Volume one of our series The Edison Legacy contained thirty-six unpublished disc recordings made a century ago under the supervision of Thomas A. Edison, inventor of the phonograph. Re-mastered from original discs housed at the sound archive of the Thomas Edison National Historical Park, here is volume two comprising an additional offering of unpublished recordings, many of which even Edison himself never heard. We are grateful to the T.E.N.H.P. for permitting us to release these fascinating documents for the first time. The recordings in this set were chosen based on the merit of the singing as well as the reputations of the various singers. Some names are well known to record collectors, while others are obscure, so we have included a biographical sketch for each singer.

One thing that makes these recordings distinctive is the sheer superiority of sound quality that Edison was able to achieve between 1910 and 1912, as he was beginning to experiment with disc recording. His process was capable, for example, of reproducing sibilants such as “s” and “ch,” almost always lacking on other acoustic recordings. He was not, however, as successful in capturing lower frequencies, and the Edison recordings tend to lack the sense of depth sometimes heard on Gramophone Company recordings from the same period. How Edison achieved the sonic clarity that was in many ways superior to his competitors’ is not fully known, for he made certain that every aspect of his recording technique was kept secret, with no paper trail. Only Walter Miller, head of the recording division, understood the entire process. Edison recorded discs for nearly three years before any were published and in fact, of the first 1000 recorded, fewer than twenty-five were ever released. Why were so many rejected? It seems to have been the result of Edison’s personal involvement with the disc-recording project and his idiosyncracies concerning the human voice.

A few facts concerning Edison and his favorite invention:

In 1877, Edison was attempting to develop an alternative to Alexander Graham Bell’s telephone that would circumvent Bell’s patents. The phonograph was actually an inadvertent byproduct of Edison’s experiments with sound transmission. He initially used strips of waxed paper to capture sound vibrations, but his first recordings that could be played back were made on a rotating tinfoil cylinder with an embossing stylus making vertical vibrations. It is a fact, but not well known, that Edison also conceived the idea of recording on to a tinfoil disc, but he did not pursue that option. After a flurry of initial publicity, Edison decided to put the phonograph aside as investors were eager for him to develop the incandescent lightbulb. It wasn’t until January 1886 that he was able to resume work on the phonograph; consequently almost a decade of potential sound documentation was lost to history. In the mid-1880s, Chichester Bell, a cousin of Alexander’s, and Charles Sumner Tainter, began making improvements to the phonograph, which gave Edison added incentive to resume work on his invention. The tinfoil cylinder was replaced by one made of wax, which permitted the grooves to be cut into the surface. This improved cylinder could be played back numerous times and could be preserved. By 1888, Edison felt confident to begin demonstrating the phonograph to public audiences. His initial exhibitions used musical selections played on a variety of instruments recorded in his laboratory. In order to further publicize his amazing invention, Edison dispatched agents to secure recordings of such important celebrities as Tennyson, Browning, Gladstone, Bismarck, Arthur Sullivan, and Johannes Brahms playing a bit of his First Hungarian Dance. Some of these recordings survive today though sadly, most of Edison’s recordings by prominent musicians were destroyed or lost. So many other important voices of the 19th century could have been preserved, but Edison’s primary objective was to promote the phonograph as a dictation machine for business. He simply did not grasp the importance of his invention as a way to capture the voices of his era. There were, however, a few forward looking individuals who acquired phonographs from Edison and began making their own celebrity recordings. Notable among these was Julius Block in Russia, whose collection dating back to 1891 is available on Marston’s The Dawn of Recording (53011).

It wasn’t until the late 1890s that Edison began issuing cylinders under his own aegis. He ignored the possibility of recording great artists, confining his repertoire to popular ditties and comic sketches. During the early years there was no way to mass-produce cylinders except by recording the same selections over and over again. Additional copies could also be made by re-recording from one machine to another, an extremely time consuming process that often yielded poor results.

The 1890s saw the introduction of Emile Berliner’s gramophone, which recorded onto a flat disc, with the cutting stylus making grooves with lateral vibrations. This system had several advantages, foremost of which was the ability to press multiple copies from a master disc. Berliner’s first discs were vastly inferior to cylinder recordings, but by 1900, such significant improvements had been made that the flat disc gramophone began to give the cylinder phonograph serious competition. The disc format increased in popularity and disc companies in Europe and America began recording many of the major operatic celebrities of the day. Not only could discs be mass-produced, but the disc also had the advantage of a greater time duration. By 1903 the twelve-inch disc had been introduced, with a capacity of four and a half minutes of music compared with the two and a half minutes of a cylinder.

Edison solved the problem of cylinder duplication in 1902 with his “gold-mold” process, and in 1905 he finally began producing two-minute operatic cylinders. While many of the prominent names had already signed contracts with disc companies, Edison managed to procure the services of such important singers as Mary Garden, Pierre Cornubert, Antonio Scotti, Anton van Rooy, and an unknown Irish tenor who would become one of the century’s great stars, John McCormack. In an attempt to catch up to the disc companies, Edison introduced a four-minute cylinder in 1908 and within a couple years had added such celebrities as Leo Slezak, Alesandro Bonci, and Sarah Bernhardt to his roster. Similar in appearance to the two-minute cylinder, the four-minute variety had a narrower groove width permitting longer duration. The new cylinder did not score an immediate success, however, since it could not be played on the two-minute machines. The four-minute cylinders also proved problematic because the narrow grooves were less able to withstand the stylus pressure of the reproducer, causing them to wear quickly. The Edison Company continued to promote its new cylinders but it was becoming evident that the disc had gained ascendancy.

For some time Walter Miller had been urging Edison to enter the disc market, but he had refused, believing his cylinder to be the superior recording medium. While Edison was in Europe in 1909, Miller conducted some disc-recording experiments on his own, and when Edison returned, Miller finally convinced him to implement plans for a disc-recording project. Edison, now in his middle 60s and almost completely deaf, suddenly became keenly enthusiastic about developing a disc record superior to all others. During the next five years, he invested an estimated two million dollars and began immersing himself in every aspect of the project.

Some technical details concerning Edison’s discs are contained in the extract below, from the article “Edison and the Diamond Disc” by Angus Joss, published in The American Record Guide, December 1947, vol. 14, no. 4, pp. 99-103:

Jonas Aylesworth, his (Edison’s) chief chemist, was instructed to find a suitable surface for the new disc. The laminated surface, a phenol condensation veneer over a body of compressed wood pulp especially treated to be impervious to atmospheric conditions, was finally decided upon. Since a longer playing record was needed, Edison chose … 150 grooves to the inch. He had known for several years that the weakness of the wax record was in its pliancy. … Celluloid was of course considered, but this material did not lend itself to the conditions of a laminated record with a cheap filler and there were also patents with which to be reckoned. … Many chemical experiments were performed in search for a new material which would be pliable, indestructible, and above all so perfectly smooth that no scratching sounds would mar the quality of the music. Edison said of his disc record: “I made a thicker record of greater solidity which would not shake and vibrate as a whole when played. I developed an extremely hard and smooth surface for the record so that the sound waves would not be flattened out when the diamond point passed over them.” … Ten months were spent in the perfecting of the recorder and the reproducer. The last five months were the most intensive for the solution was at long last in sight. The diamond reproducer, the ultimate in sound reproducing devices for many years, was developed by May 1911. From the beginning of his phonograph, Edison had tried 2300 designs of reproducers to attain the climax. The ultimate one was a most peculiar device which should be described in detail. The larger diaphragm was composed of forty thicknesses of Japanese vegetable parchment, laminated and compressed while heated, but which when cooled retained an .00511-inch thickness. A graduated cork disc was fastened by shellac to the underside of the diaphragm as a reinforcement. A silken cord kept taught by tension was tied on the top side to an ivory fixture. The lower end was fastened to a pivoted fulcrum, on the other side of which was set a diamond cone. “By adopting a permanent diamond point,” Edison said, “I got away from making the sound grooves ‘grind in’ steel needles. By mechanically feeding the so-called tone arm across the record, I eliminated having the delicate sound grooves drag the arm across. In other words I do not use delicate over-tones to move machinery.” The fulcrum was pivoted in a floating hinged weight permanently kept in alignment by a limit pin and adjusting screw. The outlet of the reproducer was fastened to a scientifically tapered tone arm connected to an oval brass horn of ample acoustical capacity. The whole formed a non-metallic transmission of the vibrations from the point to the solid gasketed diaphragm, the reproducer having no tone of its own and entirely free from metallic harshness.

Edison’s discs were vertically cut and used the same size groove as the four-minute cylinder, giving him an advantage over competitors in allowing more time to be recorded—over four minutes on a ten-inch side and as much as seven minutes for a twelve-inch side. Disc recording first took place in Edison’s New York studio but within a few months, he had set up disc recording facilities in London and Paris. Between 1910 and 1913, approximately fifteen hundred discs were recorded, mostly ten-inch but some twelve-inch sides as well. A seven-minute disc should have been a strong selling point but oddly, Edison chose to issue only ten-inch discs. Thankfully, over 150 twelve-inch disc recordings were saved and are now part of the T.E.N.H.P.’s archive. These include more than 120 wax masters that were never electroplated for pressing. Each wax master disc was stored in a sealed, sturdy-metal container, and it is likely that they were never played. A smaller number of twelve-inch test pressings made by the Edison laboratory during the 1920s also exist at the Edison archive.

Edison became increasingly interested in the musical content of his new disc recordings, despite the fact that he had little knowledge of music and no musical ability. In 1910, he purchased a set of the Grove Encyclopedia of Music, embarking upon a self-education program, reading about all manner of musical subjects. He held strong and often eccentric opinions on music in general and singing in particular. He felt that only certain types of voices were appropriate for phonograph recording, with a particular distaste for what he called “tremolo.” He was extremely critical of many of the singers whom he recorded, and was reluctant to publish records which he personally disliked. Humbert Tosi, an Edison representative, was dispatched to Europe to audition singers for the new label. Several hundred voice trial audition cylinders were made for Edison and his team to hear, but despite the fact that many were excellent, none of these singers was chosen to record for the company. The best of these tests can be heard on Marston’s The Edison Voice Trials (52025). An example of Edison’s prodigious waste of talent and money is the acclaimed Finnish soprano, Aino Ackté, for Edison published only one of the 30 recordings she made for the company. Ironically, Edison’s dislike of so many of the voices he recorded might account for the existence of the unplayed wax master discs at the Edison archive.

Some of the inventor’s notes on various auditioned recordings survive. He apparently liked soprano Adelina Agostinelli, noting that “her voice stands out.” But he did not appreciate tenor Carlo Albani in two arias by Verdi : “No good—tremolo fierce. Rotten voice. Don’t want any more of this man if this is a sample.” But he was pleased by a Spanish serenade ditty recorded by Albani: “Very loud yet good. He is a loud singer, but somewhat mellow.” One of the truly great singers who made Edison discs was the Croatian baritone Josip Kašman, who sang under the name Giuseppe Kaschmann. Unfortunately he had a tremolo. Of his recording of an aria from Verdi’s Don Carlo, Edison noted: “This singer has mellow voice but bad tremolo, but not so bad as he could be used alone, as a low baritone he’s more mellow than most of these kind of singers.”

Edison was also often bothered by the sound of the orchestral accompaniments. He preferred using a harp combined with a few strings and perhaps a woodwind instrument. (It seems likely that his strange opinions may have resulted from his hearing deficiency.) As the years passed, he seems to have given up on the harp idea and permitted the choice of instruments to be made by his musical directors.

Late in 1912, Edison introduced his first disc phonograph and the initial offering of his “Diamond Disc” records. These first Edison discs played quite smoothly with little surface noise but within a year or so, customers complained that the edges of the discs were curling away from the core, making them impossible to play. A new formula was developed, which proved to be more stable but the resultant discs were quite noisy compared to the previous issues. During World War I, Edison’s pressings became even noisier since some of the chemicals that he was using had come from Germany and were no longer procurable. By the early 1920s, the etched label was replaced by paper and the pressing material was greatly improved. With the advent of electrical recording in 1925, Edison stubbornly refused even to experiment with the new technology, continuing to record and issue acoustic discs and cylinders. It wasn’t until late 1927 that Edison finally began experimenting with electrical recording and in 1928 he issued his first electric discs in both lateral and vertical formats. By this time, however, it was too late for Edison to catch up to the competition and in October 1929, the record division of Thomas A. Edison, Inc. closed its doors.

Since 1956, Edison’s laboratory and home in West Orange, New Jersey have been designated a National Historic Park under the United States National Park Service. There, much of the equipment that Edison himself used for research and experimentation still exists and every effort is being made to restore all of the buildings to their original state during Edison’s time. Housed at the site are the Edison paper archive, comprising approximately five million documents. The most relevant of these are being microfilmed and gradually will be available through the Internet. The most important documents, together with commentary and analysis, are also being published in book form.

The site, which also contains a large collection of Edison’s recordings and related artifacts, has made it possible for us to issue important Edison recordings. During the summer of 2010, I had the pleasure of spending six days with Gerald Fabris, Curator of the Edison museum and sound archive, to make state of the art digital transfers of the surviving twelve-inch wax masters. A great deal of experimentation went into the choice of styli to be used, which made a tremendous difference in achieving optimum sonic reproduction. We carefully played each disc preserving the audio as high resolution WAV files. The playback equipment at the archive is of the highest quality, and our efforts to preserve this material is commensurate with the latest technology. What a thrill it was for me to be the first to hear many of these discs.

This issue comprises thirty-eight wax masters, four twelve-inch diamond disc test pressings, and one ten-inch diamond disc published recording. Prior to the unsealing of the metal canisters that hold the wax masters, no absolute documentation existed for dating many of these recordings. We are now able to cite precise recording dates for many of the wax masters since in their original containers, each master is accompanied by a hand-written note containing such information as the singer’s name, selection title, and often the recording date. Unfortunately, we cannot be absolutely certain of the recording dates for those items that exist only as test pressings. In the re-mastering process I have taken care to reproduce each recording at its correct pitch. I have applied just enough digital processing to reduce the high level of surface noise, but never so much as to interfere with vocal timbre and nuance. Some of these recordings are flawed by a prominent mid-range hum, probably a result of acoustic transfer of turntable motor noise. I have attempted to reduce this acoustic hum, but occasionally, it can still be detected.

Volume three of the Edison Legacy will comprise further examples from this collection featuring the voices of Dan Beddoe, Lucretia Bori, Clementine de Vere, Maria Farneti, Edoardo Ferrari-Fontana, Lucette Korsoff, Valentin Jaume, Maria Labia, Carmen Melis, Elisabeth Schumann, Marguerite Sylva, Maggie Teyte, Elvino Ventura, and Ellen Beach Yaw.

The Thomas Edison National Historical Park continues to make digital transfers of their audio archive, and as additional Edison material becomes available, we look forward to producing further offerings in our Edison Legacy series.