[Reprinted from the Journal of the Franklin Institute, June, 1888.] 
The GRAMOPHONE: ETCHING the HUMAN VOICE. 
By Emile Berliner. 
paper read at the Stated Meeting of the Franklin Institute-, 
May 16, i888.'\ 
Jos. M. Wilson, President, in the chair. 
The President introduced Mr. Berliner, who spoke as follows: 
Members of the Franklin Institute, Ladies and Gentle- 
men:-The last year in the first century of the history of the 
United States was a remarkable one in the history of science. 
There appeared about that period something in the drift of 
scientific discussions, which, even to the mind of an observant 
amateur, foretold the coming of important events. 
The dispute of Religion versus Science was once more at its 
height; prominent daily papers commenced to issue weekly dis- 
cussions on scientific topics; series of scientific books in attrac- 
tive popular form were eagerly bought by the cultured classes; 
popular lectures on scientific subjects were sure of commanding 2 
enthusiastic audiences; the great works on evolution had just 
commenced to take root outside of the small circle of logical 
minds from which they had emanated, and which had fostered 
them. Scientific periodicals were expectantly scanned for new 
information, and the minds of both professionals and amateurs 
were on the qui wive. 
Add to this the general excitement prevailing on account of 
the forthcoming centennial celebration with its crowning event, 
so dear to this nation of inventors, the world's exhibition, and 
even those who did not at the time experience the effects of an 
atmosphere pregnant with scientific ozone, can, in their minds, 
conjure up the pulsating, swaying, and turbulent sea of scientific 
research of that period. Science evidently was in labor. 
The year 1876 came, and when the jubilee was at its very 
height, and when this great City of Philadelphia was one surging 
mass of patriots filling the air with the sounds of millions of 
shouts, a still small voice, hardly audible, and coming from a 
little disk of iron fastened to the centre of a membrane, whis- 
pered into the ear of one of the judges at the exhibition, and 
one of the greatest of living scientists, the tidings that a new 
revelation had descended upon mankind, and that the winged 
and fiery messenger of heaven's clouds had been harnessed to 
that delicate, tremorous, and yet so potent form of energy, called 
the Human Voice. 
The speaking telephone had been born. 
The stimulus which this event gave to science can best be 
measured by the enormous advance made since, especially in that 
now most prominent branch, electricity, and I will show further 
on how, immediately following it, our sister republic across the 
ocean answered the magic touch by the conception of another 
invention, the scope of which cannot to-day be measured yet, 
and which only just now is starting on its career of usefulness 
among the practical arts. 
In order to show the influence which these two inventions 
had upon each other, and how their respective development 
came about in parallel steps, permit me, before entering upon 
the new methods which I am to bring before you to-night, to 
pass in rapid review on the principal events in the history of the 3 
transmission of speech electrically, and of recording and repro- 
ducing the same mechanically. 
In 1854, Charles Bourseuil, with more than usual boldness, 
advanced the idea that two diaphragms, one operating an electric 
contact, and the other under the influence of an electro-magnet, 
might be employed for transmitting speech over telegraphic 
distances. "Speak against one diaphragm," he said, "and let 
each vibration break or make the electric contact, and the electric 
pulsations thereby produced will set the other diaphragm 
vibrating, and the latter ought then to reproduce the transmitted 
sound." Outside of the fallacy which his theory contained in 
the assumption of breaking the contact, instead of merely modi- 
fying the same, Bourseuil's paper, in speaking of the diaphragm, 
laid stress upon stating that " if one could be invented so movable 
and flexible as to answer to all the undulations of sound." He 
evidently desired extreme flexibility, and diaphragms constructed 
on that principle proved fatal to the efforts of many subsequent 
experimenters; even at first to Mr. Bell, who, like Bourseuil, 
borrowed the idea from the flexible tympanum membrani of the 
human ear, and who overlooked the important modifications 
which the vibrations undergo, before reaching the auditory nerve, 
by the series of muscular hinges in which the various bony acces- 
sories of the ear are mounted, and which act as elastic dampers 
against the tympanum membrani. 
Bourseuil's ideas were immediately reprinted from French 
journals in other countries, and among the first was a prominent 
German semi-weekly journal, printed in Frankfurt-on-the-Main, 
The Didaskalia, which, on September 28, 1854, under the heading 
" Electrical Telephony," published a leading article, giving a full 
account of Bourseuil's ingenious and wonderful conception. 
Frankfurt was then a city of about 60,000 inhabitants, and 
among other institutes of learning, it supported a Physical Society, 
which counted, at the time of this publication, among its active 
and most zealous members, an enthusiastic young teacher named 
Philip Reis, who, five years afterward, actually made an appa- 
ratus such as indicated by Bourseuil (who had since died without 
executing his idea), and which apparatus has since become known 
as the Reis telephone. 
I will not now enter upon any controversy as to the scope of 4 
this invention, regarding the possibility at the present day to 
transmit speech with the same. It may suffice to state that, when 
the news of the Bell telephone reached the learned men of Ger- 
many, some of the very first scientists in Berlin who knew all 
about the Reis apparatus, doubted the possibility of the perform- 
ance as represented by the American press. It is also now a 
matter of history, that in the late decision in favor of Mr. Bell, 
the United States Supreme Court was unanimous so far as the 
Bourseuil-Reis apparatus was concerned. 
Fig. i. 
Elec, Wvrkl *V. K 
Scott's Phonautograph. 
While Bourseuil's conception was being digested by Reis, 
another invention, having also a membrane diaphragm as its 
motive principle, was patented in France in 1857. This was the 
phonautograph, by Leon Scott, which had for its purpose the 
recording of sound vibrations upon a cylinder rotated by hand and 
moved forward by a screw (Fig. 1). The cylinder was covered 
with paper, this was smoked over a flame, and a stylus attached 
to the centre of a diaphragm under the influence of words spoken 
into a large barrel-like'mouthpiece, would trace sound vibrations 5 
upon the smoky surface. Scott also employed an animal mem- 
brane for his diaphragm, and took pains, by means of an attach- 
ment called a sub-divider, to make the vibrations appear as large 
as possible. This sub-divider, however, became the prototype of 
the dampers in subsequent apparatus, like the Blake transmitter 
and the Edison phonograph. 
The next important event in electro-phonic and acoustic sci- 
ence was the publication by Helmholtz of his investigations in 
sound, and of Konig in the same line of research, but classical 
as these publications will forever remain, they for a time retarded 
the progress of apparatus for practical use, for the reason that 
they discouraged inventors by the mechanical complications 
which they apparently ascribed as indispensable to articulate 
soeech. In fact, the perusal of their work left a serious doubt in 
the mind of many a student, whether there was not something in 
articulate speech and its audibility by the human ear, beyond the 
grasp of the mechanical mind of man. 
These doubts were still increased by the attempts of Faber to 
construct a talking machine, after the system of the human organs 
of speech, a mass of intricate mechanism, levers, bellows, and 
pulleys, which gave an unearthly rendition of many words and 
sentences. 
But the Bell telephone came, and its greatness consisted not 
so much in the fact that it carried speech over hundreds of miles, 
but that it taught how simple a piece of apparatus could produce 
such perfect results, and that any diaphragm however thick, 
could be made to set up audible articulate vibrations. 
The effect of this lesson was immediate, for hardly had the 
new wonder become known when an astonishing chain of logic 
formed in the brain of a distant devotee to science. 
On the 30th day of April, 1877, Mr. Charles Cros deposited 
with the Secretary of the Academy of Sciences in Paris a sealed 
envelope, containing what in translation reads as follows: 
" Process of Recording and of Reproducing Audible 
Phenomena." 
" In general, my process consists in obtaining the tracing of 
the to-and-fro movements of a vibrating membrane, and the utili- 6 
zation of this tracing for reproducing the same to-and-fro move- 
ments, with their relative inherent durations and intensities in the 
same membrane, or in another adapted for furnishing the sounds 
and noises which resu!t from this series of movements. 
" We are, therefore, concerned with the transformation of an 
extremely delicate tracing, such as that obtained with a delicate 
stylus rubbing upon a surface blackened by a flame, to transform, 
I say, these tracings in relief or intaglio, in resisting material 
capable of guiding a moving body, which transmits these move- 
ments to the sonorous membrane. 
" A light stylus is connected with the centre of a vibrating 
membrane; it terminates in a point (metallic wire, the barb of a 
feather, etc.), which bears upon a surface blackened by a flame- 
This surface is a part of a disk to which is given a double move- 
ment of rotation and rectilinear progression. 
" If the membrane is at rest, the point will trace a simple spiral; 
if the membrane vibrates, the traced spiral will be undulating, and 
these undulations represent exactly all the to-and-fro movements 
of the membrane, with their times and intensities." 
Up to this point the apparatus as described would represent a 
modified Scott phonautograph, in which the cylinder is sub- 
stituted by a flat disk. Mr. Cros then continues: 
" By means of the photographic process which, in fact, is well 
known, this traced, transparent, undulatory spiral is converted 
into a line of similar dimensions, in intaglio or in relief, in resist- 
ing material like tempered steel, for instance. 
" This done, this resisting surface is, by means of a motor 
apparatus, made to turn and to progress rectilinearly with a 
velocity like that which was used in the registration. 
" If the reproduced tracing is in intaglio, a metallic point (and 
if it is in relief, a notched finger), held by a spring, bears upon the 
tracing at one end and is connected at the other end with the 
centre of the membrane adapted for sound reproduction. Under 
these conditions, this membrane is not any more acted upon by 
the vibrating air, but by the tracing controlling the pointed stylus 
by pulsations exactly like those to which the membrane was sub- 
jected in recording, both as to duration and intensity. 
" The spiral trace represents the successive equal periods by 
its increasing and decreasing length. There is nothing inconve- 7 
nient in this if only the outer portion of the rotating circle is 
used, and if the spirals are close together, except that the central 
part of the disk is lost. 
" In all cases, however, a helical tracing upon a cylinder is 
much to be preferred, and I am actually engaged in finding a 
practical embodiment of this." 
This paper was only read in open session at the Academy on 
December 3, 1877, in the meantime Mr. T. A. Edison appeared 
with the phonograph. 
From what we can learn by published reports, Mr. Edison, 
some time in the latter part of September in the same year, was at 
work on an automatic telephone, by which he intended to impress 
a telephone message upon a strip of tin-foil, and let the indenta- 
tions thereby produced act upon a variable resistance, such as a 
lampblack button, and thereby transmit the message over the 
wire. While one day at work on this, so the report runs, he, 
perchance, slipped the previously indented slip under the record- 
ing stylus which, as in the Scott phonautograph, was connected 
to the centre of a diaphragm, and then and there occurred the 
first actual reproduction by mechanical means of words registered 
before. 
The phonograph became then, at once, an accomplished fact, 
for to such an experienced inventor it must have taken but a 
moment to mentally cover the cylinder of a Scott phonautograph 
with tin-foil and to indent the same at right angles to the surface 
of the cylinder. 
Everybody remembers the sensation which the invention pro- 
duced, and the prognostications which were advanced for it by 
the scientific press showed that the principle of the apparatus was 
considered to contain the germ of an ultimate achievement of the 
most accurate results. 
In this respect, as well as in others, there are striking resem- 
blances in the history of the two inventions with which I am 
dealing. 
In both, the original idea emanated from Frenchmen, and both 
described one process of transmitting, and a different process of 
reproducing speech. In the Bourseuil telephone there was a con- 
tact transmitter and an eloctro-magnet receiver; in the Cros 8 
phonograph, a written record and an engraved reproducing 
groove. 
In both inventions the first realization occurred in the United 
States, and was effected with apparatus representing only the 
reproducer of the original conception. In the speaking telephone, 
the reproducing electro-magnet of Bourseuil became also the 
transmitter of Bell, and in the phonograph, the reproducing 
groove and stylus of Cros became also the record of Edison. 
Both the Bell and Edison apparatus were accepted for a time as 
containing the best mechanical and philosophical principle for 
the highest attainable results. In both, the aim at the beginning 
was to produce loud sounds, and both eventually contented them- 
selves with a much fainter voice, which then became more dis- 
tinct in articulation. Finally, in both inventions, the original 
transmitter was subsequently resurrected, and found to contain a 
pointer toward a superior principle as a transmitter and recorder, 
and it only remains now to use a Scott phonautographic record 
direct for reproduction in order to complete a parallel with the 
fact that a contact transmitter can also be used as a telephonic 
reproducer. 
In making these parallels, however, I am aware of the fact 
that Cros had a better idea of a talking machine than Bourseuil 
had of a speaking telephone. 
The paper of Mr. Cros, which can be found on page 1082, 
vol. 85, of the Comptes Rendus of 1877, appears to have been 
consigned immediately to obscurity. When ten years later, I 
filed my patent application for the gramophone, not even the 
Examiners at the Patent Office knew anything of Mr. Cros, and 
when I mentioned his name in the first publication of the 
"gramophone," even those best informed on the subject were 
surprised. Nevertheless, I considered it a duty to my friends to 
make the following statement to the Editors of The Electrical 
World, which they published simultaneously with the " gramo- 
phone," on November 12, 1887. I said: 
"On August 30, of this year, which was three months after 
the filing of my application for a patent, while in the office of 
my counsel, Mr. Joseph Lyons, I happened to look through a 
German scientific book in his possession, and reading up about 
the phonograph, I came across a remark stating that on April 30, 9 
l8//, one, Chas. Cros, deposited at the French Academy of 
Science a sealed paper which, when opened and read at a subse- 
quent session during that year, was found to contain a description 
of the author's idea that a photo-engraved phonautographic 
record, either in relief or intaglio, might be utilized * for reacting 
through a stylus on a diaphragm, and by this reaction ought to 
reproduce the original sound.' 
" Surprised as I was at this discovery, I requested Mr. Lyons 
to find out through his friends in Paris whether and to what 
extent Mr. Cros had ever carried his idea into practice, and an 
answer has since come to the effect that Mr. Cros never put his 
idea into practical operation. 
" Whether he was taken aback by the eclat which the phono- 
graph produced soon afterward; whether he became discouraged 
at the practical difficulties, of which I have found many at the 
outset of all my experiments; or whether he did not appreciate 
the peculiar advantage of the phonautogramic method-all this 
does not appear from the meagre accounts so far to hand. 
" But although, viewed in the light of equity, he had virtually 
abandoned his invention at the time when I independently and 
without knowledge of his prior idea took up the same subject, 
the fact remains that to Mr. Charles Cros belongs the honor of 
having first suggested the idea of, and feasible plan for, mechanically 
reproducing speech once uttered." 
As this statement has never been challenged since it was first 
made, 1 presume that it is substantially correct. 
If we should attempt to carry out strictly the ideas of Mr. 
Cros, we would find many obstacles to obtaining practical results, 
and while undoubtedly the correctness of the general principle 
could be proved, the effects would not be as good even as those 
obtained by the original phonograph. Even with the application 
of the various improvements which I originally introduced, the 
process requires great care, and while this would not have been 
an obstacle on account of the great advances made in photo- 
engraving, I have now abandoned the original process altogether, 
and have substituted one of great rapidity and simplicity. 
But to return to the phonograph, we find this apparatus re- 
mained in an unsatisfactory and unfinished condition for nearly 
nine years. 10 
Among those who believed that ultimately the phonograph 
could be turned to practical account, was the well-known original 
patron of the speaking telephone, Mr. Gardiner G. Hubbard, and 
being also financially interested in it, he, in 1883 or thereabouts, 
caused the Volta Laboratory Co., an association originally founded 
by Prof. Bell as a laboratory, from the funds of the Volta Prize 
awarded to him by the French government, to provide ample 
funds for the purpose of making an extensive series of experi- 
ments with the phonograph. 
Prominent among the scientists connected with the enterprise 
were Prof. Bell, Dr. Chichester A. Bell, and Mr. C. S. Tainter. 
After two years of ardent labors these gentlemen came to the 
conclusions: 
First. That the indenting process had to be abandoned and an 
engraving process be substituted-i. e., instead of pushing the 
record surface down with the stylus, as in the original phono- 
graph, it should rather be dug out or graven into. 
Second. That the best substance, answering also the various 
other requirements, was beeswax hardened by an admixture of 
paraffine, or other similar waxy substances. 
Third. That loud speaking was impracticable, and that the 
ordinary conversational tone gave better results, although reduc- 
ing the reproduction to the loudness merely of a good telephone 
message. 
In Patent No. 341, 214, of May 4, 1886, issued to Dr. Chichester 
A. Bell and Mr. C. S. Tainter, the following claims, among others, 
were granted : 
" The method of forming a record of sounds by impressing 
sonorous vibrations upon a style, and thereby cutting in a solid 
body the record corresponding in form to the sound waves, in 
contradistinction to the formation of sound records by indenting 
a foil with a vibratory style, etc. 
" 3. The vibratory cutting style of a sound recorder; sub- 
tantially as described. 
" 7. A sound record consisting of a tablet, or other solid body, 
having its surface cut or engraved with narrow lines of irregular 
and varied form, corresponding to sound waves substantially as 
described. 
" 9. The method of forming a sound or speech record, which 11 
consists in engraving or cutting the same in wax, or a wax-like 
composition ; substantially as described." 
As a final result of all their labors, there issued in the spring 
of 1887, the graphophone, the first really practical apparatus of 
the phonograph type, and which was exhibited to admiring 
crowds in Washington and elsewhere. 
To those who have never heard this instrument, I will repeat 
what I wrote about its performance in November, 1887, namely, 
that it appears to be the best instrument to take down business 
letters or dictations of any kind, in which the recognition matters 
little, so long as the words can be made out; also, that the re- 
produced sound is as loud as that of a good telephone message, 
but that the distortion produced by the engraving is sufficient to 
make the voice unrecognizable save to a strained imagination 
added to a previous knowledge of the author of the voice. The 
record ground of this machine is a thin pasteboard cylinder 
covered with wax. 
Soon after the graphophone became generally known, Mr. 
Edison, evidently encouraged by the results obtained in this instru- 
ment, took again to experimenting with the phonograph, and, 
after trying wax covered with tin-foil for indentation, he aban- 
doned that mode of recording, and also settled upon a cylinder 
of wax and the graving-out process, thus confirming the correct- 
ness of Bell and Tainter's conclusions, and the new Edison 
phonograph and the graphophone appear to be practically the 
same apparatus, differing only in form and motive power. 
I now come to the subject of the evening, the Gramophone. 
In my telephonic studies, I had become familiar with all the 
causes influencing the transmission and reproduction of the voice, 
and what had at all times struck me as forcibly as anything in 
telephonic phenomena, was the fact that the self-induction of long 
iron wires or of polarized electro-magnets acted so detrimentally 
upon the articulation. Electrical resistance alone would simply 
have weakened the sound, but self-induction meant retardation, 
and this distortion of the transmitted waves which varied in 
length and amplitude. To appreciate fully what an extremely 
small amount of energy ordinary speech possesses mechanically, 
let us consider a few well-known facts: 
A puff of air, not strong enough to extinguish a candle-flame, 12 
when blown across an empty bottle or into a whistle will pro- 
duce a sound which may be heard over a hundred feet away. 
The amount of electricity needed to operate audibly a magneto- 
telephone, is said to be less than one-millionth part of the elec- 
tricity of a standard Daniel cell. 
In considering such and other facts it became evident to me 
that if such delicate energy, subdivided into maybe several 
hundred waves, should indent or engrave itself into a solid body, 
it needed but very slight mechanical resistance to modify con- 
siderably the character of the sound vibrations. For what self- 
induction is to the telephone circuit, the variable resistance which 
impressible material offers to indentation or engraving at various 
depths is to the phonograph record sheet. Neither is proportional 
in direct ratio to the expended energy and must give cause, aside 
from a reduction in size of the sound characters, also to a distor- 
tion of the same. 
Your own Prof. Houston, in his learned remarks in the Journal 
of the P'ranklin Institute of January, 1888, says : 
"The difficulties just pointed out, it would seem, must exist 
in any instrument, however improved in its mechanical structure, 
if it make the record on the Phonogram at right angles to the 
surface thereof. Of course, if a substance was discovered for such 
a surface, that offered a resistance to indentation exactly propor- 
tional to the depth of such indentation, the difficulty would, to a 
great extent, be removed." 
All the experiments which were made with the phonograph 
and the graphophone, confirmed the correctness of all these 
assertions, for the louder it was necessary to speak when record- 
ing, the less distinct became the articulation of the reproduced 
sound. 
A change for the better was, therefore, to be obtained: 
First. By tracing the vibrations, as in the old phonautograph, 
parallel to the record sheet. 
Second. By reducing the resistance offered by the record 
medium to as near to nothing as possible. 
Both principles, although not emphasized, are contained in the 
Cros document; but for my part, I found that merely smoked 
surfaces were utterly impracticable, because, if sufficiently black 
for a photo-engraving, and with the extremely small sizes of 13 
waves obtained with records that are adaptable for the reproduc- 
tion of good articulate speech, the record lines were ragged, and, 
under a magnifying glass, looked like a set of parallel saws whose 
teeth would form a grating sound, which nearly drowned the 
articulation. 
I observed, however, in my experiments, that the grayish 
deposit of lamp-black which is obtained from the centre of a 
kerosene flame was more oily and gave a somewhat sharper line 
than the deep black deposit caused by smoking with the top of 
the flame, and this led me to the highly beneficial process of 
oiling the plate prior to smoking the same, either by applying 
printers' ink or artists' paint by means of a printers' roller or by 
brushing oil over it. The smoke would then amalgamate with 
the oil and forms a fatty ink of a rather dry consistency, which, 
when crossed by a stylus, shows, even under a microscope, a 
sharply cut transparent line. 
I still employ this process for small test plates and prepare 
them as follows: One part of paraffine oil is mixed with twenty 
parts of benzine or gasoline. This mixture is poured on and off 
a glass disk, when the benzine evaporates leaving an extremely 
thin layer of oil. This is held over a smoky flame and moved to 
and fro until the surface looks just dry. The application of artists' 
paint with a roller prior to smoking is still better. 
I also adopted for the gramophone a disk of glass as a support 
for the smoke deposit, traced the sound record from below so that 
the displaced lamp-black should fall down, varnished it after the 
tracing was done and used this disk as a negative without, there- 
fore, needing a camera or photographic chemicals outside of the 
chrome-gelatine or chrome-albumen used in developing a raised 
picture. I would refer, for a detailed account, to the already men- 
tioned issues of The Electrical World and the Journal of the 
Franklin Institute. 
The lesson of simplicity which the telephone was continu- 
ously preaching caused me at an early day to look for a simpler 
plan to attain my purpose, and in the specification originally filed 
by me I said: 
" This record (meaning the phonautogram) may then be 
engraved either mechanically, chemically, or photo-chemically." 
And although for a long time without much hope for success, the 14 
purely chemical process of direct etching haunted me continu- 
ously, and was repeatedly suggested by others. 
But it was easier suggested than carried out, because under the 
principles of the gramophone the etching ground was to offer 
practically no resistance to the stylus, and to make one which 
had no resistance mechanically, but did resist the etching fluid 
after the tracing was done, was the problem to be solved. 
You will readily see, that if we can cbver, for instance, a 
polished metal plate with a delicate etching ground, trace in this a 
phonautogram and then immerse the plate in an etching fluid, 
the lines will be eaten in and the result will be a groove of even 
depth such as is required for reproduction; such a process, of 
course, would be much more direct and quicker than the photo- 
engraving method. 
In nature provision seems to be made for all the wants of 
mankind, and confident in this belief, I kept on trying to find a 
trail which led to promising results, and I have the honor 
to-night, for the first time, to bring before you this latest 
achievement in the art of producing permanent sound records 
from which a reproduction can be obtained, if necessary, within 
fifteen or twenty minutes, and which can be accurately multiplied 
in any number, by the electrotype process. It may be termed, 
in short, the art of etching the human voice. 
The etching ground which I use is also a fatty ink, and one of 
the best I have found thus far is made by digesting pure yellow 
beeswax in cold gasoline or benzine. 
Benzine, in a cold state, will not dissolve all the elements of 
the wax, but only a small part, namely, that which combines with 
the yellow coloring principle, and the resultant and decanted 
extract is a clear solution of a golden hue, which gradually 
becomes bleached by exposure to light. The proportions which 
I use are one ounce of finely scraped wax to one pint of gasoline. 
The bottle containing the mixture must be repeatedly shaken, 
and, after the white residue has settled, the clear fluid is decanted 
or drawn off by a siphon. 
I then take a polished metal plate, generally zinc, and flow the 
fluid on and off, as if I would coat with collodion. The benzine 
will quickly evaporate, and there remains a very thin layer of 
wax, iridescent under reflected light, not solid as a coating pro- 15 
duced by immersion in a melted mass, but spongy or porous, and 
extremely sensitive to the lightest touch. 
Partly on account of the too great sensitiveness of a single 
film, and also as an additional protection against the action of 
the acids employed in the subsequent etching, I may apply a 
second coating of the solution, and this double coat I find to 
answer all requirements. 
The protection which this porous or spongy wax affords from 
the acid, is mostly due to the fact that watery solutions assume 
the spherical state on the film, while at the lines where the wax 
is disturbed the acid enters freely, and attacks the metal below. 
A difficulty, which only a short time ago appeared insurmount- 
able. was the accumulation at the point of the stylus, while tracing 
the sound record, of filamentary particles of dust which exist in 
the wax solution, and which being ever present in ordinary rooms, 
settle down and adhere to the film. These dust particles are so 
fine that they cannot, as a rule, be detected by the most searching 
inspection of the prepared plate; but they become very conspicu- 
ous, and a very serious source of annoyance when a long record 
is being made. 
It must be borne in mind that the contact which the tracing 
stylus makes with the record surface, is obtained by the elastic 
pressure from a piece of hair-spring backed by a narrow blade of 
writing paper, and which pressure amounts to about five grains. 
Therefore, as this stylus passes through the fatty ink or other 
ground, and traces the fine undulatory line, the dust particles, as 
well as small portions of the displaced ink or wax adhere to and 
accumulate at the point of the stylus and are dragged along, and 
the record thereby becomes blurred and indistinct. 
I have discovered an effective means for overcoming this diffi- 
culty, and it consists in applying to the record surface a fluid that 
slightly adheres to the etching ground, and keeps it wet while the 
record is being made. I have found commercial alcohol to be 
very effective for this purpose, and it is used by pouring it over 
the platq just before the sound record is made. The alcohol, of 
course, immediately commences to evaporate, but not rapidly 
enough to disappear entirely before the record is finished, and 
there is no difficulty in adding more alcohol while the plate 
revolves. Under this condition, the point of the stylus remains 16 
perfectly clean, and it seems as if the dust particles had not been 
present at all. 
The theory by which I explain this result is, that the alcohol, so 
to speak, lubricates both the surface and the stylus, and prevents 
the adhesion of the filaments to the latter. At any rate, the 
application is highly beneficial, and the resulting line is so sharp 
and fine that it has to be widened in the subsequent etching pro- 
cess, in order to permit the acid to bite at sufficient depth. It 
can also be proved that the resistance of the wax film is decreased 
by the presence of the alcohol, but when this has evaporated the 
wax film appears to be in precisely the same condition as before, 
even showing again the iridescent colors which disappeared on 
the application of the alcohol. 
The film of wax being so thin, it is almost transparent, and if 
th erecord was made on this it could barely be detected. As, 
however, it is sometimes desirable to examine the record prior 
to etching the same, I can smoke the etching ground slightly by 
holding it high above burning camphor, so as to prevent a 
heating and melting of the spongy wax, and the alcohol poured 
afterwards over this smoked surface does not seem to wash off 
any of the soot particles. 
We now come to the important process of etching the record. 
Etching is done on steel, copper, or brass with nitric acid, per- 
chloride of iron, or with a mixture of muriatic acid and chlorate 
of potash known as Dutch mordant. In modern photo-engraving 
nearly all the etching is done on zinc by means of diluted nitric 
acid, and these materials are preferred on account of their being 
cheaper than any other, and zinc is a metal easily obtained with 
smooth and even facings. In etching, however, on zinc, it 
is necessary continually to brush away the hydrogen bubbles 
which form and adhere to the lines, and as the etching ground is 
usually of firm and solid material (like asphaltum, hard wax, 
pitch, or rosin mixtures) no harm results from the brushing neces- 
sary in order to obtain sharp edges along the lines. 
Desiring to avail myself of the advantages offered in zinc 
plates, I soon found that no etching fluid was known that 
would be to zinc what perchloride of iron was to copper- 
namely, etch cleanly and without the appearance of hydrogen 
bubbles. To apply the brushing to the delicate spongy wax film 17 
I employed was out of the question, as the first touch would wipe 
away the whole ground, and to permit the formation of hydrogen 
bubbles without brushing them away meant uneven and ragged 
lines and a distorted record. 
While studying this matter over it occurred to me to, so to 
speak, depolarize the zinc plate by adding to the acid, bichromate 
of soda which I thought might prove efficient, as it does in the 
galvanic battery, to prevent the appearance of the bubbles while 
etching the zinc. It took, however, a comparatively large quan- 
tity of the bichromate to answer my purpose, so much that I 
concluded that the mixture had all the conditions of a chromic 
acid, or at least of a mixture of chromic acid and nitrate of 
soda. When I thereupon substituted a solution of chromic acid 
pure and simple, I found this to be a most excellent etching 
fluid, and that is what I am now using-namely, a solution of 
one part by weight o.f dry chromic acid dissolved in three parts 
by weight of water. I use the commercial acid, such as can be 
obtained from Churchman & Co., of this city, at twenty-five cents 
a pound. Such a solution etches on zinc a sharp and clearly cut 
line, and no hydrogen appears during the etching. 
The back of the zinc plate had previously been painted with 
protecting varnish or molten beeswax, and within from fifteen to 
twenty minutes from the time of immersion in the chromic acid 
solution, and without disturbing it a cleanly cut groove of suffi- 
cient depth is obtained for reproduction. This groove may then 
be deepened in the ordinary way of rebiting by covering the 
facing of the plate with rosin dust, heat the same, and then im- 
merse in diluted nitric acid. Under these conditions the brush 
may be applied until the necessary depth is obtained, generally 
in about one to three minutes according to the strength of 
the etching fluid. I have used stronger solutions of chromic 
acid with no ill effects and a more rapid etching, and there 
seems to be a wide margin on this point, provided the plate 
is watched during the etching process. The lines very gradually 
widen in the course of the etching, but the upper edges of the 
grooves remain perfectly parallel and sharply defined. 
Before proceeding with a practical demonstration of the whole 
process, I will now describe the most important apparatus of the 
gramophone, the recorder. The translation of the movements of 18 
the diaphragm into the same movements at right angles, and with 
the extreme smallness of the motion and the liability of distorting 
them, adding to them, or detracting from their value in trans- 
lating them, requires greater care to guard against error than an 
uninitiated observer would suppose, and when we examine the com- 
plex and extremely delicate mechanism which nature has provided 
in the human ear for giving a correct translation of air vibrations 
into nervous vibrations, it behooves us to be careful in the applica- 
tion of every day mechanics. Free as the telephone is, compara- 
tively, from mechanical incumbrances, it is deficient in articulation 
of the consonants, and with the simplicity of mountingas required in 
the phonograph and graphophone, these difficulties of recording 
proper do not exist, and are shifted to the other portions of their 
construction and manipulation. In having attempted, therefore, 
to do justice to all sources of error I am not yet prepared to say 
that my present recording apparatus is constructed and adjusted 
to the greatest attainable correctness. Those who are familiar 
with the tediousness of original research will admit that a new 
subject of this kind cannot be solved in its entirety within the 
space of a few months, and what I bring before you to-night 
being the hasty results of a new machine finished but ten days 
Fig. 2. 
ago, should be measured rather by the possibilities it opens, than 
by the results so far attained, whatever merit you may accord to 
them. 19 
My impression, however, is that there is very little of lost or 
added motion in my present apparatus, and whatever imperfec- 
tions may exist must be looked for in the mode of reproducing 
the sound, rather than in the recorder (Fig. 2). 
K is the diaphragm box ; A is the centre portion of the dia- 
phragm ; B is a brass post screwed to the diaphragm and slotted 
above ; E is a piece of rubber tubing held in the slot and holding 
one end of the stylus C. This stylus is made of stiff metal and 
is pivotted by the steel pivots F F. D is a blade of writing paper 
reinforced by a piece of hairspring which extends, and forms the 
tracing point. G is a piece of rubber tubing around the stylus 
which dampens its musical vibrations ; 77 is a piece of felt damper 
between the diaphragm and the diaphragm box, which acts as a 
general damping device. 
The whole is mounted on a sliding carriage, which is drawn 
by clock-work across the disk, while the latter revolves at the rate 
of about thirty revolutions per minute. 
DEMONSTRATION. 
While the plate is being etched I will now let you listen to 
some phonautograms which I prepared in Washington within the 
last two weeks. The reproducing apparatus, or sounder, is con- 
structed on precisely the same principles as the recorder, but of 
smaller dimensions and with more rigid mountings, so rigid, in 
fact, that if it was used as a recorder it would barely show undula- 
tions on a smoked surface when shouting into it. 
The stylus is tipped with iridium like the points of a gold 
pen, the object of this being to prevent abrasion by the continuous 
friction with the hard record. 
In reproducing the sound, I find that it is louder with hard 
contact substances, like metal, than with soft ones like rubber or 
plaster-of-Paris. Hard metals like copper, nickel, or brass, sound 
louder than zinc or type-metal, but the scraping sound, which is 
due to friction, is also increased unless the record surface is 
smooth and very highly polished. 
But when an iridium-pointed stylus is rubbed over clean glass 
a scraping sound is barely perceptible. I am now in communica- 
tion with a firm that is making ornamental glass tiles by im- 20 
Recording Diaphragm and Stylus. 
Reproducing Apparatus. 
Record lines (magnified 6 diameters). 21 
pressing upon red-hot glass plates fancy designs in relief or in- 
taglio by strong pressure. You will readily see that if on the 
same plan we can impress a matrix showing the sound record in 
raised lines upon a glass plate, we would get a groove, in glass, 
giving a loud reproduction with a minimum of disturbing sound 
due to friction. 
In the description of November 12, 1887, I advanced the idea 
of mounting the sounder on a carriage and rails, and have the 
record groove itself be the screw which was to guide the point of 
the stylus across the disk from periphery to centre. This has 
been improved upon by Mr. Werner Suess, the gentleman with 
me here to-night, and who is the mechanician of our little shop 
in Washington. He suggested to mount the sounder on a pivot 
at some distance from the disk and then let the reproducing 
groove guide the sounder across the disk over an arc of flat am- 
plitude. This happy idea is embodied in the present apparatus, 
and is a very ingenious adaptation indeed. 
REPRODUCTION. 
It is, I trust, pardonable if I close by foreshadowing to a cer- 
tain extent the practical applications of the gramophone. 
A standard reproducing apparatus, simple in construction, and 
easily manipulated, will, at a moderate selling price, be placed on 
the market. 
Those having one, may then buy an assortment of phonauto- 
grams, to be increased occasionally, comprising recitations, songs, 
and instrumental solos or orchestral pieces of every variety. 
In each city there will be at least one office having a gramo- 
phone recorder with all the necessary outfits. There will be an 
acoustic cabinet, or acousticon, containing a very large funnel, 
or other sound concentrator, the narrow end of which ends in a 
tube leading to the recording diaphragm. At the wide opening 
of the funnel will be placed a piano, and back of it a semicircular 
wall for reflecting the sound into the funnel. Persons desirous 
of having their voice "taken" will step before the funnel, and, 
upon a given signal, sing or speak, or they may perform upon 
an instrument. While they are waiting the plate will be devel- 
oped, and, when it is satisfactory, it is turned over to the electro- 22 
typer, or to the glass moulder in charge, who will make as many 
copies as desired. 
The electrotype shells are mounted on thick pasteboard, and 
this is backed by a stiff piece of sheet metal. There is another 
process which may be employed. Supposing that his Holiness, 
the Pope, should desire to send broadcast a pontifical blessing 
to his millions of believers, he may speak into the recorder, and 
the plate then, after the words are etched, is turned over to a 
plate-printer, who may, within a few hours, print thousands of 
phonautograms on translucent tracing paper. These printed 
phonautograms are then sent to the principal cities in the world, 
and upon arrival they are photo-engraved by simply using them 
as photograph positives. The resultant engraved plate is then 
copied, ad infinitum, by electrotyping, or glass moulding, and 
sold to those having standard reproducers. 
Prominent singers, speakers, or performers, may derive an 
income from royalties on the sale of their phonautograms, and 
valuable plates may be printed and registered to protect against 
unauthorized publication. 
Collections of phonautograms may become very valuable, and 
whole evenings will be spent at home going through a long list 
of interesting performances. Who will deny the beneficial influ- 
ence which civilization will experience when the voices of dear 
relatives and friends long ago departed, the utterances of the 
great men and women who lived centuries before, the radiant 
songs of Patti, Campanini, Nieman, and others, the dramatic 
voices of Booth, Irving, and Bernhardt, and the humor of Whit- 
comb Riley can be heard and re-heard in every well-furnished 
parlor ? 
Future generations will be able to condense within the space 
of twenty minutes a tone picture of a single lifetime. Five 
minutes of the child's prattle, five of the boy's exultations, five 
of the man's reflections, and five of the feeble utterances from 
the death-bed. Will it not be like holding communion even 
with immortality ? 
Postscript: One of the peculiarities inherent with the gramo- 
phone is the possibility to enlarge the original sound by enlarging 
the printed vibratory characters of speech and then photo-engrave 23 
the same. In this manner it should be possible to get the repro- 
duction at a much greater volume than the original sound. It 
would be interesting if some day speakers in a large hall would 
prefer to do their talking by machine, or to send speeches to 
a convention which they were unable to attend in person. 
E. B. 
[At the close of the paper and after the exhibition of the ap- 
paratus, Prof. E. J. Houston moved a vote of thanks to Mr. Ber- 
liner for his interesting and valuable communication. The 
motion was carried unanimously, and the meeting was adjourned.]