You are right, of course. I thought I'd seen more of the Century Expanded numbers than of the Cheltenham ones when I scrolled trough the releases we have, but now I see it's the other way around. And in fact, I now see that you'd declared a 1929 label to be "typical East Coast with fonts from the Cheltenham family" at https://www.45worlds.com/78rpm/record/4418us&rc=355428#355428
Funny, I'd've thought the Cheltenham {Images #3630345 & 3630346} was East Coast and the Century Expanded {Images #3630091 & 3630092} was West Coast, but . . .
The catalog numbers of the East Coast and West Coast versions (#1 and #1W) are set in very different fonts. The East Coast version numbers might be Century Expanded; the West Coast maybe a type of Cheltenham? I've written to W.B.lbl to help us identify the fonts for sure. Based on the font similarity, though, #2 looks also like a West Coast version to me.
Variant #2 (from 1932) lists the matrix number in the runout: ?35881ᴬ 4. The ? stands for what may be a small letter that I can't read,
Variant #3 (from 1934, with the inner pressing ring) lists the matrix numbers both on the labels and in the runouts. In the A side runout, it may again be preceded by some other letter: ?35881ᴬ 9 ⍑. The B side runout also has something like ⍑ in the 9 o'clock position. Couldn't read the rest on the original uncropped images.
Br 6105 - 5 label images added of 3 additional label types (#1W AB, #2 A, #3 AB; #1 & 4 already present).
#1 - Br Radio Corp., Subs. Warner Bros (at bottom).
#1W - (presumably) West Coast variant of #1, with very different layout.
#2 - Br Radio Corp. (Warner Bros. omitted)
#3 - Br Record Corp.; Patent nr under label name. Not licensed for radio broadcast. Title & credit serifed.
#4 - Br Record Corp.; no patent nr under label name. Title & credit sans serif .
Note that on West Coast pressing #1W everything is different: fonts of all lines, incl. cat.nr; 2-line description at 3; Spanish title in italics; layout of credit. It is possible that #2 is a later West Coast label. Types #1-3 have on the B-side an underlined cat.nr.
Note also that #1 is among the very first releases with this new label type. The last known release with the previous label type is 6111; this type has only a patent nr under the label name in small print and a different bottom legend.
"Nobody's Sweetheart" is the A side after all. The original 1931 release marks "St. James Infirmary" as the B side with a dash in the runout, see the images here.
Looking at it again, I've come to the conclusion that these labels definitely belong to a later pressing (1935 or after) because the original Brunswick labels from mid-1928 to 1932 didn't have matrix numbers on them and the line under the cat# of one side (you may call it Side B, Side 2 or whatever) appeared until the end of 1934, see e.g. 6917. As far as my spot checks go, the line was always on the side with the higher matrix number, without any valuation.
What I forgot to mention: At that time Brunswick actually marked the B-side (usually the one with higher matrix number) with a line under the cat#, see e.g. the labels of the nearby numbers 6074 and 6211 and my comment on Brunswick 4302. In contrast, the displayed labels here are not marked like that, possibly because they come from a later pressing.
Using the matrix number in that context makes more sense. Back in the days before computers and data bases, one of the favorite pastimes of discographers was designing discographical data sheets. A single sheet was supposed to serve to fully document one record. On those sheets, as it is on the database, it was helpful to have a shorthand means of referring to each side for purposes of discussion. Most defaulted to using side A and Side B, although some went with side 1 and side 2. Those who were set on having thoroughly uniform reporting proposed that the first slot in the sheet be allocated to the track with the lower matrix number. I personally still prefer that the sides retain whatever designation they carried on the disc, but that would probably become chaotic in a short period of time when applied over the whole of a database.
Hi Steve, I completely agree (with all three comments of yours, by the way). I don't care too much about A & B sides either. Records were expensive then and both sides had to be of value for the buyers (especially the juke box operators) - like both sides of an LP some decades later. Still we have to decide which side has to be "A" and "B" when we add them here, and if there is no other information on the labels, people tend to follow the matrix numbers, what I didn't here but wanted to explain.
The Betty Boop clip with Cab singing for Koko the clown is a blast. I never get tired of it. The cartoonist captured the essence of the way Calloway moved when he was performing. Clearly, the cartoon was done before the Hollywood standards committees got going. Ghostly Cab drops his pants at one point, and demands a shot of booze at another. Most of us who saw these cartoons during our college days had the same question: What were these guys on? It may have been nothing more than free-flowing imagination, but somehow I doubt it. Setting aside its questionable suitability for children, however, it is a masterful blend of music and animation.
Before he wrote Minnie the Moocher, Cab Calloway used "St. James Infirmary" as his theme song. It may be that the record company felt that his fans would want to have the song available, but that the familiarity factor would work against it's being the bigger hit. In retrospect, however, "St. James Infirmary" is much better suited to Cab's singing style, and the arrangement is better suited to Cab's powerhouse swing band than the somewhat dated treatment of "Nobody's Sweetheart". The latter has the feel of a mid-to-late twenties Paul Whiteman effort. I'm guessing the A&R guy picked it, and he didn't know much about the Calloway band. He probably would have gotten a better track if he'd let the band do a head arrangement of it. "Nobody's Sweetheart" was (and still is) a standard among hot musicians.
By now, most of you who care about the A side/B side decision probably know that I don't (particularly care about it, that is). So you can ignore my opinion, if you want to, and I won't take it personally. To the extent that there ever was any importance attached to the side designation, however, I am not convinced that it was ever tied to the assignment of matrix numbers by deliberately assigning numbers to the planned A side first, and only then proceeding to the B side. It is probable that some A sides got lower numbers because the A&R guy generally knew he was going to record the song when the session started, while some B side songs were only selected as the session proceeded. To the extent that A sides naturally got the first and most attention, they would tend to get the lower numbers, too, but I wouldn't rely on it as a rule to indicate which track the company valued more. The only thing the sequence of matrix numbers can more or less reliably indicate is the order in which the first takes of each track were recorded. A typical recording session at which four tunes were to be recorded might as easily have started with the B sides to get warmed up and get them out of they way, then moved to the serious challenge of locking down a hit or two.
Although the matrix numbers suggest "Nobody's Sweetheart" as the A-side and the linked German release explicitly follows the numerical order, I have decided for "St. James' Infimary" because it was the more popular side, confirmed by the A-side designation on the reissue of this coupling in 1943 on Brunswick 80018. It may have been meant as the top side here too, eventually having provided the intro of Cab's previous big hit, "Minnie The Moocher".
Videos of both sides are already on the German issue, submitted by xiphophilos, so I only have to copy them here:
Nobody's Sweetheart:
St. James' Infirmary (in fact, a different video because the one by xiphophilos doesn't work for me):
St. James' Infirmary (Cab Calloway singing as Koko the Clown in a 1933 Fleischer Bros. clip)