“Tuxedo Junction” Erskine Hawkins (1939) and Glenn Miller (1940)

“Tuxedo Junction”

Composed by Erskine Hawkins, William Johnson and Julian Dash; head arrangement organized by William Johnson.

Recorded by Erskine Hawkins and His Orchestra for RCA-Bluebird on July 18, 1939 in New York.

Erskine Hawkins, trumpet, directing: Sammy Lowe, first trumpet; Marcellus Green, Wilbur “Dud” Bascomb, trumpets; Bob Range and Edward Sims, trombones; William Johnson, first alto saxophone; Jimmy Mitchelle, alto saxophone; Julian Dash, tenor saxophone; Haywood Henry, tenor and baritone saxophones and clarinet; Avery Parrish, piano; William McLemore, guitar; Lee Stanfield, bass; Jack Morrison, drums. Solos: Erskine Hawkins first muted trumpet solo; Bill Johnson alto saxophone;  Wilbur “Dud” Bascomb open trumpet solo; Haywood Henry, clarinet; Erskine Hawkins, second muted trumpet solo.

The story of the Erskine Hawkins band is a story of brotherhood. Brotherhood in the literal sense because two of the best jazz soloists in the band were brothers—tenor saxophonist Paul Bascomb (#), and his trumpet-playing younger sibling Wilbur “Dud” Bascomb. And brotherhood in the larger sense, of friendship, commitment, mutual respect, joint sacrifice, and loyalty among the men who made up the Hawkins band.

What eventually became the Erskine Hawkins band in the mid-1930s evolved from various groupings of musicians who came mostly from Birmingham, Alabama in the early 1930s. Their youth, strength, idealism, wisdom, musical talent and solid musical training were bonds that held them together for many years. In a rigidly racially segregated society where often brutal segregation was rampant and legally (and illegally) enforced, and in a music business often run by dishonest promoters and businessmen, the Afro-American Hawkins band survived somehow, and eventually thrived. Their slowly-building success was yet another bond that held them together. (Above left: Erskine Hawkins – early 1940s.)

The remarkable story of the Erskine Hawkins band began in the Gothic Deep South of the 1920s, with all of the implications inherent in the term “Jim Crow,” including the concept later enunciated in the landmark U.S. Supreme Court case Brown v. Board of Education, that racial segregation as practiced then in the U.S. resulted perforce in separate and unequal accommodations and facilities for the races. Whites and blacks lived separately in white and black worlds, and the black worlds were inferior in almost every material sense to the white worlds. The result was continuing widespread poverty and ignorance in black communities, and separate and unequal employment and educational opportunities that all but ensured ongoing poverty and ignorance among the vast majority of Afro-Americans in the Deep South.

What the Jim Crow enforcers of racial discrimination could not degrade or destroy however was the pride of African-American communities, and the burning desire of many blacks to better themselves, despite the long odds against any such self-improvement in the Deep South. There was in the 1920s a segment of the black population of the Deep South that resolved to break out of poverty by creating opportunities for Afro-Americans to excel. The leaders of these communities established black organizations of all sorts that would allow members of their communities, especially the young, to participate in the same activities as whites. These leaders for the most part were not nationally known, but were local people who were strong-willed visionaries who were looked up to by local black people. The organizations they created, which were almost always chronically underfunded, nevertheless became havens for black youths seeking a path to a better life. Paramount among these were black schools run by idealistic and often charismatic black educators, where knowledge of many different trades and arts would be imparted to the young. A number of these visionary leaders would play fundamental roles in the story of the development of what eventually became the Erskine Hawkins band.


The music: As was so often the case during the swing era, when a band started a relationship recording with a record label, especially one as prestigious as Victor/Bluebird, many people who knew little about that band’s music were suddenly in a position to decide what kind of records that band should make. The people at Bluebird had little understanding of the musical orientation of the Hawkins band when it began recording for that label in 1938. Moreover, so much activity was going on at Bluebird then recording the many other new swing bands that had recently been signed by the label, that a great deal of thought was not invested in what would be a good fit musically for the Hawkins band in terms of material to record. The automatic default then for all A and R supervisors (later called record producers), was to make bands record current pop tunes. (Above left: L-R: Sammy Lowe, Julian Dash, Erskine Hawkins.)

When Erskine Hawkins and his band learned that the four tunes they were assigned to record on their first Bluebird session were all current pop tunes, they were disappointed. Nevertheless, with the same positive spirit that had propelled this group of musicians through earlier travails, they arrived at a stratagem that would give them a chance to record music they believed showed them in the best light. When the band arrived at the Victor recording studio, Hawkins asked the recording supervisor if his band could record a couple of its special arrangements during the recording session where they were to record four pop tunes. He was told that they could, but that few bands were able to record more than four tunes in a three hour recording session. Hawkins replied that he understood that. He and his band then efficiently recorded the four pop tunes assigned, and then recorded the two arrangements (“Weary Blues” and “King Porter Stomp”) he and his band wanted to record, all within the allotted three hours. (This bit of information was given to me in a conversation with Hawkins’s long-time saxophonist/clarinetist Haywood Henry when I encountered him at a jazz club in Manhattan in the late 1970s.)

That the Erskine Hawkins band was well-organized, disciplined, and prepared is evident from the recordings they made. Their first two Bluebird recording sessions in 1938 yielded twelve acceptable masters. As stated above, most bands were happy if they recorded four masters per session. These early recordings document the solid musicianship of the Hawkins band. Clearly, the stable personnel, strongly individual soloists, and talented arrangers (including band members Bill Johnson, Sammy Lowe and Avery Parrish) who were contributing original compositions to the band’s library, were beginning to give the Hawkins band a recognizable musical identity.

The Bluebird records the Hawkins band made in the first half of 1939 show the band improving, and acquiring a bit of assurance. Their first hit, “Tuxedo Junction,”(1) was recorded on July 18, 1939. It was released around September 1, and by early 1940 it replaced “Swing Out” as the band’s opening theme song. By the end of 1939 it became clear that its sales were far beyond the average sale of other Hawkins disks. But the metric for a hit record in 1939 was much different from what it became later. If a record sold 10,000 units in 1939, it was considered a success. Anything beyond that ranked it as a hit. Within these parameters, the Hawkins band’s recording of “Tuxedo Junction” was a hit by the end of 1939.

Despite the assertion of the Hawkins band’s jazz trumpet soloist Wilbur “Dud” Bascomb that Hawkins’s lead alto saxophonist Bill Johnson was “the guy who wrote ‘Tuxedo Junction,’” there is more to that story. Dud himself later asserted, “…the last chorus was mine.”(2) However, the composer credits for “Tuxedo Junction” read: William Johnson, Erskine Hawkins and Julian Dash. No composer credit has ever been given to Wilbur Bascomb. Erskine Hawkins himself remembered how “Tuxedo Junction” came to be “composed.” “I needed one more tune to record at that session. I told them ‘give me about a half hour.’ So I went in to tune up with Bill Johnson, and I took the rhythm section and they put it together.” (In the Mainstream…18 Portraits in Jazz by Chip Defaa, (1992) Scarecrow Press, Pages 142-145. Hereafter (ITM/page reference.) I find it interesting that Hawkins said they put it together, instead of we put it together. (Above left: Julian Dash.)

Also, Hawkins does not mention Julian Dash, the third listed co-composer of “Tuxedo Junction.” Fortunately, Dash gave his recollections about how “Tuxedo Junction” came into being to a journalist around 1970.(3) Dash remembered that “Tuxedo Junction” began as an unnamed head arrangement that the Hawkins band played on at least one theater engagement, and then at Harlem’s Savoy Ballroom, as the sign-on for radio broadcasts. When the Hawkins band needed another tune at (the recording session mentioned above), Hawkins, Bill Johnson and Dash began shaping-up the unnamed head arrangement. Then Bill Johnson, an accomplished arranger, wrote out enough music for the band to play so the recording could be made. (Above right: Bill Johnson in 1940.)

When listening to the Hawkins recording of “Tuxedo Junction,” one notices immediately that solos make up the vast majority of the performance, with simple backgrounds and melodic fragments placed strategically throughout making up the rest. In other words, there was not a lot of writing done for the ensemble.

Usually, with an original whipped-up casually by members of a swing band, composer credits were not worth that much in dollars. But if a tune took off, there could be litigation over composer credits. That never happened with “Tuxedo Junction,” so I think it reasonable to assume that the composer credits actually identify those who were involved in creating the tune.

Hawkins continued: “The band recorded it. They then said: ‘What are you going to name it?’ I asked for suggestions from everybody. My valet said, ‘Why don’t you call it “Tuxedo Junction” since that’s where you’re from?’ And that’s how it got its name.” (ITM 145)

Everyone connected with the Hawkins band was delighted with the success of “Tuxedo Junction.” While its brisk sales generated some much-needed cash for its composers, the financial ramifications of “Tuxedo Junction” were not that great at first. Then something happened that no one could have envisioned, and what had been a modest hit record was transformed in the next year into a mega-hit, and an annuity for its composers. Hawkins remembered: “Glenn Miller came in as a guest band at the Savoy. He and I alternated on a Sunday afternoon” (ITM 145)Note: Hawkins’s memory is substantiated by John Flower in his book “Moonlight Serenade…a Bio-discography of the Glenn Miller Civilian Band” (1972) Arlington House, page 108. The date when these events happened was December 24, 1939. “So when we finished, Glenn said, ‘Hawk, I want to talk to you.’ So we went in the manager’s office and sat down. He then told me that he wanted to record ‘Tuxedo Junction,’ but that RCA Victor told him that he had to see me on account of two bands couldn’t make the same tune on the same label. So he gave me his idea. He said, ‘You make it your theme so that people know that it’s still your number. I’ll play it as a regular number, but you’ll still have it as your theme.’  And that’s what I did. I also gave him the OK to record it.” (ITM 146). And he said one more thing to me: ‘Don’t let anyone record “Tuxedo Junction” for six months after I record it.’”(ITM 146).

(#) Paul Bascomb was not present in the Hawkins band when it recorded “Tuxedo Junction,” but was present throughout the 1930s and intermittently after that.


“Tuxedo Junction”

Composed by Erskine Hawkins, William Johnson and Julian Dash; arranged by Glenn Miller and Jerry Gray.

Recorded by Glenn Miller and His Orchestra on February 5, 1940 for RCA-Bluebird in New York.

Glenn Miller, first trombone, directing: John Best, first trumpet; Clyde Hurley, R.D. McMickle and Leigh Knowles, trumpets; Paul Tanner, Frank D’Annolfo and Tommy Mack, trombones; Hal McIntrye, first alto saxophone; Wilbur Schwartz, alto saxophone; Al Klink and Tex Beneke, tenor saxophones; Jimmy Abato, alto and baritone saxophones; J.C. MacGregor, piano; Richard Fisher, guitar; Rowland Bundock, bass; Maurice Purtill, drums.

The story continues: At the very time Glenn Miller was having his conversation with Erskine Hawkins, he (Miller) was about to commence his sponsored network radio series (Moonlight Serenade on CBS, for Chesterfield Cigarettes), which would provide Miller with a powerful promotional tool for increasing the popularity of his band for the next 2 ¾ years. His first Chesterfield broadcast would be on Wednesday December 27, 1939. Miller’s canny use of the Chesterfield radio show to promote not only his band, but specific tunes recorded by his band, would yield numerous hit recordings for the Miller band during their entire run on the Chesterfield show, which lasted until September 1942, when he broke up his band to enter the Army Air Force.

There is another aspect of how Glenn Miller became interested in “Tuxedo Junction.” It was recalled by Miller’s pianist Chummy MacGregor on page 8 of the transcript he prepared for publication around 1970. (It has not been published to-date.(4) The essence of it is that Miller’s musicians were favorably impressed by the Hawkins band’s performance of “Tuxedo Junction” that night at the Savoy Ballroom. During an intermission, Hal McIntyre, Miller’s first alto saxophonist, got a lead sheet (very simplified sketch of the tune) from one of its composers (probably from Hawkins’s lead alto saxophonist Bill Johnson). This was then given to Jerry Gray, Miller’s chief arranger, who worked up a very simple arrangement that he brought to a Miller rehearsal. MacGregor said that the band then added their own ideas, and voila! …one of the classics of the swing era was created.

The music: I suspect strongly that Glenn Miller’s own musical ideas guided the overall musical design of what eventually became his band’s arrangement of “Tuxedo Junction.” By the time success came to Miller, he was no longer writing arrangements for his band; he was editing arrangements written by others. And he was a great editor. The most obvious difference between the Hawkins band’s approach to “Tuxedo Junction” and the Miller band’s approach, is that the tempo of the Miller version is much slower. Indeed, it is insinuating. The Hawkins band performed “Tuxedo Junction” as a jaunty, casual-sounding framework, mainly in 2/4 time, for numerous jazz solos. Miller’s conception was much more deliberately calibrated, though minimalistic, and in 4/4 time, with the entire arrangement building slowly and teasingly to a powerful climax. Miller’s recording of “Tuxedo Junction” is a perfectly paced and balanced assemblage of very few musical parts.

The central musical device in the Miller arrangement is the tantalizing use (on the “A” section of the tune), of the low register trombone oo-ahs, created by the players flexing (oo),and then extending (ah), a held plunger in front of the bell of their instruments, playing at a very low dynamic level, almost at a whisper, developing tension, followed by explosions of open brass. The unison saxophones are also a fitting touch in this musical context. The trumpet solos in the Miller performance are by Mickey McMickle, using a Harmon mute, and Clyde Hurley on open trumpet, whose Beriganesque playing is soulful, with a couple of tasty low-register rasps.

John Best plays the first trumpet part in a most relaxed manner, scooping notes just like he did when he led Artie Shaw’s brass section before he joined Miller. Best rarely played lead with Miller because Miller gave him many solos, and because Miller generally preferred Mickey McMickle’s on-the-beat manner of phrasing the lead parts. Also, McMickle played few solos.

The first Miller performance of “Tuxedo Junction” that is documented was on the Chesterfield show on February 1, 1940. He recorded “Tuxedo Junction” for Bluebird on February 5, 1940,(4) and promoted it ceaselessly on his radio show even before his recording of it was issued. By the time it was issued in March, it was already a hit. By the summer of 1940, “Tuxedo Junction” had become a major hit, eventually selling well over one million Bluebird records. (This was just in the 1940-1942 period. Since then, Miller’s recording of “Tuxedo Junction” has sold many more millions of records. Additionally, “Tuxedo Junction” was recorded by many other artists (a lyric was soon added), generating a lot of royalties for its composers Erskine Hawkins, William Johnson and Julian Dash.)

Lightning striking twice? For what it’s worth, I have always wondered about the musical relationship between “Tuxedo Junction,” and the deathless novelty tune “Three Little Fishies.” Glenn Miller recorded “Three Little Fishies” for Bluebird on April 10, 1939. That recording, clearly intended to be a comedy-novelty, is hardly a swing era classic, though it does feature Tex Beneke singspeaking in a language that is a strange variant of English. The Miller recording sank quickly in 1939, while the version recorded two days earlier by Kay Kyser, which is even more absurd, rocketed to hit status. I think it at least possible that when Glenn Miller heard the Esrkine Hawkins band play “Tuxedo Junction” at the Savoy Ballroom, he began thinking: Hmmm, “Tuxedo Junction” is based in part on chords similar to those for “Three Little Fishies;” Kay Kyser has sold a million records with his version of “Three Little Fishies;” the public obviously likes something about “Three Little Fishies;” maybe I can repackage “Tuxedo Junction” in an entirely different way and …..??? If that was Miller’s thought process, he was obviously correct.

One final question: Were the members of Erskine Hawkins’s band aware of the first recorded version of “Three Little Fishies,” made in early 1939 by Hal Kemp’s band (which included its composer, Saxie Dowell)?

The recordings used in this post were digitally remastered by Mike Zirpolo.

Notes and links:

(1) Tuxedo Junction was the name of a ballroom in Birmingham, Alabama located at 20th Street and Ensley Avenue.

(2) The World of Swing by Stanley Dance, (1974) Scribner’s, page 197.

(3) The Swing Era 1940-1941, Time-Life Books (1970), page 57.

(4) A first or alternate take of “Tuxedo Junction” exists. It is almost identical to the issued take, except that trumpeter McMickle had a bit of trouble with his solo on take one, which he corrected in take two.

(5) I thank Glenn Miller aficionado Rob Ronzello for graciously providing me with a copy of that transcript.

Note:I must say that Glenn Miller’s Bluebird recording of “Tuxedo Junction,” for whatever reason, contained an unusual amount of extraneous noise, including hiss and hum. I did everything I could to reduce that noise, but unfortunately could not eliminate it completely without degrading the actual musical sounds in the recording.


As a special bonus, I am presenting here an unissued aircheck recording of “Tuxedo Junction” by Erskine Hawkins and his band from the Savoy Ballroom in Harlem taken from an NBC Blue network sustaining (unsponsored) broadcast which took place from 5:45 p.m. to 6:00 p.m. on Saturday January 27, 1940. As you will hear, Hawkins was by this date using “Tuxedo Junction” as his band’s opening theme song on radio broadcasts. The personnel of the Hawkins band is the same as noted above.

I must thank Karl Pearson of suburban Chicago, for this unique recording. It was digitally transferred by him from the original transcription disk. I did the clean-up of that transfer, and digitally remastered this recording.


P.S. For those who can’t get enough of “Tuxedo Junction,” I have posted at bunnyberiganmrtrumpet.com two other excellent versions of it, one a Tommy Dorsey broadcast from mid-1940, and another a Bunny Berigan broadcast from October of 1940. The TD version is in a Sy Oliver arrangement which includes a great trumpet solo by Bunny Berigan, and the Berigan version has Bunny and his band playing the same chart used by the Erskine Hawkins band, and it features more excellent Berigan trumpeting. These performances include two quite different jazz solos by Berigan. Here is  a link to them: here.https://bunnyberiganmrtrumpet.com/2017/11/28/tuxedo-junction-1940-two-different-versions/

Related Post


  1. Re Miller’s “Tuxedo Junction” and its “alternate take” — BS 046786-1 was the first issued take, on gold-print and early brass-print (not “silver”) copies of Bluebird B-10612-A. No take digit appears in the wax. This is the one where McMickle’s muted trumpet solo ends late; it also has *two* drum hits at the end of the second, quieter repeat (2:42). BS 046786-2A was released in 1942 (around March), also on brass-print Rings-label copies of B-10612-A, and was used on all reissues both 78 and microgroove (in some cases master-pressed, such as on some copies of Victor 20-1565-A; on these, ‘2A’ clearly appears at 9:00 in the wax with the catalog number at 6:00). This is the take most familiar to us, with McMickle’s solo ending neatly in time and with just one drum hit. That drum hit was so loud that the master had to be dubbed (on most copies) so as not to cause the stylus to jump. I think that’s actually the first take cut, and was likely kept only if needed. Take 1 is numbered, as was Victor’s then-current practice, in order-of-preference. Note too that the common take is 2A, not just 2. ‘2A’ was cut on the alternate or second turntable, again Victor practice. Take 1 virtually disappeared until being reissued in the Bluebird “Complete Glenn Miller” series in the mid-to-late 1970s.

    • Thanks Joe. I was hoping you would see this discussion and chime in. You have specifically laid out the key facts, and I appreciate it, as I am sure others will. As I have said so often, we learn from each other.

  2. One apparent difference between Hawkins’ and Miller’s versions is that Hawkins’ included a clarinet background and accompanying riff not found in the Miller arrangement-or so I thought. Then I heard an aircheck recording from Glenn’s final night (until the following fall) at the Cafe Rouge in New York in the spring of 1940 which includes that clarinet cum riff passage. I’ve heard a number of Glenn’s performances of Tuxedo Junction (airchecks, AAF, etc.) none other of which included this passage, which I think is a shame. I sent a copy of this cut to my friend Tom Cunningham, who leads an outstanding swing repertory band in the D.C. area. To my surprise, he was not only familiar with this version, but even had an arrangement in his books that included it. However, he said he didn’t use it because he thought the piece was better without it. I have to disagree with him on that, but all the other versions Miller did indicates that mine is a minority opinion.

  3. For a generation the only Miller ‘Tuxedo Junction’ we could hear, reissued on 45 for jukeboxes and on LP compilations, had been dubbed from an off-center source. It was a great relief when RCA finally replaced that with a transfer that did not include a horrible wow!
    Now can we hear Edgar Hayes’ original recording of ‘In The Mood’, with Kenny Clarke on drums? Miller copied Joe Garland’s arrangement.

  4. Donald, you must have ESP! I am now working on a post that will explain much of the background of “In the Mood,” and how the Miller arrangement came to be. This post will include the Edgar Hayes recording. It is a fascinating story.

    Also, the poor sound quality of Miller’s “Tuxedo Junction” has been explained as being caused by some technical problem in the studio preventing the original master recording being released. The one that we have heard for decades was the safety recording, which obviously was not as technically perfect as a usual master recording that was released.

  5. “…the poor sound quality of Miller’s “Tuxedo Junction” has been explained as being caused by some technical problem in the studio preventing the original master recording being released. The one that we have heard for decades was the safety recording, which obviously was not as technically perfect as a usual master recording that was released.” This is misleading. BOTH takes WERE released as master recordings, at least initially. The original-issue Take 1 was released 23 February 1940 and remained in print for about two years, at which time (for reasons unclear) Take 2A was used. (Bolig shows -2A as also released 23 Feb 1940 but that’s incorrect. No known US gold-print copy of B-10612-A has -2A.) It’s possible the -1 master was damaged or worn out; the Bluebird LP reissue uses a transfer from a 78 pressing rather than a metal part. Master-pressed copies of -2A exist both on Bluebird B-10612-A and on Victor 20-1565-A, as I noted above. As for “technically perfect” — there is an issue with -2A (also noted above) that resulted in it being dubbed for the majority of reissues. I haven’t listened to the dub master in some time so I can’t say for sure if that’s the source of the well-known “wow” (I do recall that wow from at least one Reader’s Digest set). One might wonder what happened to Take 2 (as opposed to -2A). Victor routinely used two lathes concurrently; one lathe’s output marked with a plain take digit, the other’s suffixed A (alternate?). There would have been a -1A also. We can only guess at why -1A (rather than either -2 or -2A) wasn’t used in place of a purportedly damaged -1 master, since it would be the same performance. Victor session sheets almost invariably show -1 as a Master, -1A as Hold or Not Processed, so it’s possible -1A wasn’t retained and wasn’t available in 1942.

    As for “In The Mood” — looking forward to reading that post. We discussed that history quite a bit on the GMA/Glenn Miller Fans group.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.