Victor Master Numbering Systems

The original Victor identification system was begun with No. 1 on April 23, 1903, and (with occasional modifications) was in continuous usage until August 1936, when the basic system had reached 104,075. At that time, a new block of numbers was established, beginning with 00 and reaching 09999 by February 1938 and 097503 in August 1958. Effective August 15, 1951, a new and much more complex matrix numbering system was introduced (revised December 10, 1954).

The matrix system that originated in 1903 consisted of three parts: an initial letter, followed by a dash and a “serial number,” followed by a dash and a “part number,” which we today call a “take.” The initial letter indicated the finished record size: A = 7”, B = 10”, C = 12”, and D = 14”, as well as E for an 8” record size introduced in 1906 and abandoned in 1909. Starting in the 1920s, the initial letter or letters began to be used for further information in addition to the record size. The place of recording could be indicated (when other than Camden); thus, PB and PC indicated 10” and 12” recordings made on the Pacific Coast (first in Oakland, later Hollywood), and the initial letters BA were used for 10” records made in Buenos Aires. When electrical recording was introduced in 1925, the letters “VE” were added to indicate the use of a Western Electric recording head; thus, 10” electrical recordings were designated “BVE,” 12” electrical recordings as “CVE.” Electrical recordings made in Brazil were “BBVE” and “BCVE”; those on the Pacific Coast were “PBVE” and “PCVE.” When RCA recording heads were used in place of Western Electric, the prefix became “BRC” and “CRC.” “BRCHQ or “CRCHQ” indicated a 10” or 12” “high-quality (fidelity)” RCA cutting head, while a designation “BSHQ” or “CSHQ” showed that a “semi-high-quality” head was used. An “L” meant “long play” (33.33 rpm); thus, “LBVE” or “LCVE.”

The chronologically assigned “serial number” remained the same for further takes of the same selection by the same artist even if made many years later, as long as the accompaniment remained the same. If a selection originally made with piano accompaniment was re-made with orchestra, it was assigned a new matrix number and began again with take 1. Thus, C-2980-1 was a recording made by Schumann-Heink on January 3, 1906; C-2980-2 was a re-make of the same selection on October 18, 1909; and CVE-2980-3 and -4 were electrical re-makes recorded June 26, 1930, all with orchestra. Parts for all these different versions, even including some unpublished takes, might well be found together in the vault. Original recording sheets and cards in the catalog-number (“historical” or “blue” or “green”) cards in New York are supposed to show by the letters “d,” “h,” or “m” which takes were destroyed, held, or mastered, but these notations cannot be trusted to reflect what was actually published and which masters were retained long-term. Sometimes a notation such as an “h” changed into a “d” (on a given date) is found, and that may have been the intention, but the take marked “h” may have been retained. Often, a take actually issued and replaced in the catalog by a later take was changed from “m” to “d” in the card files. Actual copies of published recordings said in the card files to have been destroyed are often found in collections and archives, and parts for them sometimes exist in SonyBMG archives.

Until about the middle of 1906, the first two components of the matrix number (initial letter and serial number) were hand-written in the original wax and appear on pressings in the 6 o’clock position (stamped catalog number at 12 o’clock). The first take usually was not marked, but subsequent takes from number 2 onwards were numbered and (until about mid-June 1907) could be seen hand-written in the center above the center hole, with the take to the left of the center and a brief title and notation of the artist below the center. All this information was placed so that it was covered by the label cutout and thus not visible on the finished pressing, but it is visible, of course, on the metal parts. Still later, the take number was often imprinted outside the label at 9 o’clock.

In November 1912, it became the practice to stamp a very tiny letter “R” in the inner rim at 3 o’clock to indicate a new take that replaced an earlier one on a record with the same catalog number. I have never seen this mark mentioned in Victor literature, and it is not mentioned in any of the existing files found in New York.

Late in 1912, Victor experimented with a groove enhancement program that smoothed and deepened the original record groove on masters whose grooves had rough edges. I have found no mention of this in the Victor files, nor do treated masters have any special mark showing that this has been done. Pressings made from such treated masters can be identified by sight by some rather subtle indications. We first learned of this process from correspondence found in the files of the Gramophone Co. in England, as this company participated in Victor’s Camden research.

In 1915, Victor developed an acoustical dubbing process to create new masters from pressings where damage had occurred to the originals. Such dubbings are marked with the symbol “s/8” stamped in the inner rim. These are occasionally (but not always) noted in the New York files. Pressings made from these dubbed masters are sonically inferior to the originals.

During the electrical period (i.e., after 1925), a tiny “R” was placed under the take number (9 o’clock) to indicate an electrical dubbing. Such dubbings do not appear to be noted in the New York files. This should not be confused with the “R” at 3 o’clock used in the acoustical period for take substitution, as noted above.

Beginning in 1905, Victor started sending recording teams to Asia and Latin America. Wax plates were shipped back to Camden for processing, and finished pressings, often of local interest only, were shipped to their country of origin for sale. Letter prefix numbers were assigned to each overseas trip and the resulting recordings were numbered serially for that trip, which perhaps would start in Mexico, go to Cuba, then down the west coast of South America, across to Buenos Aires, and back by Rio and the Caribbean. Paper documentation of these field trips is only fragmentary today.

Special Recordings and Tests: Over a period of 50 years or more, Victor made hundreds of pressings of artists’ trials, including artists who recorded for other companies or are not known to have left any commercial records. Up until the electrical period, test pressings we have seen appear to carry no numbers, unless some identification was written (as sometimes happens) on the exterior ring and thus does not appear on pressings. After about 1925, many (but not all?) “trial” test pressings were serially numbered, preceded by “T”; sometimes they also carry conventional matrix identification, such as BE-T-#### or CVE-T-####.

Known matrix numbers of Victor origin from 1903 through the mid-1950s number some 235,238. HMV and other imported masters probably bring the possible total to 300,000. This number probably represents close to the total number of matrix entries that will eventually be listed, mostly with backup material, in the Encyclopedic Discography of Victor Recordings.

William R. Moran
1995
Edited for inclusion in the DAHR website