Destroyed, Dispersed, or Missing Organbuilder Records: How to Document a Negative Finding

Closure can destroy most of a firm’s factory records, and even the destruction may be recorded inconsistently. Two dates are on record for the fate of the Estey office records at Brattleboro. The Estey Organ Museum states that “Most of these records were destroyed when the company closed in 1960,” while the compiler of the Estey opus list, writing in December 1973, said the list drew on “the few factory records salvaged when the contents of the Brattleboro offices were needlessly destroyed in 1961,” a differently dated account of the same clearance. The museum’s own chronology records that the Estey buildings were sold in 1961 and that the contents were destroyed, without saying what became of the business records.

The two statements may describe different events, a first clearance at closure and a later disposal, or they may simply differ by a year in recollection; nothing on the record decides between them, so a researcher records both as made and neither averages them nor quietly prefers one. Neither statement lists what was destroyed or what survived. The same list carries two provenance statements as well: the 1973 compiler described it as drawn from “many sources, especially” the salvaged factory records, while a later web version says it came “from Cost Cards,” and neither supersedes the other. Both destruction accounts also sit alongside the documented transfer of 151 boxes of Estey pipe-organ correspondence to the Organ Historical Society in 2023, which is the clearest reason not to treat a destruction account as a complete account of anything.

A firm’s factory records can sit in an institutional repository rather than with the firm, and one firm’s material can be split between holders in materially different classes. The Reuter Organ Company’s photographic, promotional, dedication-program, sales-catalog and financial material is held at the Kenneth Spencer Research Library in Lawrence, Kansas, under four call numbers, while the firm’s extensive working records went separately to the Organ Historical Society in Villanova, Pennsylvania, so anyone researching one Reuter instrument may need both repositories for a complete documentary picture. The M.P. Möller Collection at the Organ Historical Society Library and Archives, spanning 1898 to 1992, shows the same pattern from a different firm: production paperwork now held by an institution rather than by the firm that made it.

Estey material shows how far dispersal can run beyond the obvious repository. An archival aggregator search on 2 September 2026 returned 71 results, including the Julius J. Estey Papers and the Levi K. Fuller Papers at the University of Vermont, eighty-two albumen photographic prints of Estey organs with a volume of confidential price lists for 1899 to 1906 at Harvard’s Baker Library, Estey ephemera and a Fuller acoustics manuscript at the Vermont Historical Society, and a January 1918 Estey contract with correspondence about the resulting instrument at the Indiana Historical Society. None of those holdings is a shop-order or cost-card series; they bear on the firm and its people rather than on per-instrument construction records, and a result set on one date is not a census. The point for a negative finding is that a search of one repository, even the principal one, is a search of one repository.

Sometimes no published source says where the records went, and the honest statement is that custody after a given date is unestablished. As of September 1, 2026, no published source establishes custody of the 91 Wicks file cabinets after the May 2019 asset purchase. A 2017 move notice from the firm proves the cabinets’ earlier custody, and one October 2019 church account of a Wicks document being retrieved does not identify the repository it came from. That is a negative finding about custody, not about the records: the last dated evidence shows them with the firm in 2017, one later retrieval shows that at least one document was reachable in 2019, and nothing published since fixes where the cabinets are.

Published accounts of what happened to M. P. Möller’s paper after the firm closed point to two different acquirers, no published account reconciles them, and a researcher carries them as separately attributed, separately dated statements. M. P. Möller, Inc. went into involuntary bankruptcy and closed in April 1992. The first account places the records with King of Instruments, Inc. of Chicago. The Organ Historical Society’s Pipe Organ Database builder record states that the trade name and assets were acquired by King of Instruments on 13 January 1993, a note compiled from a builder guide and trade-press citations rather than from court or deed records.

Jonathan Ambrosino, writing in 1993 as an eyewitness of the January 1993 liquidation auction, states that King of Instruments purchased the company name, “archives,” post office box and telephone number. The Diapason, in a December 2011 retrospective by an organbuilder who attended the auction, reports that the Chicago businessman Paul Stuck bought the Möller name, files and “trade secrets” for $50,000 and traded as King of Instruments until the venture failed in April 1995; the database builder record dates that closure to November 1994. Wikipedia’s article likewise says King of Instruments acquired the name, archives and customer list and briefly operated as Möller Organ Company before ceasing.

A United States trademark application for the mark “M. P. MOLLER” was filed in the name of King of Instruments, Inc. on 9 April 1993 and abandoned on 1 April 1994 without proceeding to registration, as read on 1 September 2026 from a republication of the prosecution record rather than from the trademark office itself. It documents a claim to the name; it says nothing about who held the paper.

The second account places the records with the Allen Organ Company. The finding aid for the M.P. Möller Collection, published by the Organ Historical Society Library and Archives and read on August 31, 2026, states that the collection was purchased at auction by the Allen Organ Company of Macungie, Pennsylvania, and that Allen donated it to the society in early 1995. The society’s journal reported in 1995 that Allen acquired the Möller name and papers in early 1995 and donated the papers. The finding aid gives no auction date or lot detail, and no published account establishes any relationship between King of Instruments and Allen.

The two accounts are not necessarily incompatible, because the files bought with the name in 1993 may be a different body of paper from the collection bought at auction and donated in 1995, and neither account defines what it means by “archives.” But no published account says so, and which pathway a given opus file took is unresolved.

Record each account with its source and date, state that they are unreconciled, and do not merge them into a single custody narrative. Dates attach to the events each source describes, not to a verified chain of provenance, so write “the finding aid states” or “the 2011 retrospective reports” instead of a flat claim that the records went to one holder. The rule the Möller case illustrates is that a defunct firm’s name and its business records can pass to different holders, and published sources can disagree about which went where; if your question is which later corporate names a firm traded under and where each successor firm sits, that is a separate page, linked below. For a records search the consequence is narrower: a negative result from one holder in a disputed custody chain is a negative finding about that holder and nothing more.

A single name-string search does not retrieve a firm’s records, because the same firm appears under several name forms and each repository indexes whichever form it received. In the published sources that name the Westfield firm of William A. Johnson, the firm appears under at least six name strings: “William A. Johnson,” “Wm. A. Johnson,” “William Johnson,” “Johnson Organ Company,” “Johnson & Son” and “W. A. Johnson & Son.” That is an observation bounded to the sources examined on the access date, not a survey of every rendering in use, which is itself the point: the list of forms is never known to be complete. Product names drift in the same way. As of August 31, 2026, the Austin chest appears in reliable sources as “Universal Air Chest,” “Universal Airchest,” “Universal Airchest System” and “Universal Windchest,” with the manufacturer and the Organ Historical Society using different forms, and no United States trademark registration for the term was located.

A name can also lead to the wrong business entirely. In Hagerstown the Möller name attaches to a separate automobile enterprise, the M. P. Moller Motor Car Company, whose local archival traces are distinct from the organ works; the Washington County Historical Society’s online catalogue returns only motor-car items under “Moller” and no organ-company records. That result is a documented negative finding for one repository under one name form, not evidence that organ records for the firm are missing. A negative finding therefore has to list the name forms searched, because a form you did not try is a record you may not have reached.

A gap or a code against an opus number in a published work list is a statement about the list, not about the instrument. The archived Ernest M. Skinner & Son and Aeolian-Skinner opus lists use the code NB for an instrument not built and NR where no record exists, so a number that shows NR, or shows nothing, does not by itself mean an instrument is missing or destroyed; it means the compiler had no record to enter. Read the list’s own key before treating a blank as a finding, and record the blank as what it is: the absence of an entry in a named list as published on the date you read it.

A published work list can also simply not exist in public form. As of 2 September 2026 Casavant Frères publishes no complete opus list on casavant.ca: the site groups instruments as churches, concert halls and schools, offers an organ map framed as showing some of the firm’s organs throughout the world, and gives a history timeline of selected milestones, with no downloadable or paginated opus list found across the site’s own sitemap. That finding is bounded to one website on one date; it does not establish that no Casavant opus list exists, only that none is published there.

The absence of a record for a particular instrument from a firm’s archival collection establishes that the collection as held and described does not contain such a record; it does not establish that the record was never created, that the instrument does not exist, or that the firm did not build it. The same logic runs through every example on this page: destruction accounts that do not say what was destroyed, dispersal that leaves per-instrument paper unlocated, custody that stops being traceable at a date, and lists that code their own silences. Each is a fact about the evidence available to you, and a documented negative finding reports that fact at its true size.

A responsibly stated negative finding names what was searched, over what period, in which sources, and says that no record was found rather than that no record exists. In practice that means writing the target, which is the instrument, the firm, and every name form and opus number used; the date range covered and the date of the search; each repository, published work list or database consulted; and the result. State that result in words that stay inside the search, writing “no record of this instrument was found in the sources named” instead of “no record exists” as a conclusion. The rule is a method rather than a measurement of any one archive; it follows from how register disclaimers, repository descriptions and split custody limit what a search can reach. A finding written this way stays true when a second repository turns up or a name form you never tried surfaces a file.

A well-formed negative finding reads like the bounded statements that follow, each of which names its scope and refuses to claim more. A search of public sources found none that documented the wording, material or placement of a Schantz nameplate; that does not establish that no firm convention exists. Within the public sources examined for Möller, no factory instruction or survey established a general location for opus markings, and the located marking evidence was instrument-specific: one 1948 Artiste nameplate without an opus number and one 1897 instrument with the firm name stenciled on its nameboard. For Austin, no source examined established where opus numbers are physically marked or documented the wording, material or placement of a nameplate, and again that does not establish that no factory convention existed.

For Casavant Frères, no published source examined documents where the firm marks an opus number or what its builder plate says by era; searches of the firm’s own site, two reference works, a professional-association service article and several project records produced era, action and archive information but no plate or stamping description. That finding carries its own reason for caution, because the Organ Historical Society Pipe Organ Database, which commonly records such details, was not among the sources consulted.

A patent search shows the same shape. The Dictionary of Canadian Biography records that the physician-organist Salluste Duval took out a United States patent for an adjustable-combination pedal in 1889 that was later improved in Casavant’s workshops, but neither that entry nor Casavant’s own timeline gives a patent number, jurisdiction or filing date. The firm’s own claims of patents in 1895, 1903 and 1918 carry no numbers either, and as of 2 September 2026 no patent number for any Casavant patent had been located in the published sources examined. None of these statements tells you how to identify an instrument; each tells you exactly how far a search reached and where it stopped.

When a firm’s own records are gone, the evidence classes that have been found to remain are the parish or institutional records of the owning building, the physical instrument with its nameplates and markings, the trade press of the period, published work lists, and the holdings of successor firms or societies. That list is a synthesis across several cases rather than a rule any single source states, and it is not exhaustive; it establishes what has been found to survive, not how often. Reading the instrument’s own physical evidence is a physical-identification job this page does not cover, and building the owner’s permanent record, including a photographic record, is its own page, linked below.

Two routes remain once the negative finding is written. If the firm’s records may have moved with a later corporate name or a successor firm, How to Research a Defunct Organbuilder owns tracing the firm; this page does not. If the firm’s paper is gone and the instrument’s documentary record must now be created rather than found, Photographing an Organ and Building a Permanent Institutional Archive owns that work. For the shared doctrine, the other records methods, and how to request records from a repository, start at the archives and factory records hub.