Archive Finding Aids, Catalog Records, and Box-Level Descriptions Explained

A finding aid is a description of an archival resource that gives the reader context and structure: who created the material, how it came to the repository, and how it is organized. The Society of American Archivists defines the term that way and treats it broadly, so a guide, a calendar, an inventory, a box list, or an index published by a repository may all be finding aids in this sense. A typical finding aid consolidates acquisition and processing information, a provenance statement with an administrative history or biographical note, the scope of the collection including its size, subjects, and media, the organization and arrangement of the material, and an inventory of the series and folders.

The Organ Historical Society’s finding aid for the M.P. Möller Collection shows the pattern. As of August 31, 2026, it identifies the collection as MS 1, dates it 1898 to 1992, states its extent as 900 linear feet, and states in its provenance note that the collection was purchased at auction by the Allen Organ Company of Macungie, Pennsylvania, which donated it to the OHS in early 1995. Each of those elements is the repository’s description of its own holding. The historical note in any finding aid is context supplied by an archivist, not an adjudication of the firm’s history.

A catalogue record is an entry in a repository’s online catalogue, which is a different discovery tool from the one that publishes finding aids. Both are descriptions of holdings, which is why a finding aid belongs to the same broad class of description as a catalogue record, but a finding aid describes an archival resource as a whole, with its context and structure, while a catalogue record describes one catalogued item. As of September 1, 2026, the Organ Historical Society directs researchers to two separate discovery tools: an online Special Collections search running on an ArchivesSpace public interface, and a separate online catalogue, which the society calls an OPAC, for its book and periodical collections. A search of one is not a search of the other. A reader who searches the online catalogue for an organbuilder’s name and finds nothing has learned only that no catalogued book or periodical matched; the archival descriptions live in the other tool.

A museum object record describes a physical object, not a body of documents. As of September 1, 2026, the Smithsonian’s National Museum of American History object record for an Estey reed organ carries fields such as maker, date made, place made, materials, dimensions, accession number, credit line, and display status, and it need not carry or reference any archival documentation about the object’s manufacture or its first owner. That is what an object record does: it states what the institution has determined about the object in its custody. It is not a finding aid, and finding an instrument in an object collection does not mean the same institution holds any paper about how that instrument was built or sold. A reader looking for contracts, specifications, or correspondence needs an archival description, which is a different record produced by a different process.

Archivists arrange material by its provenance and original order at five levels: the depository, the record group, the series, the filing unit, and the individual document. The Society of American Archivists defines arrangement that way and states that arrangement must be completed to an acceptable degree before description is possible, so the structure you see in a finding aid is the structure the archivist established first. A series is a named division within a collection. A filing unit is the box or folder in which records sit, and a description that stops there is what this page calls a box-level description. A finding aid’s inventory commonly runs to the series and folders and stops; when the inventory ends at a box or folder title, the description has reached its chosen level of arrangement, not the end of the material.

A series description tells you what kinds of material the collection contains and how the repository grouped them, and nothing finer. As of September 1, 2026, the Kenneth Spencer Research Library at the University of Kansas arranges the Reuter Organ Company photograph collection in seven series: photographic prints; glass plate and flexible negatives; dedication programs; 75th Anniversary Open House; publications; financial volumes; and oversize materials. Those seven titles tell a reader that this is a photograph collection with some printed and financial material alongside it. At the series level the finding aid does not expose folder titles, opus numbers, or dates, so a reader cannot tell from the series list alone whether a particular church’s instrument is photographed. The Möller Collection works the same way at a much larger scale, with 29 series covering contract, sales, payroll, accounting, inventory, shipping, pipe-makers’, and installers’ ledgers, contract and correspondence files, photographs, and music-store records; the series titles tell you the classes of material, and the folder inventory is where a named instrument could appear.

A box-level description tells you which physical container holds a named unit of material and how the units inside it are titled. In the Reuter photograph collection, individual components are titled by the name and city of the receiving church; St. James’s Episcopal, Houston, Texas, for example, sits at Box 5, Folder 61, within a hierarchy that runs from photographic prints, to additional photographs, to organs, as of September 1, 2026. Read that entry as a promise that a folder with that title exists in that box, and no more: it does not state how many photographs the folder holds, what they show, or when they were taken.

The OHS Möller finding aid, as of August 31, 2026, describes contract ledgers with opus-numbered entries spanning 1899 to 1933 as held in boxes 1 to 5. It describes the contract and correspondence files as arranged by opus number, with files up to Opus 5308 held on site and the remainder off site, under the note “Allow five days to retrieve off site files.” A box range and an on-site boundary are facts about physical storage. They tell you where to point a request, not what a given opus file contains.

Where a repository has described a folder at item level, the entry can be quite specific. The Möller opus folders described in OCLC’s ArchiveGrid are single folders whose contents are typically a copy of the original factory specification sheets and stoplist plus one or more photographs; the folder for Opus 9075 of 1957 holds a photocopy of the factory specification sheets and stoplist dated January 11, 1957, on two leaves, and one black and white photograph stamped on the back with the firm’s Hagerstown name. That describes one folder. It does not establish that a folder exists for every Möller opus or that every folder has the same content.

A linear foot measures shelf space, not items or instruments. The Society of American Archivists defines it as twelve inches and notes that archives use linear feet to measure shelf space and to estimate processing cost and labor per unit of material. When the Möller finding aid reports 900 linear feet, it is telling you how much shelving the collection occupies, which is useful for judging how large the body of material is and how much of it could plausibly remain undescribed. It says nothing about how many instruments are documented, how many folders exist, or whether any of that shelving concerns the instrument you are researching. An extent statement answers a storage question, not a coverage question.

Processing is the work of preparing archival material for use, and it consists of arrangement, description, and preservation activity. The Society of American Archivists’ definition adds that processing must include prioritization among collections and decisions about the level of arrangement and description each collection receives, which is why one collection is described to the folder and another only to the series. A backlog is material a repository has received but not yet subjected to one or more of those processes. The same dictionary records the observation that backlog material is not described in finding aids, catalogue records, or other online forms, which leaves it essentially hidden from the public. Processing and backlog are two sides of one boundary: what has been processed can be described and searched, and what sits in the backlog cannot, however carefully you search. A repository can hold a great deal that no published description mentions.

When a finding aid does not mention something, it tells you that the description does not mention it, and nothing more. The published finding aid for the M.P. Möller Collection identifies no shop drawings, blueprints, scale sticks, or voicing records among its 29 series, and its series list carries no separately titled series for engineering drawings either. Read literally, that is a firm statement about the collection as described. It is not a statement that the Möller factory never made such drawings, that none survive, or even that none sit inside this collection: an item can occur within a listed series without earning its own title, the photograph series is described as off site and unprocessed, the OHS holds further unprocessed material at Warminster, and access to unprocessed Special Collections is restricted. A silence produced by an undescribed backlog is not a negative finding about the record. Turning a silence into a documented negative finding requires a bounded search whose scope is stated, which is the job of the page on destroyed, dispersed, or missing organbuilder records.

Being described in an institution’s online catalogue and being held by that institution are different conditions. An institution that publishes an online search does not thereby represent that its whole holdings are described there, so an empty result from a repository’s search means that nothing described matched, not that the repository holds nothing relevant. The same logic runs through aggregators. As of September 1, 2026, ArchiveGrid, published by OCLC Research, aggregates more than seven million archival descriptions from thousands of libraries, museums, historical societies, and archives, indexing MARC records from WorldCat together with EAD finding aids harvested from contributors’ websites; a hit there is a description a repository contributed, and a miss reflects only what was contributed and harvested. Where a repository’s own online finding aids cannot be retrieved directly, collection titles are frequently still visible in public search-engine indexes, which permits a partial enumeration of what the repository describes. Titles recovered that way are discovery leads. They establish nothing about extent, arrangement, processing status, or access, and each has to be confirmed against the repository’s own description.

A finding aid that states a collection has no access restrictions is saying that no donor or institutional condition bars consultation, and that is all it is saying. The same finding aid can and does separately reserve restrictions on use based on the physical condition of the material; the Kenneth Spencer Research Library’s finding aid for the Reuter photograph collection carries exactly that pair of statements. An access restriction governs whether you may consult the material. A use restriction governs what you may do with it afterward, such as reproduce or publish it, and it is decided separately under the repository’s reproduction and permission terms. Reading “no access restrictions” as permission to reproduce or publish is an easy misreading, and it is the wrong one. How to ask for the material, and how reproduction and permission terms are handled, belongs to the page on requesting organ records from a library, museum, or historical society.

When a finding aid and another published source disagree, keep both statements, attribute each to its source, and carry the disagreement forward rather than choosing a winner. A finding aid’s historical note is written by an archivist to give context to the material; it may be more or less precise than a firm’s own history or a trade-publication column, and it does not settle a question by virtue of being archival. The Kenneth Spencer Research Library finding aid for the Reuter photograph collection states that Franklin Mitchell and Albert Neutel acquired the firm in the early 1980s; the firm’s own published history states that Albert Neutel joined in 1980 and later partnered with Franklin Mitchell to purchase it. Those statements are compatible, but they are not the same statement, and a reader who needs the acquisition date has to say which one is being relied on.

Where sources genuinely conflict, the same rule holds. For the Hook and Hook & Hastings firms, The Diapason states that the firm closed its doors in 1935 and was dissolved in April 1937; the Peabody Essex Museum’s finding aid says the firm survived until 1936; an Organ Historical Society page says it remained in business until its closure in 1936; and the Organ Clearing House states that the Weston factory was demolished after the firm closed in 1929. Those may describe different events, such as ceasing to build, ceasing to trade, formal dissolution, and demolition, but none of the four sources says which, and this page does not reconcile them. The founding date shows the same pattern: The Diapason gives 1829 for the first organ Elias and George Greenleaf Hook built together, for Danvers, while the Peabody Essex Museum finding aid and an OHS page give 1827 for the establishment of the business, and the two dates may refer to a shop opening and a first completed instrument.

The terminal date of M. P. Möller, Inc. is a third case. Published accounts, including the Pipe Organ Database builder record, place the firm’s involuntary bankruptcy and closure in April 1992, and the OHS finding aid dates the forced bankruptcy to 1992; a Hagerstown local-history feature published in December 2024 dates the firm’s last delivery, and the end of its business, to 1993. Both positions stand with their own attribution. What the 1992 closure supports, on its own sources, is that an instrument documented as delivered by M. P. Möller, Inc. must have been contracted before that closure; it does not establish who completed a contract left unfinished when the firm closed.

Agreement across source classes strengthens a claim without changing the rule. Austin’s manufacturer history states that the Clough & Warren Company, a Detroit reed-organ maker, licensed John T. Austin’s chest patent and that the first Austin-patent organs were built in 1893, and The Diapason’s 2007 anniversary feature describes the same 1893 milestone as Austin building and selling his first organ at the Clough & Warren plant; a reader can cite both while still naming each.

Published work lists deserve the same care, because they carry their own chain of authority. As of September 1, 2026, the OHS-hosted Aeolian-Skinner Archives publishes four separate published work lists, for Skinner, Aeolian-Skinner, Ernest M. Skinner & Son, and Bassett-Skinner, and states that they are used with the permission of the OHS and with the support of Allen Kinzey and Sand Lawn, the compilers of the underlying published list, crediting Kinzey for the factory specifications and shop notes he provided. A date drawn from such a list is the compilers’ date, not the hosting archive’s independent finding, and a reader should say so when the list disagrees with a finding aid. None of these examples settles anything about another firm or another instrument; each shows only how to hold two dated statements side by side.

Which page you need next depends on what you are trying to do. For orientation to the whole family of archives and factory records and the other pages in it, start at the archives and factory records hub. This page ends at interpretation. Asking a repository for the material, including its reproduction and permission terms, is handled on the request page linked above, and recording a search that finds nothing as a documented negative finding is handled on the destroyed, dispersed, or missing records page linked above. Matching what a record says to the instrument in front of you is physical identification, which sits outside this page, as do component diagnosis, builder history as a subject in its own right, and any conclusion about who built a particular instrument.