Mechanical Music Registry Project
for Wurlitzer Regular, Roll Changer, and Long Tune Roll Frame Music Rolls
The 88-Note Wurlitzer Player Piano was introduced in 1906, two years before the Wurlitzer 65-Note Automatic Player Piano was marketed. First off, it is probably a good idea to avoid confusion by pointing out that the Wurlitzer 88-Note Player-Piano music roll was unique. It used a multi-tune roll that was 10-7/16" wide, against a 100 hole tracker bar with 10-to-the-inch spacing. In sharp contrast, the common 88-note home player piano roll was 11-1/4" wide, for a tracker bar with 88 holes with 9-to-the-inch spacing. The Wurlitzer 88-note player piano and the standard home player piano were not at all interchangeable, nor the rolls compatible with each other. There were two Wurlitzer 88-note models: The Style A was fitted with a coin slot for commercial use, while the Style B was push button operated for home use.
About 65 units were manufactured in 1906. The year 1907 was the best year for production, with some 183 units produced. When the 65-Note Automatic Player Piano was introduced in 1908, it soon outsold the 88-Note Player Piano. By 1912 a grand total of only five 88-Note units were made, consisting of 2 Style A pianos and 3 Style B pianos. The Wurlitzer Automatic Roll Changer was introduced in 1910. In the Wurlitzer 10,000 Series Disposition Ledger there are only four 88-note style A pianos with the notation "RC," indicating that a roll changer had been installed. Oddly enough, all four of these roll changer equipped pianos were shipped to Kentucky in 1911, one to Frankfort, and the other three to Lexington. One of these, #14489 dated Aug. 7, 1911, in the Wurlitzer ledger, is still in Lexington today and has the oldest known roll changer, #199.
In parsing through the available Wurlitzer Monthly Roll Bulletins, the last observed newly arranged Wurlitzer 88-Note Player-Piano roll seems to be #266, listed in the December, 1921, roll bulletin. In the years leading up to 1921 a particular roll bulletin might list none, or only show one or occasionally two 88-note rolls, with perhaps as few as 3 or 4 new rolls for the entire year. After 1921 only one other 88-note roll has been observed in a roll bulletin, this one in the March, 1922 roll bulletin. However, this was not a new roll, but rather a rehash of older unsold stock, roll #245 (Irish Selections). The disappearance of new 88-note rolls in the roll bulletins after 1921 suggests that roll production for the 88-Note Player-Piano ceased altogether in 1921, although there may have been some remaining unsold stock on hand for several more years. Wurlitzer 88-note music rolls were available in two formats: (1) the regular 5-tune size that was fitted with a tabbed roll leader, and (2) the 5-tune size fitted with a cloth leader with attached metal bar for the roll changer equipped instruments.
Over the manufacturing lifespan of the 88-note and 65-note instruments the roll frames utilized evolved dramatically, starting out with a simple deKleist designed single roll mechanism to the elaborate Wurlitzer automatic roll changer, which held six music rolls and, after automatic rewind, changed to the next (or a selected) music roll without human intervention. But the single roll mechanism also evolved, becoming a rugged and dependable roll frame suitable for sustained commercial use. Although these robust mechanisms still utilized a single music roll the rolls were much larger in diameter and could therefore accommodate much more music than the early 5-tune music rolls. Now the number of tunes per roll generally ranged from 10 to 20 tunes, although some classical rolls had less than 10 tunes due to the length of the classical arrangements. Nevertheless, regardless of the number of tunes, these large multi-tune music rolls were commonly referred to as 10-tune rolls.
The tracker scales remained consistent over time, but the later 65-note scale simplified the construction of the player mechanism by eliminating vacuum expression controls as employed in the earlier 88-note player pianos. Having dispensed with vacuum expression controls, the 65-note pianos had only a hammer soft rail to provide any shading in the music, but the 65-note tracker scale did provide for register controls for additional instrumentation, such as for a rank of violin and/or flute pipes, plus snare and bass drum perforations. Some instruments built in the 1920s used multiplexing of certain control perforations to turn on and off a set of orchestra bells and also a xylophone.
Wurlitzer Tracker Comparison Table | |
---|---|
65-Note Tracker Scale | 88-Note Tracker Scale |
1. Shutoff 2. Hammer rail up, mandolin on. 3. Hammer rail down, mandolin off. 4. Sustaining pedal on. 5. Sustaining pedal off. 6. Snare drum, reiterating. 7-71. Playing notes: A through C#. 72. Flute on, violin off. 73. Violin on, flute off. 74. Bass ldrum and triangle. 75. Rewind. |
1+2. Rewind (when opened simultaneously). 3. Cancel 2nd intensity. 4. Cancel 1st intensity. 5. Hammer rail down. 6. Sustaining pedal. 7. Shutoff. 8. Hammer rail up. 9-96. 88 playing notes, A through C. 97. Cancel 3rd intensity. 98. 3rd intensity on. 99. 2nd intensity on. 100. 1st intensity on. |
When available, the color of the music roll paper is recorded. This can be an important clue when trying to identify a music roll when the label is missing. The various paper colors tend to define a date range when the roll was arranged and cut, therefore giving a clue as to where to look when trying to match it with already established roll information. Also, by knowing the paper color, anyone experienced with Wurlitzerized music can pretty much know in advance how the tune will be arranged, and sound, as the roll arrangers in any specific period of time had their own peculiar way of arranging, but all of it was definitely Wurlitzerized.
Wurlitzer APP music rolls have been observed in the following paper colors:
For regular spooled (non-roll changer) Wurlitzer music rolls, and after they were spooled and ready to stuff into a box, the factory wrapped each roll with a piece of roll paper, usually ranging from 18 to 24 inches long. On the wrapper for early red paper rolls the roll number was rubber-stamped in large numbers on the leading end, and the wrapper was also rubber stamped in large letters with the brief notation: “Remove this wrapper to play roll.” By the time that the white paper rolls came into use the roll number was no longer rubber-stamped, but rather just quickly handwritten on the wrapper. Then by the time that the green waxed paper rolls came into use, circa 1918, the wrappers seem to be neither rubber-stamped or marked up in any way. That someone had to be warned about removing the wrapper raises the question as to what kind of complaints Wurlitzer might have received to warrant such a prominent and bold notice? Perhaps it might seem ridiculous that anyone would have to be instructed to remove the short length of wrapper paper before the music roll could be used, but apparently such a notice was deemed necessary.
Most collectors today have probably never seen a Wurlitzer roll with the original wrapper still intact, let along know that such a thing ever existed. No doubt the rarity of wrappers today is due to them routinely being tossed in the trash, since at the time they would have served no useful residual purpose to a route operator or piano owner. For example, do you bother to keep and preserve the wrapping paper used for shipping when you receive some new product today? Nevertheless, surprisingly some music roll wrapper specimens do still exist, all of the currently known examples from a horde of new old stock Pianino rolls found by the late Jerry Doring, a long time, avid Southern California collector.
Wurlitzer's Music Roll Department from time to time issued what were called Standard Instructions, which the music roll department were to follow. At least one of these instruction sheets survives and is interesting in that it sets a new standard for music roll arranging and roll cutting prompted by customer complaints. The document describes how new music arrangements, starting with roll #1709, are to be arranged to the same scale, how stencils (roll masters) are to be identified, and then rolls perforated using a standard gear on the perforating machine. Any collector who has listened to a lot of Wurlitzer rolls can testify to the fact that the tempo from tune to tune is often inconsistent, and apparently a lot of Wurlitzer customers noticed this difference, too, and complained loudly about it. To quote from the Standard Instructions, "We have had so many complaints regarding the tempo of our music that we have found it necessary to make some changes in our present methods. After a thorough investigation we have found that most of our difficulties can be charged to the fact that we do not have a uniform way of making our rolls. In order to correct this condition we have prepared the following standard instructions which are not to be deviated from without the consent of the General Superintendent."
From the rollography, roll #1707 and roll #1712 both have tunes with a 1923 copyright date, which suggests that roll #1709 would also have tunes from the same time period. This then suggests that roll #1709, and by inference, the Standard Instructions issued to the Music Roll Department, were issued circa 1923, or soon thereafter. It is surprising that resolution of the tempo issue waited until such a late date before any corrective measures were taken. Perhaps the tempo problem was a minor thing until it was eventually exacerbated by the fact that by 1923 Wurlitzer was manufacturing a much wider variety of music rolls, which included not only rolls for long ago established automatic pianos and orchestrions, but also for photoplayer use and some rather new, exotic rolls for pipe organs players. Early on (after the de Kleist takeover) roll arranging was probably a relatively simple matter, with only a few categories of automatic instruments to be considered, but later on the music roll types and the uses to which they were applied had expanded greatly, with widely differing requirements for the various categories. As such, old original standards might have become confused and intermixed with the new, which by 1923 required reorganization to quiet customer complaints.
The above confirms a phenomenon many students of Wurlitzer music have noticed for a long time: Mixing early and late cut Wurlitzer band organ rolls on a duplex roll frame led to very noticeable tempo problems. Early rolls and later rolls don't play well together. It has been reported that the late Rich Olsen was investigating this problem and he had zeroed in on the year 1923 as the year when the mismatched tempo problem began. This same kind of tempo disparity has also been noted for APP and other Wurlitzer coin piano rolls when rolls of vastly different ages are intermixed on a roll changer, although perhaps the tempo issue is not so readily apparent as it is with band organ music where a strict cadence is expected.
All music rolls are identified by a number located on the music roll leader. Early Wurlitzer cut music rolls were supplied in sequentially numbered, labeled boxes containing rolls that were also labeled. Sometimes the roll was additionally rubber stamped with a roll number above the label. Later rolls also came in labeled boxes, but Wurlitzer often dispensed with a numbered program label on the roll leader altogether, with no more than a rubber stamped number centered on the roll leader. Hence, if the box was misplaced or lost the tune program for the roll was also lost. Moreover, it is not uncommon to find a music roll whereby the leader has been torn off due to some mishap. So, then, how do you identify such a damaged roll?
In the Wurlitzer Roll Department a production line of perforators were kept busy, each perforator simultaneously punching out a dozen or so copies of a particular roll type. When the perforating job was finished (and before the rolls were sent to the spooling area) they needed to be identified, and so the roll number (or some shorthand version of it) was scribbled on the tail-end of the newly cut batch of rolls. This means if you unspool a roll and examine the tail-end you will find the roll number written in pencil. However, what you find may not be the entire roll number, but only a shortened version of it. Here is one observed example: From a cutting batch for roll #2070 the number inscribed on the tail-end was the number 70. Such abbreviated numbers would not have been an issue for the roll department, because they would have been considered no more than a temporary identifier before the batch of rolls were sent on to the spooling room. So, while it is possible to identify a Wurlitzer roll by looking for the tail-end number, one must also be aware that what is observed might be a contraction or shortened number. This complicates the process of identifying a roll, but it is usually possible to determine the full number by noting the vintage of the tunes on the roll, as well as the paper color, and then comparing the mystery roll to other rolls with the same paper color and/or tunes of a similar vintage.
From the 1913 Wurlitzer Monthly Roll Bulletins comes the following messages:
WURLITZERIZED MUSIC is demanded by all musicians who want the best and know which is the best. This week's roll is far above the average. Good, tuneful melodies and rich harmonies. Don't fail to hear this roll. You will surely want it, and of course it's Wurlitzerized.
And...
Wurlitzerized Music educates the masses to a keen appreciation of that which is the best in music. No matter whether it be a popular song, a ragged rag, or a classical selection, each has its devotees, and we have the music for them. This week's offering consist of five very popular hits, all good musically and poetically. It should be in every library, and to miss it would be to lose a treat. And don't forget it's Wurlitzerized. 'Nuf said.
And so it has been said for Wurlitzerized music...
The Music Roll Database incorporates the following important attributes:
Typos, misspellings, and misprints in music roll titles, individual tune titles, and in composer names are silently corrected when noted. When an error in an original Wurlitzer roll catalogue and/or label is suspected, or when two or more spelling variations are observed when compared to other Wurlitzer sources, the possibly errant word, title, and/or composer name is checked against various Internet resources for accuracy, with the correction, if any, applied to all discoverable instances within the APP rollography, as well as in other associated rollographies.
But please note that no bracketed indications of a correction, or [sic] notices are included. Why? While notating corrections that deviate from some original Wurlitzer label or catalogue might be interesting to some readers, noting such corrections could interfere with indexing and look-up options for that particular item. A case in point is Internet searches. Very often, when searching out information for some obscure tune, what turns up in the search is a rollography entry already available on the Mechanical Music Press web site. Thus the rollography information becomes a resource for anyone looking up information. For this and similar reasons, it was deemed important to have the information in a rollography be as accurate as possible—even though the corrected result may vary slightly but insignificantly from a Wurlitzer printed reference—because a rollography entry is often the only usefully referenced source returned in an Internet search.
The database will be updated periodically if error/correction reports are received. To submit an error report please send an email to
Verified music rolls appear in database reports, while unidentified rolls (where the roll number is unknown) may or may not, depending upon the purpose and structure of the report. A roll is considered verified when the roll number, roll title, various other kinds of header information, and the tune number, tune order, tune title, and composer information is verified as reasonably correct by comparing it to whatever most trusted reference source might be available. Reference resources primarily include, but are not limited to (1) original Wurlitzer roll catalogues, (2) original Wurlitzer roll or box labels, (3) Wurlitzer program cards, (4) Wurlitzer monthly roll bulletins, and, secondarily, (5) personal roll lists with a reputation for accuracy, such as by Himpsl, Conway, Kitner/Reblitz, Sprankle, etc., (6) re-cut roll catalogues by Player Piano Company (Durrell Armstrong), Ray Siou, and/or the Player Piano Centre (Doyle Lane), and (7) roll labels for re-cut rolls by the aforementioned entities.
Unidentified APP Rolls Report: This special diagnostic report is used to study and share information about rolls for which the roll number is missing. For these rolls little about them can be reliably ascertained. This report makes it easy for anyone to assist in discovering a roll number, and in turn the accuracy of the tune and composer information.
Backfilling is when re-cut roll data is recorded not only for the re-cut roll, but the musical program data is then, if necessary, also backfilled into the Wurlitzer APP data segment of the database. This happens whenever there is presumably reliable re-cut roll information that was derived from original Wurlitzer music roll labels, but for which no official Wurlitzer label or other official reference is currently available. Thus, Wurlitzer rolls are entered into the database that are based solely upon the information provided by a re-cut roll catalogue and/or roll label. The criteria that determines whether backfilling occurs is as follows:
The primary and other (secondary) sources of roll information are shown on most database reports. How and in what order sources are reported is based upon a series of so-called Trust Levels, which are based upon the idea that the most trusted source is original Wurlitzer documentation (i.e., roll labels, program cards, and roll catalogues—although Wurlitzer information is known to contain errors). Next is information from various roll catalogues by Ralph Tussing, Durrell Armstrong, Ray Siou, Doyle Lane and the Herschell Carousel Museum, followed by accurate personal roll lists made from music rolls on-hand and/or actually observed. Lastly are the compilations of music rolls made up through reports by multiple sources, but largely for rolls never physically observed. Ironically, the Julie Porter rollography, which is the initiating seed compilation that inspired this APP database into existence, falls into the last category simply because the majority of the information was reported, with the more "original" reporting sources taking precedent. However, being in this category does not diminish the utter importance of Julie Porter's work in making possible this particular rollography database.
The Frank Himpsl APP roll information occasionally contains both the composer pseudonym/name (as printed on the roll label) plus the added composers actual legal name, i.e., "Bud Manchester (E.J. Stark)." In this example the composer's true name is in parenthesis. According to Mr. Himpsl the practice of using pseudonyms was common with ragtime writers, and female composers as well, who often used pen names because sheet music publishers had the idea that tunes by women didn't sell as well.
All generated PDF format reports are fully searchable. This means that you can initiate a search for any string of characters desired without having to first OCR or otherwise adjust the report. When viewed using a standard PDF viewer/reader, pressing CTRL + F should bring up a small dialog box that allows the input of a search string. This "string" can be any combination of alpha-numeric and/or special keyboard characters.
There is one Microsoft Excel formatted report available for those individuals who want to vary and customize the way they deal with and sort tunes and/or composers. The Excel file incorporates the following field items, which are ordered using a specialized music roll identifier code in the general format of "WAPP - 12345 (a)". Here is how to understand the identifier code:
The database is far from complete, and so any additional, or more reliable, information is both welcome and appreciated. For instance, if a music roll is verified by means of a re-cut roll label, and then a Wurlitzer roll label is discovered, the verification entry will be upgraded to indicate the increased level of trust. Moreover, partial information from damaged roll labels can be very useful, and can often be used to match up and complete otherwise incomplete tune title and/or composer information. For examples and suggestions on how to submit images click here or on the thunmnail image at right.
Very often music rolls are NOT in their correct box. Thus, it is necessary to look inside all boxes and see if the box and roll labels correspond. If they do not, then a "double yield" of information results, from both the box label and a differing music roll label.
—Please send an e-mail message to and include the following information:
Thank you for any assistance you may provide. Information submitted will be added to the music roll database and/or will be very helpful in confirming that data already collected is correct. Some of the catalogued data has come from old typed lists or nearly illegible box labels, for which no known original roll exists, and so every bit of new data can be very useful in compiling a more complete and accurate database of rolls.
Adobe Reader (or its equivalent) is required to open, fill-in, or print the Music Roll Survey Form. The free Adobe Reader can be downloaded by clicking on the icon at left.
All database report information is offered "as is," without any guarantee or warranty whatsoever of any kind, neither stated, implied, nor inferred, as to the accuracy, correctness, exactness, suitability, or usefulness of any content.
Music roll information resources: Art Reblitz, Julie Porter, Frank Himpsl, Jack Conway, Mark Reinhart, Matthew Caulfield, Don Teach, Dana Johnson, Douglas Hershberger, Terry Smythe, Jeff Wood, Mikey Mills, Charles "Rusty" King, several roll lists representing unidentified authors, and miscellaneous bits and pieces from other long forgotten sources. Examination and recompilation of the Julie Porter Excel Spreadsheet and the addition of new music roll information by Terry Hathaway. Database content proofing and error checking by Mark Reinhart.
Wurlitzer Catalogues #7 LTRF, #8 LTRF, #7 Regular and Roll Changer, and #8 Regular and Roll Changer; Wurlitzer Monthly Roll Bulletins; Wurlitzer Program Cards; Wurlitzer roll labels;.Play-Rite roll labels; Player Piano Company catalogue; Ray Siou Catalogues and labels; Player Piano Centre catalogue and labels; Ed Sprankle Catalogue; Jack Conway list, Kitner/Reblitz list, and other sources of unidentified provenance.
Terry Hathaway.