Appendix E: Discographical Sources

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


The author has long maintained that discographers should be more specific about their sources. It is one of the scandals of modem discography that many compilers give the reader virtually no clue as to the origin or reliability of their data. As a result mistakes live on, and in some cases thrive, in subsequent literature. Some have criticized this position, maintaining that it is “impractical” or simply too much work (for them), but having taken the position I feel obligated to try to live up to my own standards. The following is an attempt to provide background on the sources of the data found in the discographical sections of this book. See also the Bibliography, which contains information on what was drawn from specific secondary works.

Entries in the discography are of two general types.

1. Where a take number is shown (e.g., “4000-3”), actual copies of the records listed have been seen by the author or reported by one of the many persons who contributed data over the years. Records have been kept of who contributed what data, although for reasons of privacy that in information is not given here. In a very few cases (generally foreign issues), entries with take numbers may come from published discographies listed in the Bibliography.

2. If no take number is shown (e.g., “4000-”), the corresponding data may have come from one of the following sources, in approximate order of likelihood:

a. Original catalogs and supplements or release lists published in the Columbia Record, and other contemporary sources such as Sears’ catalogs. All known Columbia catalogs and most supplements published between 1901 and the 1920s have been examined, along with those of many other companies that used Columbia masters. A ca. 1902 single page list of Climax discs, presumably from a retailer catalog, has been particularly helpful since it contains artist identifications for many early discs. In the case of single-faced records the matrix number and catalog number are generally the same, making identification from catalogs relatively easy. Double-faced records usually bear a different catalog number, and in those cases the matrix shown here has been assumed to be the source only if the label habitually used Columbia masters and title and artist match.

b. Actual copy seen but the take number was not visible in the wax, or was not reported by the contributor. Absent takes are most common for very early issues, especially on Climax, which almost never showed a take number. If aural information is given for the disc (for example, an announcement), then of course the disc has been seen.

c. Auction lists published by modern dealers. Occasionally these include matrix and take.

d. The matrix lists compiled by Helene Chmura at Columbia Records during the 1950s and 1960s (see Introduction). These are very incomplete for pre-1910, and contain a good many errors, and have been used only when there was no other source.

e. Previously published discographies (see Bibliography).

In a few cases a note indicates that a title has been inferred from “parallel cylinder releases.” This refers to the fact that during certain periods Columbia released all titles on both cylinder and disc, in the same numerical sequence in each format. Therefore a title that is untraced on disc can sometimes be inferred from the title on the corresponding cylinder, if there was one. Since the disc version was probably unissued this is merely an inference, and is noted as such.

Artists have been identified from the spoken announcement, the label, or original catalogs, in that order of priority.

Titles are from the same sources, although errors and eccentricities in Columbia’s spelling and syntax have been corrected. The goal has been to reflect what was actually on the record, not necessarily what Columbia called it (major differences are noted). The following musical reference books have been consulted to determine proper spelling, as well as shows and composers: Bordman (1978), Burton (1962, 1965, 1969), Cohen-Stratyner (1988), Charosh (1995), Delaney’s Song Book, Fagan and Moran (1983,1986), Fuld (1995), Jasen (1973), Kinkle (1974), Mattfeld (1962), Spaeth (1948), Talking Machine World. See the Bibliography for full citations.

Spoken announcement information is from actual copies heard. Announcer identifications are tentative, and attempted only in those cases where the voice seemed readily identifiable.

Recording dates, present in only a handful of cases, are from the Columbia files or other source as noted.

Release dates are defined as the date of the monthly supplement in which the record first appeared. Only the first release of each master is dated.


Back to Appendix D: Special Catalogs | Forward to Appendix F: Bibliography

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.