire, the extremities of which pass through the handle M at _f f_, and are connected by the binding screws _I I´_ with the line wire C C´. Close to the coil covered end of the magnet is a very thin diaphragm of iron, L L´, and when this is thrown into vibration by the voice speaking into the trumpet-mouth opening, R R´, its movements produce currents in the coil according to the principles that have already been explained, for it will be observed that the iron disc is magnetized by the inductive action of the permanent magnet N S. These currents passing through the coil of the receiving instrument raise or lower the intensity of the magnetic force in it, so that the distant disc reproduces the vibrations of the transmitter. Such is at least an obvious explanation of the action of this very simple arrangement; but from a number of experiments and observations that have been made with modifications of the instruments, it would appear that other and much more complex phenomena concur in producing the effects. It has indeed been suggested—and the idea is supported by numerous experiments—that, in these telephonic transmissions of speech, vibrations are concerned which are not at all of the mechanical kind we have been dealing with in these explanations, but are _molecular_.

The Bell telephone is used by speaking distinctly before the mouth-piece of the transmitter, while the listener at the other end of the line applies the mouth-piece of his instrument to his ear, and one wire is sufficient with good earth connections, although sometimes a second wire is employed to complete the circuit. It is also found advantageous to have two instruments in the circuit at each end, so that one may be held to the ear while the operator is speaking through the other. In this way, a rapid conversation can be carried on with the greatest ease, or again, an instrument may be held at each ear, by which arrangement the words are more distinctly heard. It is not necessary to shout, as this has no effect, but to speak with a clear intonation, and some voices are found to suit better than others. The vowel sounds are best transmitted, except that of the English _e_, which, with the letters _g_, _j_, _k_, and _q_, are always somewhat imperfectly transmitted. A song is very distinctly heard, both in the words and the air, and the voice of the person singing is readily recognized. Several instruments may be included in one circuit at different stations, so that half a dozen persons may take part in a conversation, and questions and answers may be understood even when crossing each other. If two distinct telephone circuits have their wires laid for a certain distance (two miles) near each other, say a foot or more apart, and without any connection whatever, listeners at the end of the one line will hear the conversation exchanged through the other line. Other forms of the instruments have been arranged, by which a large audience may hear sounds produced at a distance, as, for instance, when a cornet-à-piston was played in London, it was heard by thousands of people assembled in the Corn Exchange at Basingstoke.