This project was by no means an easy task. Over four years in research, it traces Columbia’s chronological release of the 78 rpm albums sets released by Columbia between the late 1920′s and early 1950′s.

During the initial setting up of this project, I was certain that the project should be as user-friendly as possible because essentially most discographies are used by everyday collectors. My discography is therefore a little different in scope and layout than other discographies. I wanted to cram as much useful information into it as possible so that collectors could use my discography as a companion to their collecting activities. As a result, you’ll notice a few differences – contemporary reviews of some sets for one.

As all discography writers ask themselves, when is some information too much? During the meticulous compiling of this discography, there came a point in my research when I had to decide what to include and what to exclude. In some cases, making the right decision was critical to the integrity of the document. Take for example, Set 29 (and Set 83). In essence, I could have written a book devoted to these two sets alone.

When is some matrix information too much?

Take the example of Set 29. Set 29 is an acoustical recording released by Columbia US in 1927. It was recorded and rerecorded various times both acoustically and electrically over a period between 1923 and 1926. During the period of its release, records from the set were occasionally substituted for newly recorded takes without notice or change to the record number.

For the most part, collectors are not always interested in such complicated details.

As I got further into the project, I discovered that many recordings were missing from reference guides. It seems that Columbia did not always release a recording even though they advertised its imminent release. For example, M-62 is a case at hand.

As well, Columbia sometimes released a set in different formats and with different record numbers. Some sets were initially released under the ‘M’ series and then became part of the ‘X’ series. Often sets were released briefly and then withdrawn just as quickly – which makes it hard for researchers to track down playable copies of a recording in question. [For an example of this see page 20 for information about M-23.]

It should also be noted early in this discography that the use of the ‘M’, ‘X’ and ‘OP’ prefixes to Columbia 78 rpm sets was not a common practice until fairly late in the issuing sequence of sets.

US Columbia only used the “M” prefix after CBS bought the company around 1938. It is generally understood that the first actual “M” prefix began around the set number M-295. And to make things more difficult for researchers, I’ve noted that some very early record reviews (mainly around the early 1930′s) labelled the Columbia Modern Music Album Sets as “MM” – which was later used for the ‘drop type’ automatic record changer sequence of records – for example – MM-900.

Confused yet?

Technically, to use the ‘M’ prefix for any set issued before approximately M-295 is an error, except for a reissue/repressing of an early set. To keep with this prefixing anomaly, I have not used the ‘M’ prefix until M-295.

Some sources state that the ‘M’ prefix was introduced at set M-365. Personally, I have never agreed with this – as I have viewed sets later than M-295 with tombstone covers, clearly stating the ‘M’ prefix.

Please note, that in some cases throughout some of the indexes and notes in this discography, I have used the ‘M’ prefix for sets before 295 purely for clarity.

For each entry, I have tried; if known; to add as much detail about that recording as possible- 

  • including composition(s)
  • composer(s)
  • performer(s)
  • various record sequence numbers
  • matrix numbers
  • recording dates
  • international record numbers (and) 
  • topical excerpts from contemporary reviews

I recently participated in a somewhat heated discussion as to the quality of the Columbia US legacy. My opponent in this discussion was adamant that on average, Columbia’s US recording output was poorly recorded. I refuted this claim on the basis that his opinions were based on what he'd heard of the recordings now. I argued that his perception of what sounds good today compared with yesteryear is biased. It’s easy to listen in judgement now to a recording made; in some cases; over seventy years ago and say it sounds poor. Everything recorded that long ago sounds poor to modern-day listeners.

© S.Hopper 2010-2015 - Webspace kindly provided and funded by the 78rpm Collectors' Community - The recordings used on this educational site are for illustrative purposes.