----- Original Message -----
From: Copeland, Peter
To: [log in to unmask]
Sent: Friday, March 19, 2004 6:07 AM
Subject: Re: [ARSCLIST] Harmony label (and other things)
>>>>>    For engineering reasons, I have been compiling lists of American
matrix numbers and recording-dates (and later, with microgroove, "mastering
dates"), as the most unambiguous way of determining the ideal equalisation
curves at any time in America. This is a long way from being at a
satisfactory state; but at least I can add the following points to Dick
Spottswood's posting below.
    According to Brian Rust's "American Dance Bands on Record" (under Paul
Specht), matrix 351164 was recorded on March 21st 1932 and issued on the
Harmony label.
    But personally, I don't think the 350000 series was really a "Harmony"
series. The majority seem to be in the 140000 and 150000 series. In this
country these were issued under the British "Columbia" label, which is not
surprising because British Columbia had a controlling interest in the
American Columbia group at that time. I have hypothesised that these are
"true" Harmony matrixes, but being on the other side of the pond where US
78s don't grow on trees, I'd welcome confirmation of this idea. In which
case, the latest I have found in Rust is matrix 151507, recorded April 13th
1931 (under Britten), and this was published under Clarion and Velvet Tone -
but *not* Harmony.
    So the first "Other Thing" from my subject-box is, can anyone supply a
meaningful description of US Columbia's matrix number-blocks? For example, I
hypothesize that numbers 2000 to 2999 weren't used at all; the 47000 and
77000 blocks and the 49000 block were concurrent but the first two were
ten-inch and the third twelve-inch; and so on.
.   Another point : I don't think "The Columbia Master Book Discography"
(Greenwood Press) lists the 110000 series of "Foreign" issues, which as far
as I know were all recorded in New York. Metalwork was sent to Britain so
the stuff could be pressed for continental Europe. This is to tell you that
I am gradually piecing together some information about this series for the
benefit of the "World Music" community, and there are a number of
dating-clues on the EMI microfilms here at the British Library Sound
Archive.
    If anyone needs access to information like this, or can help me
(particularly with microgroove before RIAA standards were adopted), I would
very much like to correspond with knowledgeable people off-line. My email
address is :
[log in to unmask]
<<<<<

Specifically, I have seen it suggested that the 351xxx series was dubs of
other sides specifically
for Harmony (et al) use; however, I have no further information on this and
have never verified it.
It is worth noting, however, that most of the 351xxx sides were also issued
on Okeh with 40xxxx
matrix numbers, suggesting the 351xxx sides were at the very least taken
from different mothers.

For your general question, I believe the Columbia Master Book does list the
numbering groups,
although doesn't provide item-by-item lists for many of them. Until
Columbia(UK) went to the
A- system of numbering, the matrix groups were assigned world-wide...some of
the gaps in
US numbers, for example, represent numbers assigned in England (or, in some
cases, in other
countrues). 2000-2999 seem to have been intended for a special use, and very
few if any
were actually issued so it remained skipped. Columbia, of course, used
matrix series to
indicate both size and content. Record Research printed a list, many years
ago, of these
number groups insofar as they could identify their use...in some cases,
either so few
of a group were issued or the series appears to be an "odds & ends" series.

Also, the ledgers for the various ethnic series were apparently discarded
many years ago
and no longer exist...thus there is no source material from which to compile
a list!
Steven C. Barr