But Helmholtz did not stop after a.n.a.lyzing sounds of so many kinds: he invented a method of synthesis, by which the sounds of any kind of an instrument could be imitated. A tuning-fork, when made to vibrate by an electric current, gives out a tone without harmonics or overtones. So if a series of forks with vibration periods equal to the numbers of the series of overtones given on p. 86 be so arranged that any of them may be made to vibrate at will, it is evident that the resulting compound tone would be comparable with that from an instrument having such overtones. Thus, if with a tuning-fork giving a fundamental C, other forks giving two, three, and four times the number of the fundamental were a.s.sociated, each one giving a simple tone, we should have for a resultant the tone of a flute, as shown on p. 91. If one, three, five, seven, and nine, were all sounded, the resulting tone would be that of the clarionet, and so on. This he actually accomplished, and now makers of physical apparatus advertise just such instruments.
Helmholtz also contrived a set of tuning-forks, which, when bowed, will give out the vowel sounds like the voice.
It was remarked upon p. 89 that it has generally been considered that age has a mellowing effect upon the sound of a violin. Once in possession of the facts concerning sound that have been alluded to on the preceding pages, it is easy to see how such an opinion should arise, and also the fallacy of it. It is proved conclusively that the ability to hear high sounds decreases as one grows older. As the violin gives a very great number of overtones, even up to the limits of audibility, it is plain that if such an instrument should not change in its quality of tone in the least degree, yet to a man who played upon it for a number of years it would seem to change by subtracting some of the higher overtones from the sound; that is, it would seem to become mellower.
There is no evidence that such a physical change takes place in the instrument. It is not here affirmed that no change does take place. It may be probable; but all the evidence we have is the opinions of individuals whose hearing we know does change; and this change is competent to modify the judgment as to the quality of the sound in the same direction. Before it can be affirmed that such a physical change does take place in the violin as to make a perceptible difference in the quality of its tone, it will be needful to determine accurately the number and intensity of the overtones at intervals during many years, and then to compare them. This has not yet been done.
FORM OF A COMPOUND SOUND-WAVE IN AIR.
Upon p. 63 is given a picture of the form of a simple sound-wave in air, which, as described, consists of two parts, a condensation and a rarefaction. All simple sound-waves have such a form; but when two or more sound-waves that stand in some simple ratio to each other, as do the sounds of musical instruments, are formed in air, the resulting wave is more or less complex in structure; and where there are many components, as there are where a number of different kinds of instruments are all sounding at once, it is well-nigh impossible to figure even approximately the form of such wave-combinations. It is generally given in treatises upon sound with ordinates representing the factors with their relative intensities. When the extremities of the ordinates are connected, there is drawn a curved line with regularly recurring loops. This cannot give a correct idea of the form of the wave, because the motion of a particle of air is not up and down like a floating body upon waving water, but it is forward and back, in the direction of the motion of the wave.
In Fig. 10 three simple sound-waves are thus represented at 1, 2, and 3, these having the wave-length 1, 2, and 3. In 4, the three are combined into one compound wave, and better show the form of a transverse section of such a sound-wave in the air. The organ-pipe called the princ.i.p.al gives out such a compound wave as is seen by referring to the table on p. 91. The second overtone, however, is quite weak in that pipe, which would so modify the form as to lessen somewhat the density at _b_, and increase it at _a_.
[Ill.u.s.tration: FIG. 10.]
In like manner the s.p.a.ce in the length of the fundamental sound, whatever it may be, is divided up into a number of minor condensations and rarefactions, which may strengthen each other, or so interfere as to change the position of both; as is seen in the figure at _b_, where the condensation due to wave 2 interferes with the rarefaction of 3.
CORRELATION.
HAVING treated at some length of the three factors involved in telephony,--namely, electricity, magnetism, and sound,--it remains to follow up the various steps that have led to the actual transmission of musical sounds and speech over an ordinary electric circuit.
It is stated upon p. 31, that, when a current of electricity is pa.s.sed through a coil of wire that surrounds a rod of soft iron, the latter is made a temporary magnet: it loses its magnetic property the instant that the current ceases. If the rod be of considerable size, say a foot or more in length, and half an inch or more in diameter, and the current be strong enough to make a powerful magnet of it, whenever the current from the battery is broken, the bar may be heard to give out a single _click_. This will happen as often as the current is broken. This is occasioned by a molecular movement which results in a _change_ of _length_ of the bar. When it is made a magnet, it elongates about 1/25000 of its length; and, when it loses its magnetism, it _suddenly_ regains its original length; and this change is accompanied with the sound. This sound was first noticed by Prof. C. G. Page of Salem, Ma.s.s., in 1837. If some means be devised for breaking such a circuit more than fifteen or sixteen times a second, we shall have a continuous sound with a pitch depending upon the number of clicks per second. Such a device was first invented by the same man, and was accomplished by fixing the armature of an electro-magnet to a spring which was in the circuit when the spring was pressing against a metallic k.n.o.b, at which time the current made the circuit in the coil of the electro-magnet. The magnet attracting the armature away from the b.u.t.ton broke the circuit, which of course destroyed the magnetism of the magnet, and allowed the spring to fly back against the b.u.t.ton, to complete the circuit and reproduce the same series of changes. The rapidity with which the current may be broken in this way is only limited by the strength of both spring and current. The greater the tension of the spring with a given current, the greater number of vibrations will it make.
[Ill.u.s.tration: FIG. 11.]
Suppose such an intermittent current to pa.s.s through the coil surrounding the soft iron rod, 256 times per second; then the rod would evidently give 256 clicks per second, which would have the pitch of C.
When these clicks are produced in the rod hold in the hand, the sound is hardly perceptible, being like that of a sounding tuning-fork when held thus. In order to strengthen it, it is necessary to place it on some resonant surface. It is customary to mount it upon an oblong box with one or two holes in its upper surface, inasmuch as such a form is found to give a louder response than any other, and is the shape usually given to aeolian harps. The accompanying cut shows the combination of battery B, the circuit-breaker, and the rod mounted upon the box. The wire W may evidently be of any length, the magnetized rod and box responding to the number of vibrations of the spring S, how long soever the circuit may be.
HELMHOLTZ" ELECTRIC INTERRUPTOR.
In some of Helmholtz" experiments, it was essential to maintain the vibrations of a tuning-fork for a considerable time. He effected this by placing a short electro-magnet between the p.r.o.ngs of the fork, and affixing a platinum point at the end of one p.r.o.ng in such a manner, that, as the p.r.o.ng descended in its vibration, the platinum point dipped into a small cup of mercury that completed the circuit. When the p.r.o.ng receded, it was of course withdrawn from the mercury, and the current was broken. As it is not possible for a tuning-fork to vibrate in more than one period, such an arrangement would evidently make and break the current as many times per second as the fork vibrated. When, therefore, such an interruptor is inserted in the circuit with the click-rod on its resonant box, the latter must give out just such a sound as the fork is giving. With such a device, it is possible to reproduce at almost any distance in a telegraphic circuit, a sound of a given pitch. It is therefore a true telephone.
REISS" TELEPHONE.
The ease with which membranes are thrown into vibrations corresponding in period to that of the sounding body has already been alluded to on p.
80; and several attempts have been made, at different times, to make membranes available in telephony. The first of these attempts was made by Philip Reiss of Friedrichsdorf, Germany, in 1861.
His apparatus consisted of a hollow box, with two apertures: one in front, in which was inserted a short tube for producing the sound in, and indicated by the arrow in the cut, Fig. 12; the other on the top.
This was covered with the membrane _m_,--a piece of bladder stretched tight over it. Upon the middle of the membrane, a thin piece of platinum was glued; and this piece of platinum was connected by a wire to a screw-cup from which another wire went to a battery.
[Ill.u.s.tration: FIG. 12.]
A platinum finger, S, rested upon the strip of platinum, but was made fast at one end to the screw-cup that connected with the other wire from the battery. Now, when a sound is made in the box, the membrane is made to vibrate powerfully: this makes the platinum strip to strike as often upon the platinum finger, and as often to bound away from it, thus making and breaking the current the same number of times per second. If, then, a person sings into this box while it is in circuit with the afore-mentioned click-rod and box, the latter will evidently change its pitch as often as it is changed by the voice. In this apparatus we have a telephone with which a melody may be reproduced at a distance with distinctness. But the sounds are not loud, and they have a tin-trumpet quality. If one reflects upon the possibilities of such a mechanism, and upon the conditions necessary to produce a sound of any given quality, as that of the voice or of a musical instrument as described in preceding pages, he will understand that it can reproduce only pitch. It might here be inferred that something more than a single pitch is transmitted if the sound is like that of a tin trumpet as stated: but the reason of this is that, whenever a current is pa.s.sing between two surfaces that can move only slightly on each other, there is always an irregularity in the conduction, so as to produce a kind of scratching sound; and it is this, combined with the other, the true pitch, that gives the character to the sound of this instrument.
Dr. Wright found that a sound of considerable intensity could be obtained by pa.s.sing the interrupted current through the primary wire of a small induction coil, and placing a conductor made of two sheets of silvered paper placed back to back in the secondary circuit. The silvered paper becomes rapidly charged and discharged, making a sound that can be heard over a large hall, and having the same pitch as the sending instrument.
GRAY"S TELEPHONES.
In 1873 Mr. Elisha Gray of Chicago discovered that if an induction coil be made to operate by the current from any automatic circuit-breaker, and one of the wires from the secondary circuit be held in the hand while the dry finger of the same hand is rubbed upon a sonorous metallic plate, the other wire being in connection with the plate, a musical sound would be given out by the plate, appearing to come from the point of contact of the finger with the plate. He therefore contrived a musical instrument with a range of two octaves, in which the reeds were made to vibrate by electro magnets, the current entering any one by depressing the appropriate key. This circuit is sent through the primary wire of an induction coil while one of the terminals of the secondary coil is connected with the thin sheet metal that forms one head of a shallow wooden drum about eight inches in diameter, which is so fixed as to be rotated like a pulley. The other terminal is held in the hand while one finger of the same hand rests upon the metallic surface. While the drum is turned with the other hand, the sounds given out have considerable intensity. The faster the drum is turned, the louder do the sounds become, though the pitch remains the same.
In this case, as in the case mentioned on p. 105, we have an electric current pa.s.sing between two surfaces that are moving upon each other; the contact not being uniform, the current is varying as well as intermittent.
Mr. Gray has also invented a musical telephone by means of which many musical sounds may be simultaneously transmitted and reproduced. The actual mechanism used is quite complex, and requires considerable familiarity with electrical science in order to understand it; but the fundamental principle involved is not difficult to one who has comprehended the preceding descriptions.
Suppose that we have a series of four steel reeds, each one fixed at one end to one pole of a short electro-magnet, while the other end is left free to vibrate over the other pole of the magnet and not quite touching it. Each of the reeds is to be tuned to a different pitch, say the 1, 3, 5, and 8 of the scale. These electro-magnets with their attached vibrators are to be attached each to a resonant box (see p. 93), which can respond to that particular number of vibrations per second. This is the receiving instrument. The sender consists of a like set of reeds tuned to the same pitch, which can be made to vibrate at will by pressing a key which sends the current of electricity through its electro-magnet, which makes and breaks the current. Imagine one of these keys to be pressed down so as to make the circuit complete: the sending instrument then has one of its reeds, let it be the 1 of the scale, set in vibration; the intermittent current traverses the whole line, going through all four of the receiving instruments. Now, we know from the study of the action of sounding bodies, that only one of the four receivers is competent to vibrate in consonance with this tone, and this one will respond; that is, the vibrations are truly sympathetic vibrations. If, instead of making the 1 of the scale in the sending-instrument, the 3 had been made, the current would have gone through all of the receiving instruments just the same as before, but only one of them could take up that vibratory movement: three of them would remain at rest, the 3 responding loudly. In like manner, any number of vibrating reeds in the sending instrument can make a corresponding number of reeds in the receiving instrument to vibrate, provided the latter be exactly tuned with the former. Each transmitter is connected with but a part of the battery, so that several tones may be transmitted at the same time. If the performer plays a piece of music in its various parts, every part will be reproduced: thus we have a compound or multiple telephone. This instrument has been used during the past winter to give concerts in cities when the performer was in a distant place.
It has also been used as a multiple telegraph; as many as eight operators sending messages simultaneously over the same wire,--four in each direction,--without the slightest interference.
BELL"S TELEPHONE.
Prof. A. Graham Bell of Boston independently discovered the same means for producing multiple effects over the same wire; but it appears he did not practically work it out as completely as did Mr. Gray. But while the latter was chiefly employed in perfecting the method as a telegraphic system, Prof. Bell had set before himself the more difficult problem of transmitting speech. This he has actually accomplished, as we have so often been reminded during the past year.
Thoroughly conversant with the acoustic researches of Helmholtz, and keeping in mind the complex form of the air vibrations produced by the human voice, he attempted to make these vibrations produce corresponding pulsations in an electric current in the manner a.n.a.logous to the electric interrupter.
Observing that membranes when properly stretched can vibrate to any kind of a sound, he sought to utilize them for this purpose. So did Reiss; but Reiss inserted the vibrating membrane into the circuit, and it was quite evident that such a plan would not answer, therefore the current must not be broken; but could an electric current be interfered with without breaking the connections?
The well-known re-actions of magnets upon electrical currents, first noted by Oersted, and fully developed by Faraday, gave the clew to the solution. A piece of iron should be made to vibrate by means of sound vibrations, so as to affect an electro-magnet and induce corresponding electrical pulsations.
FIRST FORM OF SPEAKING-TELEPHONE.
A membrane of gold-beater"s skin was tightly stretched over the end of a speaking-tube or funnel; on the middle of this membrane a piece of iron, N S, Fig. 13, was glued. In front of this piece of iron an electro-magnet M is so situated that its poles are opposite to it, but not quite touching it. One of the terminal wires of the electro-magnet goes to the battery B; the other goes to the receiving instrument R, which consists of a tubular electro-magnet, the coil being enclosed in a short tube of soft iron; the wire thence goes to the plate E", which is sunk in the earth. On the top of R, at P, is a rather loose, thin disk of iron, which acts as an armature to the electro-magnet below it.
[Ill.u.s.tration: FIG. 13.]
Supposing that all the parts are thus properly connected, the current of electricity from the battery makes both M and R magnetic; the electro-magnet M will inductively make the piece of iron N S, a magnet, with its poles unlike those of the inducing electro-magnet; and the two will mutually attract each other. If now this piece of iron N S be made to move toward M, a current of electricity will be induced in the coils, which will traverse the whole circuit. This induced electricity will consist of a single wave or pulse, and its force will depend upon the velocity of the approach of N S to M. A like pulse of electricity will be induced in the coils when N S is made to move away from M; but this current will move through the circuit in the opposite direction, so that whether the pulsation goes from M to R, or from R to M, depends simply upon the direction of the motion of N S.
The electricity thus generated in the wire by such vibratory movements varies in strength proportional to the movement of the armature; therefore the line wire between two places will be filled with electrical pulsation exactly like the aerial pulsations in structure.
Fig. 10, p. 98, may be used to ill.u.s.trate the condition of the wire through which the currents pa.s.s. The dark part may represent the strongest part of the wave, while the lighter part would show the weaker part of the wave. The chief difference would be, that electricity travels so fast, that what is there represented as one wave in air with a length of two feet would, in an electric wave, be more than fifty miles long.
These induced electric currents are but very transient (see p. 31); and their effect upon the receiver R is to either increase or decrease the power of the magnet there, as they are in one direction or the other, and consequently to vary the attractive power exercised upon the iron plate armature.
Let a simple sound be now made in the tube, consisting of 256 vibrations per second: the membrane carrying the iron will vibrate as many times, and so many pulses of induced electricity will be _imposed_ upon the constant current, which will each act upon the receiver, and cause so many vibrations of the armature upon it; and an ear held at P will hear the sound with the same pitch as that at the sending instrument. If two or more sound-waves act simultaneously upon the membrane, its motions must correspond with such combined motions; that is, its motions will be the resultant of all the sound-waves, and the corresponding pulsations in the current must reproduce at R the same effect. Now, when a person speaks in the tube, the membrane is thrown into vibrations more complex in structure than those just mentioned, differing only in number and intensity. The magnet will cause responses from even the minutest motion; and therefore an ear at R will hear what is said at the tube.
This was the instrument exhibited at the Centennial Exposition at Philadelphia, and concerning which Sir William Thompson said on his return to England, "This is the greatest by far of all the marvels of the electric telegraph."
The popular impression has been, concerning the telephone, that the _sound_ was in some way conveyed over the wire. It will be obvious to every one who may read this, that such is very far from being the case.
The fact is, it is a beautiful example of the convertibility of forces from one form to another. There is first the initial vibratory mechanical motion of the air, which is imparted to the membrane carrying the iron. This motion is converted into electricity in the coil of wire surrounding the electro-magnet, and at the receiving-end is first effective as magnetism, which is again converted into vibratory motion of the iron armature, which motion is imparted to the air, and so becomes again a sound-wave in air like the original one.
This was the first speaking-telephone that was ever constructed, so far as the writer is aware, but it was not a practicable instrument. Many sounds were not reproduced at all, and, according to the report of the judges at the Philadelphia Exposition, one needed to shout himself hoa.r.s.e in order that he might be heard at all.