Jan. 3, 1878] 



NATURE 



191 



brass cylinder supported on a shaft which is screw- 

 threaded, and turns in a nut for a bearing, so that when 

 the cyUnder is cauFed to revolve by the crank, C, it also 

 has a horizontal travel in front of the mouthpiece, A. It 

 will be clear that the point on the metal diaphragm must, 

 therefore, describe a spiral trace over the surface of the 

 cylinder. On the latter is cut a spiral groove of like 

 pitch to that on the shaft, and around the cylinder is 

 attached a strip of tinfoil. When sounds are uttered in 

 the mouth-piece, A, the diaphragm is caused to vibrate, 

 and the point thereon is caused to make contacts with the 

 tinfoil at the portion where the latter crosses the spiral 

 groove. Hence, the foil, not being there backed by the 

 solid metal of the cylinder, becomes indented, and these 

 indentations are necessarily an exact record of the sounds 

 which produced them. 



It might be said that at this point the machine has 

 already become a complete phonograph or sound writer, 

 but it yet remains to translate the remarks made. It 

 should be remembered that the Marey and Rosapelly, the 

 Scott or the Barlow apparatus, which we recently de- 

 scribed, proceed no further than this. Each has its own 

 system of caligraphy, and after it has inscribed its peculiar 

 sinuous lines, it is still necessary to decipher them. Per- 

 haps the best device of this kind ever contrived was the 

 preparation of the human ear made by Dr. Clarence J. 

 Blake, of Boston, for Prof. Bell, the inventor of the tele- 

 phone. This was simply the ear from an actual subject, 

 suitably mounted, and having attached to its drum a 

 straw, which made traces on a blackened rotating cylinder. 

 The difference in the traces of the sounds uttered in the 



Fig. 2. 



ear was veiy clearly shown. Now there is no doubt that 

 by practice and the aid of a magnifier, it would be possi- 

 ble to read phonetically Mr. Edison's record of dots and 

 dashes, but he saves us that trouble by literally making it 

 read itself. The distinction is the same as if, instead of 

 perusing a book ourselves we drop it info a machine, set 

 the latter in motion, and, behold ! the voice of the author 

 is heard repeating his own composition. 



The reading mechanism is nothing but another dia- 

 phragm held in the tube D on the opposite side of the 

 machine, and a point of metal which is held against the 

 tin foil on the cylinder by a delicate spring. It makes no 

 difference as to the vibrations produced, whether a nail 

 moves over a file or a file moves over a nail, and in the 

 present instance it is the file or indented foil strip which 

 moves, and the metal point is caused to vibrate as it is 

 affected by the passage of the indentations. The vibra- 

 tions, however, of this point must be precisely the same 

 as those of the other point which made the indentations, 

 and these vibrations, transmitted to a second membrane, 

 must cause the latter to vibrate similar to the first mem- 

 brane, and the result is a synthesis of the sounds which, 

 in the beginning, we saw, as it were, analysed. 



In order to exhibit to the reader the writing of the 

 machine which is thus automatically read, we have had a 

 cast of a portion of the indented foil made, and from this 



the dots and lines in Fig. 2 are printed in, of course, 

 absolute facsimile, excepting that they are level instead 

 of being raised above or sunk beneath the surface. This 

 is a part of the sentences, " How do you do ?" and " How 

 do you like the phonograph?" It is a little curious that 

 the machine pronounces its own name with especial 

 clearness. The crank handle shown in our perspective 

 illustration of the device does not rightly belong to it, and 

 was attached by Mr. Edison in order to facilitate its exhi- 

 bition to us. 



In order that the machine may be able exactly to 

 reproduce given sounds, it is necessary, first, that these 

 sounds should be analysed into vibrations, and these 

 registered accurately in the manner described ; and 

 second, that their reproduction should be accomplished 

 in the same period of time in which they were made, for 

 evidently this element of time is an important factor in 

 the quality and nature of the tones. A sound which is 

 composed of a certain number of vibrations per second 

 is an octave above a sound which registers only half that 

 number of vibrations in the same period. Consequently 

 if the cylinder be rotated at a given speed while register- 

 ing certain tones, it is necessary that it should be turned 

 at precisely that same speed while reproducing them, else 

 the tones will be expressed in entirely different notes of 

 the scale, higher or lower than the normal note as the 

 cylinder is turned faster or slower. To attain this result 

 there must be a way of driving the cylinder, while de- 

 livering the sound or speaking, at exactly the same rate 

 as it ran while the sounds were being recorded, and this 

 is perhaps best done by well-regulated clockwork. It 

 should be understood that the machine illustrated is but 

 an experimental form, and combines in itself two separate 

 devices — the phonograph or recording apparatus, which 

 produces the indented slip, and the receiving or talking 

 contrivance which reads it. Thus in use the first machine 

 would produce a slip, and this would for example be sent 

 by mail elsewhere, together in all cases with information 

 of the velocity of rotation of the cylinder. The recipient 

 would then set the cylinder of his reading apparatus to 

 rotate at precisely the same speed, and in this way he 

 would hear the tones as they were uttered. Differences 

 in velocity of rotation within moderate limits would by no 

 means render the machine's talking indistinguishable, but 

 it would have the curious effect of possibly converting 

 the high voice of a child into the deep bass of a man, or 

 vice versa. 



No matter how familiar a person may be with modern 

 machinery and its wonderful performances, or how clear 

 in his mind the principle underlying this strange device 

 may be, it is impossible to listen to the mechanical speech 

 without his experiencing the idea that his senses are 

 deceiving him. We have heard other talking machines. 

 The Faber apparatus, for example, is a larj^e affair, as 

 big as a parlour organ. It has a key-board, rubber larynx 

 and lips, and an immense amount of ingenious mechanism 

 which combines to produce something like articulation in 

 a single monotonous organ-note. But here is a little affair 

 of a few pieces of metal, set up roughly on an iron stand 

 about a foot square, that talks in such a way, that, even 

 if in its present imperfect form many words are not clearly 

 distinguishable, ihere can be no doubt but that the inflec- 

 tions are those of nothing else than the human voice. 



We have already pointed out the startling possibility of 

 the voices of the dead being reheard through this device, 

 and there is no doubt but that its capabilities are fully 

 equal to other results just as astonishing. When it be- 

 comes possible, as it doubtless will, to magnify the sound, 

 the voices of such singers as Parepa and Titiens will not 

 die with them, but will remain as long as the metal in 

 which they may be embodied will last. The witness in 

 court will find his own testimony repeated by machine, 

 confronting him on cross-examination— the testator will 

 repeat his last will and testament into the machine s4 



