The Recordings of the Columbia Phonograph Company, 1889-1896

Post written by Mason Vander Lugt, National Recording Preservation Board. This is the second in a series of studies of the early recording industry. Read the first about the North American Phonograph Company and Edison Phonograph Works here. The author is seeking catalogs of any other North American local subsidiaries for the third part. Please write if you can help.

The National Recording Preservation Board is pleased to announce a discography of the cylinder recordings of the Columbia Phonograph Company 1889-1896, available now at Complementing Tim Gracyk’s “Cylinder Lists”, which the NRPB posted to August of last year, this new resource completes documentation of the cylinder output of the pioneering company who “founded the record industry” in the early 1890s.

Columbia’s original headquarters at 627 E. St. NW still stand, today housing Central Liquors. Left image from Phonogram 1:3, 1891, right taken by author, 2016.

Tim Brooks’ 1978 ARSC Journal article Columbia Records in the 1890’s: Founding the Record Industry provides the authoritative story of how Columbia rose to dominate the recording industry through Edward Easton’s vision and grit (and a strategic alliance with American Graphophone). The article tells the story so well, in my opinion, that I’m going to ask that interested readers download it from the ARSC AMP database and focus here on some other aspects of the story.

Why study old catalogs?

The first point I want to address is – what’s the point of studying old catalogs for which few corresponding recordings survive? Original catalogs allow us to learn about the development of the industry through primary sources – the artists and repertoire, the breadth and depth of material and relative popularity of different genres over time.

They can also give us valuable information about the few recordings that do survive, such as those in the UCSB “Recorded Incunabula” playlists and Glenn Sage’s CDs, many of which are sourced from the Library of Congress collections. Because recording logs don’t survive and consistent permanent numbering wasn’t applied in this period, when a selection was added to or dropped from catalogs can provide a natural range for issue dates.

John Yorke Atlee and the United States Marine Band were Columbia’s main attractions in its early years. Images from Phonogram 1:8 and 1:10.

Although most of these recordings don’t survive in their original form, for reasons I’ll touch on below, we can still experience approximations of many of these through later issues. In 1895, as Columbia moved into new offices in New York City, the United States Gramophone Company began recording many of the same artists in Washington. Actor David C. Bangs, auctioneer W.O. Beckenbaugh, popular singer Maud Foster, the Brilliant Quartette, banjoists Cullen and Collins, clarinetist Felix Iardella (Jardella on Berliner) and cornetist W. Paris Chambers are among those who would record for Berliner, but not later Victor or Columbia discs.

Many performers from these catalogs recorded prolifically in the following years, and these lists provide insight into their early repertoire. Others, like George H. Diamond, Dan Kelly “Pat Brady” and Al Reeves “The Famous Banjo Comedian” were famous and/or widely recorded in this period but leave us unfortunately few or no recordings today. Similarly, much of the band and tin pan alley vocal repertoire remained popular into the disc recording era to survive to the present, but a few, like “Maggie Murphy’s home”, “Down went McGinty” and “Ta ra ra boom de ay!” were popular enough at the time to warrant parody records but faded before the more durable formats or prolific output of later generations could save contemporary performances.

Even when we can’t hear contemporary recordings, reading about what was recording can feed the imagination – whether it’s the calliope imitations of the Brilliant Quartette, descriptions of Dan Kelly’s and Russell Hunting’s skits, or the 1892 Republican campaign song “Democracy’s Going to Grass”, imagining these lost sounds may not be an academic exercise, but it can be a fun one.

Russell Hunting (as Michael Casey) and Dan Kelly (as Pat Brady) were two of the most famous recorded humorists of the early 1890s. Both began with other companies (Hunting with New England, Kelly with Ohio), and were added to Columbia’s catalogs beginning April 1893. Images from Phonogram 2:8-9 and 3:3-4.

Finally, I hope I can revive some interest in the music of this period. Within the relatively small community of people who are interested in historical American music, I find that folk music is revered as authentic and worthy of preservation and study while popular music, excepting maybe Sousa’s marches and a few great hits of tin pan alley, are ignored and forgotten. Still, sheet music remains for many works when recordings don’t. Searching Youtube reveals a few artists working in this direction, but I’d like to encourage musicians to consider reviving these (or let me know if I’m off base and this is a thing…)

Columbia’s extent and position within the industry

Just how important was Columbia compared to the North American parent company and the other sub-companies? Columbia may have been the first sub-company to institute their own recording program. The first document reproduced in the discography advertises “whistling solos by artistic whistlers” alongside recordings taken at the Edison Works. This is dated November 15, two months before Edison agreed to allow the sub-companies to manufacture their own recordings. Other companies may have been recording on a small scale for local demonstrations, or purchasing from independent recordists – every phonograph sold or leased at this time would have been capable of recording – but this Columbia advertisement is probably the first surviving documentation of the public sale of records (Edison’s would have been sold only to the sub-companies).

In these first 8 months, entertainment recordings would have been used to demonstrate the phonograph to potential business customers, or in public exhibitions, as the November ’89 pamhplet describes “At the Bull Run Panorama the Phonograph delivers lectures descriptive of the scenes there displayed”. By July 1890 Columbia had installed their first automatic phonograph, in the historic Ebbitt House Drug Store at F St. NW and 14th St. NW, only a few blocks from their headquarters at 627 E. St. NW[i]. By November, they had placed more than a hundred across the city[ii].

This quickly became the dominant model for exploiting the technology and by April of the following year, according to Phonogram, Columbia was shipping records across the country[iii]. In the same issue, Columbia began running ads boasting “we sell more records than all other dealers combined”. This might easily be dismissed as advertising rhetoric except that when Columbia’s president Edward Easton said the same at the 1892 convention of the North American sub-companies nobody challenged him. In fact – the context of the discussion was whether the other sub-companies were entitled to royalties on Columbia’s sales within their territories.

Columbia identified titles and performers on their cylinder records using a printed paper band, similar to North American’s title ring, but without requiring the special channeled rim. Photo by the author.

Why do so few survive?

If Columbia sold so prolifically, why do so few of their recordings survive to the present day? The answer is probably a combination of factors –

  1. The entire industry was relatively small in scale. Columbia bought less than 70,000 blanks from Edison between 1889 and 1894[iv], and some portion of these would have been resold to Columbia’s business customers for dictation. For comparison, according to Antique Phonograph Monthly, Edison sold approximately 2 million cylinder records in 1900 and 4 million in 1902. As discussed in the previous post the market for recordings outpaced supply in this period, so Columbia’s output was probably limited by their capacity to record and/or duplicate, or Edison’s ability to supply blanks.
  2. The format is unstable. Brown wax, designed for ease of recording rather than durability, wears easily. The Edison factory would consider a master record too worn to duplicate from after 200 plays[v]. Those that were saved might have been degraded by mold or chemical instability (efflorescence).
  3. They were seen more as working material than cultural record. Not that they didn’t understand this potential for the technology – both Edison and Columbia advertised the phonograph’s utility to preserve voices of family or “great artists”, and Columbia advertised some artists such as cornetist Jules Levy as great artists, but they probably regarded most of their records the way we might think today of old CDs or cassettes. If anyone at Columbia preserved special recordings like Edison did, there is no record of it.

Where’s Len?

Histories of recording pioneer Len Spencer, such as this printed in the relatively contemporary Phonoscope, usually place him virtually at the founding of the Columbia Company in January 1889, but his first appearance in these catalogs is in two undated catalogs from ca. 1895, after the company had opened a New York office and begun recording New York artists. A 1930 article by Columbia Manager Frank Dorian printed in Phonograph Monthly Review similarly suggests Spencer recorded for them “by the round” before his catalog debut[vi].

The most likely explanation is that Spencer recorded briefly in the uncredited vocal series that ran between December 1890 and January 1892 (and/or the “recitations” series, variously uncredited). A sudden drop in both series after January 1892 may corroborate this in addition to the statement in Phonoscope about his move to the United States Phonograph Company (at that time, in fact, the New Jersey Phonograph Company, another North American subsidiary) and Spencer’s headlining role in a 1892 New Jersey catalog in the Library of Congress.

Other possibilities include lost catalogs – unlikely since the found and included catalogs seem to present naturally growing and declining series – or that Spencer recorded as a kind of independent contractor, as Atlee was known to, whose records would supply the local trade but wouldn’t be distributed.

P.S. “New Paris Waltz” by E.M. Waterbury is available for adoption at the UCSB Cylinder Audio Archive


The First Book of Phonograph Records

Written by Mason Vander Lugt, National Recording Preservation Board

The First Book of Phonograph Records is a founding document of the recording industry, representing a log of recordings taken at Edison’s laboratory leading up to the first sales of musical records by the North American Phonograph Company in 1889 and 1890. It is now available in a digital edition here, presented by the National Recording Preservation Board, with scans courtesy of the Thomas Edison National Historical Park and National Park Service.

When the North American Phonograph Company was incorporated in July 1888, the Edison Phonograph Works were granted an exclusive right to manufacture phonographs and accessories, and North American the exclusive right to market them. North American wouldn’t deal directly with the public, but would license exclusive territorial rights to regional sub-companies, which were formed in the fall of 1888 and spring of 1889. These licensing agreements established that the Works’ exclusive right to manufacture phonograph supplies would include “special extras”, like recordings of music [1].

North American’s first list of recordings for sale, May 1889

The first of these were sold in May 1889, in mixed lots of 6 and 12. A. Theo. E. Wangemann began logging recordings in the First Book only a few days before these were advertised, but had in fact been developing his process over the preceding year [2]. In December Thomas Lombard, North America’s vice president, urged the Works on the sub-companies’ behalf to publish a catalog of available recordings and allow them to choose what selections they would receive. The Works obliged and in mid-January of 1890, North American published the “Catalogue of Musical Phonograms for the Phonograph, First Edition[3]. Little more than a week later, on Jan. 25th, Edison announced that he would be discontinuing the Works’ recording operation, apparently upset by complaints sent by the sub-companies [4], and suggested they should begin making their own recordings.

North American’s first catalog, January 1890

Roughly the first three quarters of the First Book represent the period between the first sales in May 1889 and Edison’s suspension of recording in January 1890. It provides insight into the state of the art, from cylinder shortages and machine malfunctions, to rates of success and productivity. It gives us a sense of the repertoire, some of which you’d expect, some which seem to be performers’ specialties. It documents the first sessions of future recording stars, like Edward Issler and George Schweinfest, and gives a rare glimpse at a generation of pioneer recording artists about which unfortunately little is known or likely to be found. The remaining quarter presents scattered clues about the Works’ come-and-go recording activities in the following years – a few more recording sessions, shipments for exhibitions, orders for repairs or duplication, and other various notes.

It’s worth noting that this is not the first time the First Book has been made available to the public. Allen Koenigsberg transcribed it for Edison Cylinder Records, 1889-1912, first published in 1969. As this important work becomes rare or expensive, we hope the First Book Digital Edition will ensure this portion remains available, as there is still important work to be done in researching this period in recording history. To this day, the recording activities of the North American Phonograph Company and its affiliates are poorly understood, partially due to the complicated relationships between the companies and partly due to patchy documentation. This resource and article represent the first in a series of studies on this topic. Readers are invited to send additions or corrections to

At the third annual convention of the sub-companies, Lombard noted that North American had made arrangements with the New York Phonograph Company to supply records shortly after Edison’s announcement [5]. North American began their own recording operation in February of 1890, or maybe January. Two letters dated February 24, 1890 offer clues – one to Edison from his secretary A.O. Tate saying “I think the North American people have started in on music business”, and another from the Vice President of North American Thomas Lombard to the sub-companies assuring they were “arranging to increase [their] facilities for making these records”. They published three lists of available recordings that year, one each in June, August and October, which are reprinted in Edison Cylinder Records.

Several of the sub-companies took Edison’s offer to begin their own recording programs. In February 1890 the New York and Columbia Phonograph Companies published versions of the January 1890 North American catalog with additions or substitutions of their own. By the following summer the trade journal Phonogram noted “The securing of musical selections has grown to be quite an industry. It is done mainly by the Columbia Phonograph Company of Washington, D.C., the New Jersey Phonograph Company of Newark, N.J., the New York Phonograph Company and the Ohio Phonograph Company” [9]. The sub-companies role in developing the industry will be a future installment in this series of studies mentioned above, but in general, the sub-companies dominated the records market from fall 1890 until at least spring 1892.

In the meantime, Edison returned his attention to developing technologies to duplicate records, understanding that the limited scale of recording “by the round” was expensive and unnecessary. At the second annual convention of the sub-companies in June 1891, Edison invited the attendees to visit the Laboratory and Works, and proposed launching a service to duplicate their recordings. North American reluctantly agreed to permit the service in August after negotiating a royalty with Edison.

Judging from the discussion at the convention, the sub-companies were under the impression that the Works would either duplicate for a fee and return records to the sub-companies to market or that North American would sell the records and pay the sub-companies a royalty. In fact, the Works would surreptitiously acquire recordings taken by the Columbia and New England Companies and advertise them for sale [6]. The Works’ first and only catalog, published sometime in late 1891 [7], lists recordings taken at the Works between April and October (First Book pp. 180-190, 206, 208) alongside band, whistling, and recitation recordings recorded and marketed by Columbia. When Columbia and New England protested in December 1891, Samuel Insull (North American’s president) wrote to Tate that this approach was unsustainable, and requested he draft a response.

In his response, dated 1/12/92, Tate suggested they insist that the Works maintained the exclusive right to manufacture recordings, but would allow the sub-companies to continue making original (one of a kind) recordings if they would deal exclusively with the Works for duplication. The Chicago company had begun offering a duplication service at a lower rate [8]/[9], and Tate may have staged the event as a way of protecting the Works’ proposed service.

He told Insull that “The prices which now prevail are absurdly high, and the character of the records is very poor. It is only by employing an adequate method of duplication that the prices of these records can be brought down to a reasonable point, but this cannot be accomplished unless the work is concentrated in one spot, so that whoever takes it up can have the full benefit of that whole class of manufacture”.

Tate must have had an interesting evening, because the day after writing the above, he wrote to Edison proposing nearly the opposite – that the Works discontinue their recording program and duplication service (to the local companies) and enter into an agreement with independent recordist and entrepreneur Charles L. Marshall [10] to supply a large number of duplicate recordings for the New England, New York, and New Jersey coin-slot markets. Insull wrote to Edison to give his support to the idea in late January, and a draft of the proposal was written in which the Works would manufacture recordings and duplicate Marshall’s, sell them to Marshall with North American’s consent, and Marshall would distribute them across the northeast with Edison’s legal protection.

The agreement seems to have fallen through, but the concept of re-centralizing the business had stuck. Shortly after, the Works and North American would again join forces to manufacture and sell recordings, anticipating a larger consolidation of the company later that year [11]. The Works resumed recording in February, and in April, North American published “Bulletin No. 1 – The North American Phonograph Co. List of Musical Records for the Phonograph”. Advertisements in Phonogram were headed “The ‘Edison Records’ / Manufactured exclusively for this company at Orange, N.J.”.

The first catalog of North American’s “Numbered Series”, which ran April 1892 – August 1894. Image from Phonogram Vol. 2 No. 10

The first of these must have been from the recording sessions listed in the First Book, pages 192-211. After this, it’s likely that master records were taken at the North American headquarters in New York, but that duplication continued at the Works. At the annual convention in June 1892, Lombard told the sub-companies that their goal was to compete with them directly with the advantage of low-cost, high quality duplicates [12].

North American advertised heavily in 1893, including a brochure advertising phonographs for home entertainment. By March 1893, recordings were taken at a dedicated facility at 56 E. 14th Street, New York [13]. At least one surviving example – Silas Leachman’s “Daddy Wouldn’t Buy Me a Bow Wow” was recorded at the company’s offices in Chicago.

North American’s December 1893. The last known to the author.

A September 1893 letter from Tate to Edison gives us a rare overview of the business – the joint North American / Works recording operation had made 18,600 records and had shipped 7,100 [14]. They had 120 selections to choose from. He said that there had always been more of a demand for recordings than available supply, and estimated that the market could support 150 to 200 thousand records per year. By November the catalog showed that the number of selections had cumulatively reached more than 800. By December, 880.

Tate’s gamble seemed to have worked. North American’s recording program grew steadily, while at the fourth and final convention in September 1893, R.T. Haines, president of the New York Company complained that demand for high class records always outpaced supply and that high-quality duplication was unavailable to the sub-companies [15]. Columbia, as usual, was the major exception, having purchased Leon Douglass’ patent in June 1892.

North American’s accounting ledgers show continued recording and sales until the company’s collapse in August for reasons having little to do with the recording business [16]. Coming full-circle, the company supplemented their supply of original recordings by purchasing from the New York Phonograph Company. The August 20 “closing entries” in the ledger estimate $5,200 worth of musical records unsold.


[1] (Image 725)
[2] Patrick Feaster’s dissertation “The Following Record” provides the most detailed history I’ve found of this period. For this citation, see p. 141.
[4] (images 150-153)
[5] Transcriptions of the proceedings of each of these conventions are available on Patrick Feaster’s website here. In this instance, see page page 83 (original pagination).
[6], also “Following Record” footnote 125 (p. 231)
[7] – the Feb. ’92 date supplied is for an order if you look carefully
[8] Regarding “available”
[9] Regarding “Chicago”
[10] More on Marshall in Phonogram 1:3 p. 63
[11] In the summer of 1892 Edison and Tate would become President and Vice President of North American, respectively, and would offer a “suspension agreement” to the sub-companies in which North American would market phonographs and supplies to the public. See Ray Wile’s 2004-2005 ARSC Journal articles for details.
[12] Page 84 original
[13] Phonogram 3:3-4, p. 383. “13th St.” in Phonogram. Ledgers and journal refer to recording at 14th St.
[14] It’s unclear whether these figures are for the year or the program to date
[15] Regarding demand, p. 114, regarding duplication, p. 55
[16] See Ray Wile’s 2004-2005 ARSC Journal articles for details about North American as a business.

The “Big Tent” of the Antique Phonograph Society

Post written by George Paul, Antique Phonograph Society President and co-author of “The Talking Machine: An Illustrated Compendium”

Among those of us who have a serious interest in the history and surviving artifacts of sound recording and reproducing history, there is an elephant in the room. The diversity of collectors and academics often polarizes us into two broad groups: one which regards the other as knuckle-dragging dilettantes, and another group which regards the first as pointy-headed bores. I’d like to make the case for a “big tent” approach which includes a broad-enough scope to satisfy serious collectors/historians of our field as well as practical tips on collecting/repair/enjoyment to benefit those who don’t know a Berliner from a Brunswick. The Antique Phonograph Society is endeavoring to make this “big tent” a reality.

For the typical ARSC member (is there such a thing?), probably the most noteworthy APS achievement in this regard is the searchable online archive. The ever-expanding web site has recently obtained one-half terabyte of storage capacity from a world-class internet service provider and uploaded in excess of 15,000 pages (so far) of documents ranging from January 1890 to the present. Some of these documents consist of primary resources (period articles, original catalogs) while others are articles published in journals around the world from the 1950s to the present on a dizzying variety of topics. The APS was formed in 2011 from the merger of the Michigan Antique Phonograph Society and the California Antique Phonograph Society, and all past issues of their respective publications are included in this archive. There are modern book features by R.J. Wakeman on Brunswick and Sonora – each presenting new research, hundreds of pages in length, and copiously illustrated – as well as historical reprints, like Henry Seymour’s The Reproduction of Sound (1918) and Ogilvie Mitchell’s Talking Machines (1921). Articles are divided into “intermediate,” “advanced,” and “technical” categories to simplify browsing, and new articles and features are added regularly. Members can even view videos of presentations and a large number of interesting talking machines/records in action.

The APS’s Mission Statement and Bylaws (under the “Governance” tab on the home page) will make clear the organization’s primary thrust: education and public awareness. Our 52-page full-color quarterly journal The Antique Phonograph also demonstrates the variety of interests and depth of knowledge of our members. You can view a sample issue on the website. The full text of the journal is searchable in our archive but we’ve also prepared a subject index (for the journal only) if you prefer to browse. I think you’ll be impressed by the breadth of topics (there’s even a “records” section!).

Other APS member activities include development of the CPS1 (“Cylinder Playback System 1” – a device which can play virtually any type/size/speed cylinder record with minimal wear and wide frequency response), restoration of cylinder records, and contribution to the UCSB Cylinder Audio Archive. APS members include Grammy and Oscar winners, descendants of record label founders, jazz recording pioneers and even museums and libraries.

There will always be a wide range of folks who share our field of interest, but if the memberships of ARSC and the APS could be represented as a Venn diagram, I feel certain that a significant portion of each organization would find our interests are shared. ARSC members are encouraged to browse our web site ( and consider joining. The online archive alone is worth the annual dues ($30 in the U.S.).

The 2016 Annual Meeting of the Antique Phonograph Society