B: Incidental Chorus by Billy Murray. Intro Tip-Tip Tippy Canoe (Hanley)
A side (mx. 78623, take 2) recorded New York, NY, September 5, 1919.
A side (mx. 78623, take 8) recorded New York, NY, October 18, 1919.
B side (mx. 78774, take 1) recorded New York, NY, October 30, 1919.
B side (mx. 78774, take 2) recorded New York, NY, October 30, 1919.
(DAHR)
Label printing date codes: LX (= December 1919) / AW (= January 1920).
The Talking Machine World, March 15, 1920, page 239: Advance Record Bulletins for April, 1920: Mid-Month Records (i.e., released March 10, 1920)
I have looked around a bit online but have not yet found a copy with mx. 78623, take 2 rather than 8 on the A side. I assume the DAHR editors must have seen one, though, or they'd have declared the status of that take 2 "unknown."
Point taken. I've removed the note "used for represses."
On Discogs, at least, there are the take combinations
take 8 on A, take 2 on B with LX / AW date codes
take 8 on A, take 1 on B with AW date code
take 8 on A, take 2 on B with AW date code
All of these have two-letter mother letters and high stamper numbers. Columbia must have printed a lot of copies of this record in advance of the official release. Maybe they had a pile of pre-orders.
I have a copy of this record that uses mx: 78774 (take 1) on the B side. DAHR lists the status of this take as "unknown", but I can confirm that it was issued.
I'm curious about the comment concerning mx. 78623 (take 8) being used for represses. A copy of this record with date code LX (= December 1919) is listed on Discogs with take 8 on the A side, so this version was also used on the earliest pressings I've seen. Also, are records with a date code AW really represses? They were produced months before the record was released, near the end of March 1920. LX and AW labels were printed on different dates, but does that prove the records they were affixed to were pressed on different dates?