Explanation of Matrix Numbers

 [Editor's Note: This brief explanation of matrix numbers has been superseded by Howard S. Friedman's detailed article, Matrix and Catalog Numbers in G&S Discography. However, it has been retained on the site as a shorter discussion of the topic.]

Almost every 78rpm side ever recorded has a matrix number. The number was normally etched onto the ungrooved surface near the center of each side of the disc. However, some very early recordings did not have matrix numbers, and some record labels—such as Victor, for its American-made recordings—would keep their matrix numbers secret and/or obliterate the numbers that had been present on other publishers' pressings.

Matrix numbers continued to be used on LPs and even CDs, but they have special significance for 78's, as they are the one and only sure way of identifying a particular performance. Editing was essentially impossible in those days, and a side had to be recorded intact. The matrix number was stamped onto the master disc when the recording was made, and when the master was copied, so was the matrix number.

This number is vital information to collectors of old records, because:

• Sometimes, different takes were issued under the same manufacturer catalog number. Only by examining the matrix numbers would one know that the records were in fact different.

• Sometimes, the same take was re-issued under another label, perhaps even with the names of the performers changed. Only by examining the matrix numbers would one know that they were in fact the same recording.

Matrix numbers were generally assigned consecutively. One can assume that, if the matrix numbers of a set are put into numerical sequence, one will know the order in which the sides were recorded. In most numbering systems, there was a suffix on the end of the number which identified the "take". To give a simple example, consider the 1927 Trial By Jury. Here are the details of that set, arranged in the order recorded:

1927 Trial By Jury
Side
Nbr
Matrix
Number
SelectionRec.
Date
1Cc11602-2 Hark, the hour of ten is sounding20 Sep 27
2Cc11603-2 When first my old, old love I knew20 Sep 27
4Cc11604-2 Swear thou the Jury!20 Sep 27
8Cc11605-2 The question, gentlemen, is one of liquor20 Sep 27
7Cc11606-2 A nice dilemma we have here20 Sep 27
3Cc11644-2 For these kind words accept my thanks29 Sep 27
5Cc11645-2 Oh never, never, never29 Sep 27
6Cc11646-1 That she is reeling is plain to see!29 Sep 27

We see here that five sides were recorded on September 20th, 1927, the remaining three on September 29th. The "-2" suffix on all but the last matrix number indicates a second take.

Sometimes, examination of the matrix numbers and recording dates allows one to piece together some of the circumstances under which a set was made. For example, the 1930 Patience was recorded on 22 sides, and 21 of them were recorded between September 24th and October 8th, 1930. However, side 13 ("I hear the soft note...") was recorded on November 9th. Its matrix number is Cc19798-6, suggesting that six takes were needed. Normally, one would not schedule a session to record just one side, so it's safe to assume that the original sessions failed to produce an acceptable take, so the company had to be assembled to redo it.

A similar case is presented by the 1929 Pirates, where details are as follows:

1929 Pirates
Side
Nbr
Matrix
Number
SelectionRec.
Date
1Cc15910-9A Overture, part 115 May 29
2Cc15921-1A Overture, part 221 Feb 29
3Cc15911-5 Pour, oh pour the pirate sherry19 Apr 29
4Cc15943-5 Oh, better far to live and die25 Mar 29
... ... ... etc. ... ... ...

Every side of this recording, except for 1 and 3, was recorded between February 21st and March 25th, 1929. The unusually high number of takes for sides 1 and 3 suggests that some problem was encountered, forcing the company to return to the studio at a later date.

The seven operas D'Oyly Carte recorded in 1949-50 (the mono sets with Martyn Green et al) were originally issued on 78rpm records. However, Decca was by now using a different system, and the sequence of matrix numbers no longer signified the order in which the items were recorded.

Marc Shepherd, oakapple@cris.com