Why Two Builders Can Share the Same Opus Number

Independent firms can collide because nothing coordinates their numbering. Each firm maintains its own series, so two series that overlap in range share the numbers in that overlap. The Saint-Hyacinthe pair makes the band concrete: Casavant’s low numbers overlap Létourneau’s entire published range, so the ambiguity is real from 1 through 143, while a number above 143 implicates only Casavant on present evidence. That upper boundary moves each time Létourneau completes another instrument, which is why the firm’s published list, rather than a remembered figure, is the thing to read.

The overlap is a matter of arithmetic, not of firm size or era. A number in the low hundreds can belong to a firm that reached it in the nineteenth century or to one that reached it recently, and nothing in the number itself says which.

A bare opus number establishes a position in a series and nothing about whose series it is. Within the Skinner family alone, a number in the 500s belongs either to the Skinner Organ Company of the mid-1920s or to the separate Ernest M. Skinner & Son of 1936 to 1942, and the Aeolian Company’s own pipe-organ series passed 1700 while Aeolian-Skinner’s ended at 1495. The number becomes useful the moment a builder is attached to it, because it then selects one entry in that builder’s list and returns whatever that list records. Reading the entry itself, with its columns and gaps, is the subject of How to Read an Organbuilder Opus List.

Two neighboring questions resemble this one and are answered elsewhere. Job numbers, shop-order numbers, and contract numbers are sorted in Shop Orders, Job Numbers, Contract Numbers, and Other Non-Opus Identifiers. A second number on one instrument, whether a suffixed opus number that attaches later work to an earlier base number or a rebuilding firm’s own opus number on a console it rebuilt inside another builder’s organ, is a layer on a single instrument rather than a collision between two firms, and Rebuild Plates, Secondary Opus Numbers, and Layered Instrument Identities owns that case.

A builder family can contain several distinct firms, each with its own opus series, so one number can belong to two of them. The Skinner family is the working example. Seven distinct corporate identities appear in print and have to be kept apart when reading a record: Ernest M. Skinner & Co., Skinner & Cole Organ Co., Ernest M. Skinner & Company, Ernest M. Skinner Company, Skinner Organ Company, Aeolian-Skinner Organ Company Inc., and Ernest M. Skinner & Son. Alongside them stands The Aeolian Company, which ran its own separate pipe-organ opus series and contributed only its organ division to the 1932 merger.

One encyclopedia summary gives the main line as Skinner Organ Company in 1901, Skinner & Cole in 1902, Ernest M. Skinner & Company in 1904, and then Aeolian-Skinner Organ Company, Inc. from the 1932 merger with the pipe organ division of the Aeolian Company, with operations ceasing in 1972; that article is subject to revision and says nothing about where the companies’ records went. Whatever the exact order, a Skinner Organ Company instrument is not an Aeolian-Skinner instrument, and Ernest M. Skinner & Son is a further separate entity.

The collision follows directly from the separate series. The archived Ernest M. Skinner & Son opus list, as of August 31, 2026, shows that firm using its own 500-series numbers, with entries visible from 507 to 551 carrying list years between 1936 and 1942, a range the Skinner Organ Company had already used for different instruments in the mid-1920s. The Washington National Cathedral instrument of 1938 is documented as Ernest M. Skinner & Son Opus 510, so a reference to Skinner Opus 510 names one of two different instruments unless the corporate identity is stated. The Aeolian side has the same shape at larger scale: the Aeolian Company’s own series passed 1700 while Aeolian-Skinner’s ended at 1495, and because the two are separate series, a number that falls in both ranges needs the firm name before it means anything.

Low numbers can appear again in a later corporate era, but that does not by itself prove that the firm restarted its series. Schantz is the documented case. A 1909 instrument of the earlier firm, A. J. Schantz, Sons & Co., is catalogued at Opus 139, while eight instruments carrying a 1946 year in the Pipe Organ Database carry Opus 4, 6, 7, 10, 11, 17, 18, and 26; later instruments then reach Opus 185 in 1953, Opus 1488 in 1978, Opus 1684 in 1982, Opus 1989 in 1992, and Opus 2292 through 2305 in 2010. The 1946 low numbers are consistent with a new series beginning when the firm was renamed Schantz Organ Co., but the sources do not state that the series was restarted. A low Schantz opus number is therefore ambiguous between the pre-1946 and post-1946 series until the year in the record is read beside it, and this one documented example establishes that reuse happens, not how often.

Separating the two series is as far as the numbers themselves go. What a year column in a list actually records, and what it leaves out, is explained in How to Read an Organbuilder Opus List, and the hub page carries the date-type distinctions that apply across the family.

One firm’s own records can assign a single opus number to two different instruments. Austin’s published opus list records duplicate numbers; Opus 1003, for example, appears as two separate entries, one for Westminster Presbyterian in Buffalo, New York, and one for a cathedral in Puebla, Mexico. The list does not say whether the duplicate reflects two contracts, a divided instrument, or a list error, and the example establishes only that an Austin opus number by itself does not uniquely identify an Austin instrument.

The Pipe Organ Database shows the same thing for Reuter. As of September 1, 2026, two distinct original Reuter Organ Co. instruments both carry Opus 1368: one recorded with a 1962 year at the AGO Convention in Houston, Texas, 25 ranks and 3 manuals, marked not extant, and one recorded with a 1963 year at Northminster Presbyterian Church in Indianapolis, Indiana, also 25 ranks and 3 manuals. Each has its own separate downstream rebuild record by a different firm, so the two chains stay distinct only because location, year, and later work were recorded beside the number. The number on its own would not have separated them.

A duplicated opus number appearing twice in an organ database is not, by itself, evidence of either a numbering collision or a database error. A database that creates one record per location, or per state of an instrument, shows the same opus number on every record in that instrument’s chain, so a number that appears twice may describe one instrument at two addresses or in two conditions. Duplication alone does not discriminate between the three explanations, a genuine collision, an error, and a record structure, and it does not rule out a genuine collision either. The records’ own notes and cross-references have to be read before concluding which it is.

The Reuter pair above is a collision because the two records describe different original instruments, at different locations, with separate rebuild histories that lead to different firms. A pair of records that trace one instrument through a relocation or a rebuild is the other thing, and the difference is visible only in the record fields, not in the number.

The same firm is routinely rendered under several names by the systems a reader searches, so one firm can look like two and two firms can look like one. Möller is the clearest case. The Organ Historical Society finding aid heads its collection “M.P. MÖLLER COLLECTION”; the society’s ArchivesSpace resource is titled “M.P. Möller Collection”; Pipe Organ Database record titles use “M. P. Möller”; OCLC ArchiveGrid catalogues instrument records as “M.P. Möller organ, Opus NNNN”; a 1933 United States patent names the assignee “M. P. MOLLER, INC., OF HAGERSTOWN, MARYLAND, A CORPORATION OF MARYLAND”; Austin Organs’ own services page writes “Moller”; and reference works use “M. P. Moller” and “M.P. Moller Pipe Organ Company.” Every one of these is the Hagerstown firm, and a search that uses only one form misses records filed under the others.

Kilgen shows the reverse risk, where a name form suggests a company that may not have existed. A console nameplate reads “George Kilgen & Son”; a Petit Ensemble case decal reads “Petit Ensemble built by Geo. Kilgen & Son, Inc. St. Louis, U.S.A.”; database and venue text reads “Geo. Kilgen & Son, Inc.” and “George Kilgen & Sons, Inc.”; and project text reads “Kilgen Organ Company.” The plural “Sons” form appears mainly in venue and society prose rather than on documented plates, and whether it was ever an actual corporate style is not established, so it should not be treated as evidence of a distinct company without a primary document.

Databases add a third complication: the catalogued builder entity can split one firm in two. As of September 1, 2026, the Pipe Organ Database files instrument 73792 as “Schantz Organ Company (Opus 2351, 2025)” and instrument 73874 as “Schantz Organ Co. (Opus 2350, 2025),” so consecutive opus numbers from one firm sit under two builder entities. As of September 2, 2026, the same database indexes Odell under at least two builder identifiers, one titled “J. H. & C. S. Odell & Co.” and one titled “J. H. & C. S. Odell,” so a count of Odell records has to be taken across both; whether those identifiers mark a deliberate distinction between corporate eras or an unmerged duplicate in the data is not established, and further identifiers may exist.

A database entity split is a different matter from a firm’s own succession of names, which for Schantz includes A. J. Schantz, Sons & Co. and the Tschantz spelling of the family name, and the two should not be confused when a search returns two builder pages for what looks like one firm.

Record the corporate identity exactly as the source renders it, then the number, then the year and location the record carries. Those three things separate every case described above: the firm’s name separates independent series such as Casavant’s and Létourneau’s, the exact corporate identity separates the Skinner-family and Schantz-era series, and the year and location separate an Austin or Reuter duplicate. Where a database record is involved, note the builder entity it is filed under and any cross-reference to a relocation or later record, because the record’s own notes are what distinguish a collision from a chain. Where more than one rendering of the firm’s name exists, note the settled form and the variants, so that later searches reach the records filed under each.

None of this settles what a particular instrument is. The examples establish the evidence method and its limits; an instrument-specific conclusion needs that instrument’s own records and its physical evidence, and where plates and marks sit on the instrument itself is a separate question with its own page. A service or rebuilding project record that names the corporate identity beside the number, rather than the number alone, is the plainest protection against the ambiguity described above, and it is the form of record worth asking for whenever an instrument’s identity is being documented.