Appendix B: Notes on Individual Series

Columbia Master Book, Volume I, Tim Brooks, ed.

1 – 4999 | 5,000s | 06000s | 6,000s | 7,000s-9,000s | 10,000s | 11,500s | 12,500s | 13,500s | 14,500s | 15,000s | 16,000s | 17,000s | 18,000s | 20,000s | 21,000s | 22,000s | 25,000-29,000s | 31,000-33,000s | 34,000s | 35,000s | 36,000s | 37500-38099 | 40,000s | 41,000s | 42,000s | 43,000-44,000s | 45,000s | 47,500s | 48,000s | 50,000s | 51,500s | 52,000s | 53,000-54,000s | 55,000s | 56,000s | 57,000s | 58,000s | 59,000s | 60,000-64,000s | 65,000s | 66,000s | 67,000-68,000s | 69,000s | 70,000s | 71,000s | 72,000s | 73,000s | 74,000-76,000s | 82,000s | 83,000s | 84,000-89,000s | 90,000s | 91,000s | 93,000-97,000s

Return to Appendix B: Columbia Matrix Series (1901-1934)

1 - 4999

Columbia’s first, general purpose matrix series. Included both 7” and 10” discs, plus a few 12” and 14”. The 2000-2999 block was reserved for Far Eastern location recordings, few of which have turned up in the U.S. Other smaller blocks were also evidently reserved for foreign location recording, for example 1260-1350 (Shanghai). Top


All items in this series appear to be Spanish language, made for the Mexican and Mexican-American market. Curti’s Mexican Orchestra is frequently seen, at least in the U.S. Most 5,000s were originally issued on the single-face black and silver label, and were later doubled in the Spanish “C” series. Starting point for the series is uncertain, the lowest known number being 5357. As for location, a Mexico City studio was active as early as 1904 (Columbia Record, April 1904), although Bauer lists some 5,000s as recorded in Madrid, and others are of U.S. origin. After 5999 the series jumps to 13500. Top


Early Spanish series. All known examples are by the Banda Española, and have been found either on Marconi or on the double-faced Argentinian “T” series. One, in fact, is “Himno Argentino” (06059). The 06000s may be Marconi catalog numbers, with matrix numbers unknown. These Marconis may have been intended for export to Argentina. Top


Twelve-inch masters recorded in England, some of which were also released in the U.S. The earliest were released in the U.K. in February 1908 (see Taylor, Columbia Twelve-Inch Records in the United Kingdom, 1906-1930, 229). Top


The purpose of the next three blocks is uncertain. They may have been the first series reserved primarily for trials and special records, although they also include a few foreign selections. In the 7,000s the Columbia files contain cards for 7000-7049 only, used for experimental vertical-cut dubbings (?) from regular masters. Dates for 7000-7036 are uncertain, although all were drawn from masters made no later than 1904. A second batch (7037-7049) is dated December 1909-February 1910, and used masters from 1909-1910 [end p. 421] (after that the experiments jumped to 8071). Therefore the 7000 series may have begun as early as 1904. None of the vertical-cut tests appear to have been released; all known Columbia discs are lateral-cut.

Bohemian and French selections have been reported in the range 8002-8067, while 10” vertical-cut tests dated February 1910 to December 1911 are found from 8070-8096. The file card for matrix 8097 is dated August 1912 and labeled “Special for V.H.E. (Up and Down),” meaning, apparently, a vertical cut test made for recording director Victor H. Emerson. A private recording of a Masonic song called “Hello Bill” by J. W. Myers has been reported on 8007. Nothing else is known in the 8,000s.

The 9,000s were apparently the 12” equivalent of the 7,000s/8,000s, although very few examples have been reported. No. 9000 is a 12” Bohemian selection by Prince’s Band, recorded and released in 1907. The files indicate that 9001-9004 were 12” vertical tests made in early 1910, and presumably not released. No other examples are known. Top


This series was for Italian titles recorded in Milan. Early numbers were released in the U.S. as single-faced discs (1904-1906), while higher numbers turn up on the double-face “C” and “E” series. Bauer claims an early example (10045) was recorded in 1903; in April 1904 the Columbia Record reported that Frank L. Capps was in Italy recording artists who appeared later in this series. The series continued into the lower 11,000s. Top


Known examples are Portuguese titles numbered in the 11,700s and higher, released in the Brazilian “B” series. One master remaining in the Columbia vaults (11506) is dated February 1908, which could indicate the series’ start date. The series probably continued into the lower 12,000s, as similar material is found there. Top


All known examples are Austro-Hungarian (and German?), some recorded in Vienna and some apparently in the U.S. Many of the earliest numbers (in the range 12501-12600) were released in the U.S. as single-faced discs during 1906-1908, while higher numbers turn up in the “E” and sometimes “A” series. The author has one of these masters (13192) on one side of a double-faced black & silver test pressing made no later than 1908, and another (13199) on a post-1908 “notes” label test pressing. Top


Spanish titles for the Mexican market, most of which were released in the “C” series. One unidentified part in the Columbia vaults (14257) is dated September 5, 1908. Top


Matrices 14500-14528 were political campaign records by William Howard Taft, William Jennings Bryan and Prohibition Party candidates, released in the fall of 1908 prior to that year’s general election. Matrices 14529-14999 were used from September 1908 to September 1909 [end p. 422] for tests and personal records. Among these were such varied fare as a “paper horn test” (October 26, 1908), vertical-cut tests, a recording by Booker T. Washington of his Atlanta Exposition Speech (14605, December 5, 1908), an unidentified title by songwriter Gus Edwards (14941, ca. July 6, 1909), vocal instruction records, and many private recordings by unknowns who paid for the privilege of making a record. In addition, there are (test?) recordings by a number of regular Columbia artists, including Vess L. Ossman, Fred Duprez, William McEwan, George J. Gaskin, Merle Tillotson, W. Francis Firth, and Carroll Clark. Several sides by French artist Louis Verand were released in the “E” series. Some entries in the files are tantalizingly unclear; could “Miss A. Neilson” (sic) on 14757-14760 (Mar. 19, 1909), be operatic soprano Alice Nielsen? Top


A mixture of trials and foreign language recordings, both 10” and 12”. Matrices 15000-15128 are Spanish language items, for (or from?) Puerto Rico, issued in the “C” series. Higher numbers include trial recordings, vertical cut tests, Chinese recordings, and more Spanish material. Top


Trials and specials with a known range of 16430-16949. All known examples are 12”, with the odd exception of 16949, which is the famous 1910 10” Double Disc Advertising Record (“… perfect in surface, perfect in tone, and extraordinary in durability!”). Some of the more interesting examples:

Table 7:

16,000s Matrix Examples

Return to Columbia Tables

16454 Oct. 1910? “My description of the big fight by Jack Johnson, Heavyweight Champ of the World. July 24, 1910.” Part of a multi-disc set the by first black heavyweight champion, used as the soundtrack for a film. The date of the fight is in error; it took place on July 4, 1910.
16472 Mar. 9, 1911 Mme. Nordica and Barron Berthald (duet).
16473 Mar. 14, 1911 Mary Garden (no titles).
16494 Dec. 28, 1912 Speech of Napoleon before the Battle of Austerlitz, by Len and Harry Spencer.
16523 Jun. 25, 1913 “The Maniac Wife,” special for Scheindelman.
16524 Jun. 27, 1913 “The Elopement” (special), by Jones, Spencer & Porter.
16888/9 Apr. 29, 1914 Lew Dockstader, Parts I & II (remade May 7).



The only masters reported in this range are two college songs by the Nassau Quartette of Princeton (17262-17263), issued on Columbia A1053, which was released in late 1911. Based on this slender evidence this would appear to be another block for trials and [end p. 423] specials. In the mid-1930s this same block, along with the 15,000s, 16,000s and others, was used for general U.S. issues, sometimes using prefixes (CO-, XCO-, P-, etc.) and sometimes not. By the late 1930s all Columbia matrix numbers carried prefixes, so this may represent a transitional period. In addition, one British master from the late 1920s (17536) has been reported in this range. Top


According to the Columbia files, allocated to Manila. Examples reported are in the range 18085-18129, and are found on South American “C” and Filipino “M” series issues. Top


Early Armenian and Czech titles have been reported in the range 20003-20032, most of them dated December 1904. At least two of these (20022-20023) were released in the U.S. in 1906. Nothing else is known. Top


Hawaiian titles reported in the lower range (21251-21332). One Hawaiian item, issued in the “Y” prefix series, according to the label was recorded in Hawaii, pressed in Brazil and for sale in the U.S. only! The rest of the block seems to have been material for the Latin market, released in the “A”, “C” and “P” (Peruvian) series. Most are Spanish language, although there are some Deiro accordion renditions of Neapolitan songs. Blacker reports Polish language items, but this has not been confirmed. The lowest dated master is 21380 (October 3, 1911), so the series may have begun prior to 1911. The lowest identified master is 21251. Top


Appears to have been reserved for Spanish recordings, at least the portion above 22500. A number are reported in Bauer, with Madrid, Barcelona or just “Spain” given as the recording location, ca. 1906-1907. Only one matrix below 22500 survives in the Columbia vaults (22160), dated November. 23, 1911, showing that the 22000-22500 range was also used, but its contents are unknown. Top


This block was reserved at the outset of disc production for 10” masters made by Columbia’s British affiliate, and it proved sufficient for 13 years’ output. Some of these London masters, by artists such as Peter Wyper, Joe Hayman and Billy Williams, were released in the U.S. as single-faced discs and later in the double-face “A” series. Top


I’m going out on a limb here, since data is sparse, but I suspect that this entire block was used for 10” tests and personal recordings in the early- and mid-teens. This would fill the “specials” gap between the 14000-15999 block, in use from 1908-1913, and the 60,000s, which began around 1915 or 1916.  [end p. 424]

There are only a handful of examples in the 31,000s, including “Boy Scout Patrol Calls” by Ernest Thompson Seton (31352-31353, released on A1331 in July 1913) and “Hello Honey” by actress Elizabeth Brice (31896), who performed the song in the Ziegfeld Follies of 1913. The latter was clearly made for Miss Brice’s professional purposes, as she does some vocal exercises along with the song.

Many examples have turned up in the 32,000s, including two versions of the 1913 Columbia Demonstration Record, one for the U.S. (32030) and the other for Canada (32210). A vinyl test exists of Irving Berlin doing a specialty number for the Friar’s Club (32229). There is also a series of 1914 dance instruction records distributed in conjunction with newspapers, Henry Burr doing a pitch for Climax Chewing Tobacco, and many personal recordings, most of them dated 1914 in the Columbia files.

A few examples have surfaced in the 33,000s, including Swedish titles by J. A. Hultman, the Yale University Quartette, and two promotional recordings for the Metropolitan Life Insurance Company (33509-33510), the latter dated October 11, 1915. Top


All known examples are language instruction records made for the International College of Languages and International Correspondence School, between 1913 and 1917. Known range is 34000-34325. Top


This was reserved at the outset of disc production for recordings made in Russia. The most famous are probably those made by a number of eminent artists at the Court of Czar Nicholas II in St. Petersburg, beginning in June 1903. Among the artists were Frey, Sharonov, Labinski, Michailowa and the Czar’s private band. Columbia was extremely proud of these recordings, and promoted them heavily (Columbia Record, February 1904, et seq). The series also included a few recordings made in Moscow, Polish titles made in Warsaw and some Italian selections, although the latter could have been recorded in St. Petersburg. Some of these masters were later issued in the U.S. in the “E” series, but many were not. Matrices 35700-35999 were reserved for London (1915-1916). Top


The only known items below 36300 are a few Russian titles released in the U.S. in the “E” series in the early teens, including some by “artists of the Imperial Theatre.” This block may therefore be a continuation of the Russian recordings in the lower 35,000s. A few Dutch items are found between 36300-36349. Regular 12” U.S. masters begin at 36350, continuing the 30000 series. Top


Spanish (and Portuguese?) titles, of uncertain origin, found on both single-face and “C” series issues. Matrices 38000-38099 appear to be exclusively 12” discs. The regular 10” U.S. domestic series picks up at 38100, as this series jumps to 48000.  [end p. 425] Top


Reserved at the outset of disc production for German titles, made in Berlin, Stuttgart and Vienna. Many were released in the U.S. as single-faced discs, and later in various double-face series on a variety of labels. Top


This is also a very early foreign-language series, a mixed bag of German, Austrian, Bohemian and Scandinavian items recorded in Europe. Bauer reports recording locations in Kristiana, Copenhagen, Prague, Norway, Stockholm and Budapest. As a colleague of mine once remarked, “Probably the recording studios were in a boxcar...”

Analysis of titles and artists suggests there may have been sub-blocks within the 41,000s. The following is a very tentative breakdown of how this might have worked.

41000-41299: Austro-Hungarian (Prague, Budapest, Vienna)

41300-41400: Scandinavian (Copenhagen, Kristiana, Stockholm)

41400-41749: German (Berlin, etc.)

41750-41999: Italian (Milan) Top


Italian titles apparently recorded in Milan (e.g., the Banda Muncipale de Milano). Known range is only 42019-42374, however, so other portions of the 42,000s may have been used for other purposes. Top


Around 1915 Columbia appears to have discontinued most of its scattered foreign language series and made this its principal 10” series ethnic series for all languages except Spanish. The series began about the time Columbia had to discontinue recording activities on the European continent due to the outbreak of war. The starting point may have been 43500, as only two Swedish titles are known below that number while many have been identified above it. Among the artists are Rhoda Bernard, Cantor Josef Rosenblatt, Torcom Bézazian and the Urania Quartette Club of New York, a principal source for the domestic “E” series during 1915-1917 (roughly, E2700-E3300). After [matrix] 44999 ethnic recording jumped to 58000. Top


French titles reported in lower range, probably recorded in the teens. The regular 10” U.S. domestic series picks up at 45500. Top


The Columbia files indicate that recordings were made in this series ca. 1906, but no details are given.  [end p. 426] Top


Matrices 48,000-48,499 used for Spanish/Cuban titles issued in the export “C” series. The regular 12” U.S. domestic series picks up at 48500. Top


This was reserved at the outset of disc production for French recordings, and quite a few made in Paris were released in the U.S. as single-faced discs during 1904-1906. To confuse matters one Hungarian march (50490) is listed in the files as “made in England”; perhaps this meant it was pressed there. Known range is 50003-50853, but the series may continue into the lower 51,000s as Bauer reports a ca. 1905 Paris recording on matrix 51044. Top


Probably a continuation of the 16,000 12” trials and personals series. Some interesting tests involving “transmitters” (sic) took place in this series in March, 1920. Could have been a radio hookup, or possibly an early attempt at electrical recording? (See also the 63,000s.) Later, in 1924-1925, numerous tests using the Western Electric system, which was ultimately adopted, were conducted in this series. The first such test was an electrical recording of “Oh Gentle Savior” by Lucy Van de Mark, made on October 2, 1924.

Personal recordings in the 51,500s range from such curios as “The Shredded Wheat Band” (June 1916) to piano solos by Percy Grainger (September 1931). Matrices 51775-51781 is a live recording of President Calvin Coolidge’s inaugural address, presumably taken from a radio broadcast on March 4, 1925. Unfortunately the Columbia engineers had “cutter problems” part way through, and wound up with only 24 minutes of the 40-minute address. At the end there is a bit of applause, some background noise, and then a distant band plays “Hail to the Chief.” Top


Not used during the acoustic era. The first hundred numbers are classical masters made in Europe, some of them issued in the early 1930s in the 67000-D/68000-D series. After that many ten and 12” electrical recordings are indicated in the Columbia files, some prefixed “P” (for personal?), but with no titles or dates given. This may represent a continuation of the 51,500s trials and personals series in the mid-1930s. Top


No examples found, and no cards in Columbia files. Top


Initial recordings in this block were apparently made in Buenos Aires, perhaps as early as 1905 (e.g., tango legend Gabino Ezeyza on 55098, Jose M. Silva on 55003). Italian, Spanish and Greek titles have been reported dating from ca. 1908 until at least the mid-teens. Many “C” and “E” series issues were drawn from this block. Guido Diero, Francesco Daddi and Emilia Vergeri are among the artists featured. A recording of Handel by the Tokyo Academy of Music Orchestra issued in the late 1930s was assigned matrix. [end p. 427] 55000/55001, but this may have been a later attempt to re-use the block. Top


A few widely scattered examples are all Spanish, mostly found in the “C” and “T” [catalog] series. Dates are uncertain, although it may have begun ca. 1920; by the end (56999) the series had reached April 1923. Among the artists is José Mardones (56008-56009). Top


Scattered examples in the lower half of this block (57000-57499) are Spanish titles issued from the late teens onward in the “C” and “T” series. Recording dates in the Columbia files jump all over the place, ranging (non-sequentially) from 1912 to 1923. Examples seen above 57500 are Cantonese, with a known range of 57525-57703. At least one (57539) is labeled “Cantonese-Made in China.” Top


A continuation of the domestic foreign language series, containing a wide range of material issued principally in the “E” series. Russian, Scandanavian, French, German, Greek, Italian and Yiddish items have been reported. The series next jumps to 84000. Top


This seems to be split into sub-blocks. The first few numbers (59000-59010) are 12”, including the intriguingly named “Sky Devils” set-whatever that was. The rest of the first 400 numbers are 10”, dating from 1914 onward according to the files, including Lovey’s Band of Trinidad (ca. 59300-59380), recorded in Port-of-Spain. A Greek selection is found on 59061. Numbers above 59400 are 12” foreign language recordings made in the U.S. between 1917-1924, and issued in the “E” series. Many German, Polish and Yiddish items have been reported. Top


These were set aside for 10” trials and personal records, and were presumably a continuation of the 33,000s. Although a few numbers have been reported in the low 60,000s, the first dated master is 60898, from June 1916. Given the large amount of recording Columbia did in these series, the block probably began no earlier than 1915.

Some of the recordings in this block are tests by soon-to-be-famous artists and early jazz bands, and sound quite interesting. Unfortunately the Columbia files usually do not give the titles recorded, and the parts have long since been destroyed. Copies of some may survive in private collections. Top

Table 8:

60,000-64,000s Matrix Examples

Return to Columbia Tables

60902 Jun. 1, 1916 Van & Schenck*
61070 ? W. G. “Gus” Haenschen’s* Banjo Orchestra of St. Louis (“Maple Leaf Rag”). Haenschen later changed his professional name to Carl Fenton, recording on Brunswick in the 20s.  [end p. 428]
61545/6 Mar. 29, 1917 Arthur Stone’s Jazz Band (Arthur Green)
61633 May 12, 1917 Karl Jorn
61642 May 12, 1917 (Opal) Cooper & (Noble) Sissle*
61648/9 May 16, 1917 E. F. Fuller’s Jass Orchestra*, Rectors
61659/60 May 22,1917 Frisco Jazz Band. (This was Rudy Weidoeft’s first group, which had just begun to record for Edison at this time.)
61708 Jun. 18, 1917 Rudy Wiedoefft (sic)  
61723 Jun. 28, 1917 Eddie Cantor*
61750 Jul. 2, 1917 David Belasco
61948/50 Oct. 31, 1917 Handy’s Saxophone Band
61981 Nov. 17, 1917 Duncan Sisters*
62013 Nov. 30, 1917 George J. Gaskin
62092/3 Jan. 15, 1918 Lionel Belasco’s Trinidad Orch. (“Shim Me Sha Wabble”)
62116/7 Feb. 1 & 2, 1918 Edna White Trumpet Quartet
62160 Mar. 9, 1918 Vernon Dalhart
62224/5 Apr. 4, 1918 Louisiana Five*, c/o Mr. Geo. Buck
62247 Jun. 19, 1918? Jas. A.N. Caruso Jazz Band
62251/2 Apr. 19, 1918 Don Richardson Jazz Orchestra
62254 Apr. 18, 1918 Creamer & Layton
62333 Jun. 21, 1918 Hick’s Jazz Band, Boston, Mass.
62388 Aug. 22,1918 John Steel*
62469 Nov. 1, 1918 Shelton Brooks*
62507 Dec. 19, 1918 Phil Baker

[end p. 428 in chart]

*Predates the first commercial recordings by these artists.

A particularly interesting series of “personals” was made by black tenor Roland Hayes between December 1917 and December 1918. Hayes recorded at least five classical and concert selections in this 10” series and four in the 91000 12” series, had [end p. 429] them pressed in quantity and sold them himself. These were the first classical recordings by an African-American concert artist in the U.S., and it is a commentary on the times that Hayes had to pay to have them made, since no company would support such music.

Intermixed with the foregoing, of course, were hundreds of non-professionals who paid for the privilege of making a record. Occasionally copies of these “personal records” turn up, and they are usually dreadful.

Files are missing for 62551 (January 1919) to 63911 (November 1921), and only a few scattered examples are known in that range. A series of experimental electrical recordings were made in early November 1921, by artists including Gladys Rice, Wilfred Glenn, Charles A. Prince (piano) and Weiss’ Band. The first one noted in the files is “The Rosary” by Miss Rice, recorded on November 3, 1921 (63912). Could this have been Columbia’s first attempt at electrical recording?

The only known title in the 64,000s is “Schubert Serenade” done both vocally and instrumentally on matrix 64384, which appeared on one side of the widely distributed Columbia Demonstration Record issued in late 1923 (catalog number 1-S). Recording continued in this series at least until matrix 64664 (November 1924), but contents are unknown. Top


The first 200 numbers were reserved for Polish recordings made in Warsaw, some of which were issued in the U.S. as single-faced discs and in the early “E” series. Numbers from 65200 up were London masters, although some Canadian releases (recorded in London?) are found here as well. Among the British artists released in the U.S. were Joe Hayman and Clara Butt. Top


German and Hungarian material, recorded in Europe during 1911-1913. Top


Examples suggest another “mixed” foreign language series from the early teens. Recording locations include Prague, Vierula, Warsaw and Hungary. Some of these were released in the “E” series, while some can be found in the continental “D” series. Top


Ten-inch London masters from ca. August 1917 to August 1920. Among them is “From the Battlefields of France,” recorded in March 1918 in France by General John J. Pershing (69333) and issued in the U.S. on Nation’s Forum. There is also a report of a field recording by Marshall Joffre (69331) in this series, as well as more prosaic recordings by British artists including Fred Duprez and Dame Clara Butt. Some of these were released in the U.S. in the “A” series.  [end p. 430] Top


Another mixed foreign block, which seems to have been assigned to European studios. Scattered examples reported in the 70000-70100 range are British; and from 70260-70815, Italian. Most of the latter were released in the “D” series on British Columbia or on its La Cigale operatic subsidiary. Top


Ten-inch London masters, some (e.g. Joe Hayman) released in the U.S. Top


Reserved for an unknown European studio, probably Paris or Milan. Hungarian and Italian titles, released ca. 1923, have been reported. Top


Last 10” London masters prior to the adoption of the “A” matrix prefix in 1923. Known range is 73000-73450. Top


These blocks were reserved mostly for English Columbia 12” recordings from 1916 to 1923, and are a continuation of the 6,000s. For some reason recording did not proceed in a logical, numerical sequence but jumped back and forth in several large blocks. The reader will note that the dates, if rearranged chronologically, cover the entire period from October 1916 to June 1923, however. So far as numbers are concerned, everything is accounted for except for 74400-74999 and 75500-75850. Italian and a few German/Austrian titles (all 12”) have been reported in those ranges, and it is likely that they were used by some continental studio (Milan?) prior to the advent of the English series. I am indebted to Brian Rust for sorting out the following crazy quilt.

Table 9:

74,000-76,000s Matrix Numbering Order

Return to Columbia Tables

Order usedNumerical order Recording dates 
3 74041—74400 Apr. 1920-Sep. 1921
  74401—74999 Italy (early mid-teens)
4 75000—75250 Sep. 1921-Jan. 1923
1 75251—75499 Oct. 1916-Mar. 1917
  75500—75850 Italy (ca. 1909-?)
2 75851—76850 Mar. 1917-Apr. 1920
5 76851—76999 Jan. 1923-Jun. 1923  [end p. 431]

In June 1923 the series was replaced by the AX-prefix series. Top


All known examples are Spanish language, ca. 1917-1920. Jumps to 93000. Top [end p. 431]


Twelve-inch Spanish titles recorded between 1919-1931, although the known range used is only 83002-(W)83128. Apparently there wasn’t much demand for 12” Spanish language discs. Top


Continuation of general foreign language series, recorded in the U.S. (Some masters in the 88,000s were recorded in Germany.) Scandanavian, Polish, Serbian, Russian and Yiddish titles have been reported, and there are probably others. Issued mostly in the “E” series. Jumps to 105000. Top


Educational records, which were previously included in the general U.S. series, were given their own matrix series in 1915. l1lere weren’t many educational recordings, however, and only 90000-90299 (10”) and 90500-90635 (12”) appear to have been used. Repertoire included children’s songs and stories, and music used for teaching purposes. Top


Evidently this split series of specials and personals (10” up to 91499, 12” above that) was used concurrently with the 62000s-64000s in the late teens and early 1920s. If there was a differentiation in function, it is not apparent. A few are marked in the files, “made in Chicago.” Among the more interesting items are the following. Top

Table 10a:

91,000s-Interesting 10” Items

Return to Columbia Tables

91115/6 Feb. 11/13, 1920 Paul Specht Orchestra (“Dreamy Orient”/”Wishing Land”). These appear to predate Specht’s first commercial recordings.
91308 Mar. 31, 1922 H. P. Vallee, Orono, Maine, saxophone: “Japanese Sunset.” This was of course Rudy Vallee, then a freshman at the University of Maine (at Orono), who came to New York City during a school vacation to meet his idol Rudy Weidoeft and make a personal recording. The incident is recounted in Vallee’s books, My Time Is Your Time and Vagabond Dreams Come True.
91326 Aug. 19, 1922 Frank Crumit, w/piano accompaniment, “Where Did You Get That Hat?” (Hershberg & Co., 339 5th Ave.) Perhaps a special record for a hat maker?
91458 Mar. 17, 1924 Scrap Iron Four (Lawrenceville School Glee Club, 1924), “Liza Lady” (Plantation).
91460-67 Mar. 1924 Series on “Wisdom,” “Joy,” etc., for the Right Thinking Record Company of Spokane, Washington. [end p. 432]
91468-69 Apr. 1924 Dartmouth Jazz Band, Dartmouth College, Hanover, N.H. “(a) Love for a Day, (b) Rockabye Baby Blues”; “Mother Eve.” The second title is noted as “taken from Dartmouth Show.”
91470-71 Apr. 1924 Society Syncopators of Hotchkiss, Lakeville, Conn. “Somebody Loves You After All, intro. My Fair Lady”; “Linger Awhile, intro. Stack o’ Lee Blues.”
91472-73 May 5, 1924? “Sea of Dreams”; “Ships That Pass in the Night.”
91475-76 May 13, 1924? “Join the Navy”; “Pirate Gold.” Preceding four by the Princeton Triangle Jazz Band. Dates are unclear; may be one day later in each case.

Table 10b:

91,000s-Interesting 12” Items

Return to Columbia Tables

91510 Feb. 13, 1919 Senator Towne—talking. For R&T Talking Picture Co., 1600 Broadway, New York City. Whoever they were.
91583 May 6, 1925 First of a long series made at the Western Electric labs. From here on all sides electrically recorded and prefixed “W.”
91790 Oct. 16, 1929 Paul Whiteman Orch., “Moonlight and Roses.”
91791 Oct. 18, 1929 Paul Whiteman Orch., “Southern Melodies.” Both marked “personal record for Mr. Whiteman.” Although not noted, Bing Crosby sings the vocal refrain on “Southern Melodies” and may be in the chorus on “Moonlight and Roses,” making these two of his rarest recordings. Two takes were made of each and all are marked “OK.” For an interesting article on “Melodies” see Bing magazine, September 1973, pA8.
91930-41 Sep. 1930 Electrical transcriptions for the Columbia Tele-Focal radio program, which used dubbed masters by Paul Whiteman, Ben Selvin and other leading Columbia artists.
91999 Jan. 20, 1931 Radio program for National Association of Macaroni Manufacturers. Many radio transcriptions appeared in this series. [end p. 433]



Ten-inch Spanish language items from the 1920s, probably a continuation of the 82,000s. Some may have been recorded abroad, e.g., an address by the President of the Philippine Senate issued on Columbia 3194-X (matrix 96056). For some reason the 94,000s (above 94221) appear to have been skipped. Top


Back to Appendix B: Foreign Language Recordings | Forward to Appendix C: Phantom Matrix Numbers (1901)

Columbia Master Book Table of Contents | Columbia History | Tables | Appendices

The Columbia Master Book Discography, 4 Volumes, Complied by Brian Rust and Tim Brooks. Reprinted by permission.