Please note: The Library of Congress does not provide search capabilities on codes for countries that have their own assignment agency. Other assignment agencies are:
The MARC Code List for Organizations contains short alphabetic codes used to represent names of libraries and other kinds of organizations that need to be identified in the bibliographic environment. This code list is an essential reference tool for those dealing with MARC records, for systems reporting library holdings, for many interlibrary loan systems, and for those who may be organizing cooperative projects on a regional, national, or international scale. There are a number of data elements in the MARC formats that call for institutional identifiers, the chief ones being those that identify the organization assigning the record control number, the agency responsible for creating or modifying a record, and the agency holding a copy of the item. In particular, it is a key to codes for holding institutions represented in the Library of Congress National Union Catalog (NUC) and other union list publications which contain holdings for reporting institutions.
This code list for organizations is the latest update to the list, previously published under the title USMARC Code List for Organizations (1996 edition). It includes new codes assigned since the last edition. A small number of existing codes have been changed or made obsolete. In all cases, previously valid codes are given as references. The large number of new codes can be attributed to continuing expansion of the use of standard identifiers, nationally by school libraries (particularly for statewide projects) and internationally as information is shared globally via the Internet.
This new edition contains over 37,000 codes, of which approximately 33,000 are valid identifiers for general use. The list also includes references from invalid codes, nearly 3,000 of which are references from codes taken from other published lists which are mentioned in the "Historical Developments" section below.
The codes listed in this publication are used to designate United States and, to a lesser extent, non-U.S. libraries and other institutions. While organizations are free to employ these codes for their own purposes, the primary use of them by the libraries and other organizations is still for bibliographic applications. The list is augmented through applications for new codes, not as the result of unsolicited assignment.
The International Standard Identifier for Libraries and Related Organizations (ISIL), ISO 15511, assigns unique identifiers to libraries and related organisations, such as archives and museums. The ISIL version of the MARC codes can be used in the same bibliographic and holdings fields that are populated by MARC organization codes. The Danish Agency for Culture is the Registration Authority and maitenance agency for ISO 15511. The Library of Congress serves as the ISIL National Registration Allocation Agency for the United States. The ISIL standard specifies that a country code identifies the country in which the library or related organization is located at the time the ISIL is assigned. As such the Library of Congress only assigns ISIL codes for organizations located within the United States. The country code shall consist of two uppercase letters in accordance with the codes specified in ISO 3166-1. The ISO 3166-1 code for the United States is US. ISIL codes assigned by the Library of Congress assume the following structure illustrated below, with the US prefix added to the MARC organization code (see below for further information about how a MARC organization code is structured):
ISIL code structure for organizations within the United States:
US-[MARC organization Code]
e.g. ISIL code for the Library of Congress:
For more information about ISIL codes visit the ISIL Registration Authority website.
The MARC Code List for Organizations grew out of the need for standard institutional codes to serve various bibliographic projects in the early 20th century. The first of these came into existence in the early part of the last century, when Charles Evans wanted to report that a certain library held one of the titles listed in his American Bibliography (published 1903-1959). He assumed that the abbreviation "BPL" would be identified immediately as representing the Boston Public Library, or that "HC" could only mean Harvard College to the reader. Prior to Evans developing his list of library codes, Joseph Sabin had used codes of his own to identify libraries in his Dictionary of Books Relating to America (published 1867-1936). He had used "B" and "H" to represent Boston Public Library and Harvard College, respectively. Others used codes such as "BoP" and "Har", or "B.P." and "CH". As bibliographies appeared one after another, it became incumbent upon each editor to include an explanation of the codes used to identify libraries and other institutions. Over the years, when one library wished to refer to other libraries, it adopted abbreviations used by respected reference works or it invented new abbreviations of its own.
In 1916, the discovery that the Universities of Illinois, Minnesota, Nebraska, and Wisconsin were all about to publish lists of their serial holdings, independently of one another, gave rise to a cooperative movement on a national scale which eventually produced the first Union List of Serials (ULS) (which was first published in 1927). The scope of this project necessitated the creation of many new identifiers for libraries. Since reporting institutions were to be found in every state, Frank Peterson, of the University of Nebraska, worked out a mnemonic system of codes that identified a library's location by the use of geographic prefixes. His geographic prefixes predated the familiar U.S. state and Canadian provincial abbreviations used by postal services and in many other applications.
Peterson's scheme was of special interest to the Library of Congress, which had greatly expanded its Union Catalog by "Project B", an effort financed by a Rockefeller Foundation grant in 1926. By April 1932, at the conclusion of the Rockefeller project, a modest pamphlet entitled Key to Symbols in Union Catalogs was published by the Library of Congress. That publication identified the 367 libraries represented in the Union Catalog at that time and is considered the first edition of what would eventually become this MARC list. A slightly enlarged second edition appeared in January 1933. A more substantial Key to Symbols used in the Union Catalog, containing 685 entries, was issued in August 1936 as the third edition of the list.
As additional libraries began to send reports to the Union Catalog the Library of Congress assigned them unique identifiers composed of letters of the alphabet, usually including a geographic prefix. Eventually some of the codes assigned for the Union List of Serials began to conflict with those needed for new contributions to the Union Catalog. For many years there were attempts to harmonize LC codes with those of the Union List of Serials, or, conversely, to harmonize the ULS forms with those of LC. The situation was exacerbated in 1937 with the appearance of Winifred Gregory's American Newspapers, 1821-1936, which assigned codes from Peterson's scheme without securing concurrence with the Union Catalog.
These problems were given a new dimension in October 1939 when Douglas McMurtrie published Location Symbols for Libraries in the United States. His list was prepared "for the use of ... the staff of the American Imprints Inventory of the Historical Records Survey Program". It contained 12,000 identifiers representing every public, college, and university library in the United States. Unfortunately, most of these libraries did not report to the Union Catalog. Years later a regional cooperative project, all in good faith, adopted the McMurtrie codes for its own use without consulting the National Union Catalog, only to learn that LC had shortened the seven-letter McMurtrie codes to three letters in the interim.
In 1941, McMurtrie published the Proposed List of Location Symbols for Libraries in All Countries of the World Except the United States. In this list he divided the world by continents, countries, and cities (e.g., Eu-Fr-P for Paris, France; An-C-T for Toronto, Canada in "America, North"). Although this proposal did not survive the massive revision of boundaries and geographic names resulting from the Second World War, some of its Canadian codes were put into use before the creation of the national Canadian scheme. Some of these codes are still found in Canadian reference sources.
Symbols Used in the Union Catalog of the Library of Congress appeared in 1942 as the fourth edition of the Key to Symbols. The title changed slightly with the fifth edition, published in 1953. Its title, Symbols Used in the National Union Catalog of the Library of Congress, was kept through the ninth edition, published in 1965. The publication of the sixth edition in 1954 coincided with the inauguration of New Serial Titles and the conversion of Canadian codes from the McMurtrie format to a new system sponsored and maintained by the National Library of Canada. To keep the size of the list within reasonable bounds, organizations were dropped in later editions if they had not reported to the Union Catalog since the previous edition. One of the principal values of the list was negated by this editorial decision. The assignment of duplicate codes became almost inevitable unless each of the previous editions was consulted. Fortunately, most of the organizations that were dropped did not begin reporting again later. Some of the more complicated problems encountered in assigning codes arose because the absence of cumulative records of the codes that had been assigned.
As the Library of Congress continued to issue new editions of its code list it gradually became apparent that the National Union Catalog was the one nationwide project that outlived the publication of individual bibliographies or specific bibliographic projects. In the absence of any other organized effort on such a large scale, it fell upon the Library of Congress to assign new codes to organizations even if they were to be involved only in regional projects of some kind. For this reason, beginning with the tenth edition published in 1969, the list of institutional identifiers came to be known by the title, Symbols of American Libraries, a title it carried for another quarter of a century. Beginning in the early 1970's, it became common for cooperating institutions to secure a unique identifier before participating in any bibliographic project that required one. The responsibility for assigning codes to organizations was held by various divisions within the Library of Congress, including the Catalog Publication Division, its successors, the Catalog Management and Publications Division, and Enhanced Cataloging Division. In 1994 the Network Development and MARC Standards Office (NDMSO) assumed the task of assigning new codes, maintaining the list, and making it available. Since in recent years the codes in the list are used most heavily in MARC records, it is here that the work of Frank Peterson, Douglas McMurtrie, the compilers of the Union List of Serials, and many others found a new home.
The 1996 edition of the list (the 15th since its inception) was published under the title USMARC Code List for Organizations. The title of the 2000 Edition of the list was changed to MARC Code List for Organizations in recognition of the harmonization of USMARC and CanMARC to form the new MARC 21 format family, and the expanded use of these codes in other MARC formats.
With the development in the late 1960's of the MARC formats for machine-readable data, uses other than identifying holding institutions emerged for organization codes. In the machine-readable records, organization codes identify cataloging agencies, record creators, inputting institutions, updating institutions, agencies assigning the record control number, and other institution-specific data, in addition to holdings. In the last decade, the use of institutional identifiers from the MARC Code List for Organizations has increased internationally, with many foreign requests for codes for national libraries and major university libraries that participate in American-based projects. The list now includes nearly 6,000 codes for non-U.S. organizations, nearly 19% of the total. The advent of cooperative cataloging projects, which allow institutions to share machine-readable bibliographic information, has put even more emphasis on the importance of a single system of organization codes. In addition interlibrary loan systems use organization codes extensively. An International Standard for organization codes (ISO 15511--International Standard Identifier for Libraries and Related Organizations (ISIL)) embraces many of the features of the codes in this MARC list. ISO 15511 was published in 2011.The Library of Congress currently serves as the ISIL National Registration Allocation Agency for the United States. For more information about ISIL codes, see above.
The MARC organization codes are structured and employ mnemonics to assist in visual location and identification of the institutions represented. The ULS system of mnemonic codes based on geographical location was adopted by the Library of Congress for many institution codes, particularly libraries. While city, state, national, and university libraries do not tend to change location, commercial firms do move occasionally from city to city and state to state. Therefore, identifiers created for them in the early years of the list are usually based on the name of the organization alone rather than a combination of a geographical and name elements. In recent years, a more consistent assignment policy has been implemented which allows for all new codes to begin with a geographic prefix.
For ISIL code information, see above.
The following general principles are used for code assignment.
- Character set - Codes are composed of basic Latin alphabetic characters and the dash ( - ). Parentheses occur in some obsolete codes but are avoided in newly assigned codes. Dashes are counted in determining the maximum length of a code, but are ignored in sorting and in determining uniqueness.
- Case - Codes generally consist of a combination of upper and lowercase alphabetic characters, although they may consist of all upper or all lowercase letters. The use of upper and lowercase letters helps to improve the readability of the codes. Case is ignored in sorting and in determining uniqueness.
- Length - Codes are variable in length but should not generally exceed ten characters. The dash ( - ) that is used in some codes is included in the character count. Beginning in 1989, codes found to contain more than ten characters were either shortened or made obsolete in favor of a newly-assigned (shorter) code conform to a new policy agreed upon by various MARC users.
Structure of Subunits
Each code consists of from one to four subunits. The first subunit usually represents a higher-level geographical jurisdiction. Subsequent subunits generally represent the next-lower jurisdictions and/or the name of the organization itself. The initial character of each subunit is an uppercase letter, although the institutional subunit, the final one, may consist of multiple uppercase letters.
- First Subunit - For institutions in the United States, the first subunit is usually for a state or territory. For non-U.S. organizations, the first subunit indicates the country. This geographic unit is usually one or two alphabetic characters. U.S. states that were most active bibliographically at the time the system was developed were given a single-letter designation (for example: codes for some organizations in New York state use the geographic prefix "N" as the initial unit of the MARC code). Other jurisdictions in the United States are identified by a modified two-letter US Pstal Service abbreviation, where the second letter of the abbreviation is lowercase (for example: "Nc" for North Carolina). Earlier codes for organizations in jurisdictions outside the US generally used a two-letter uppercase/ lowercase prefix corresponding to codes in the MARC Code List for Countries (for example "Ag-" for Argentina).. Currently, first subunit is from the ISO 3166-1 country code list (for example "AG-" for Argentina).
- Second Subunit - In most cases, the subunit for a highest-level jurisdiction is followed by a second subunit representing a geographic entity, such as a city or town. This subunit begins with an uppercase letter and is composed of one or more alphabetic characters. Subunits for cities and towns are usually one or two characters (for example: "TxDa" for Dallas, Texas), although others are longer (for example: "NNopo" for Northport, New York, and "MStoc" for Stockbridge, Massachusetts). The current policy is to prefer two-character prefixes for cities. This has resulted in the "retirement" of some three and four-character geographic prefixes for certain cities, and the expansion of some one-character city prefixes to two characters.
- Third Subunit - The third subunit usually represents the name of the institution. This unit also begins with an uppercase letter and is composed of one or more upper or lowercase letters. The letters are taken from words in the organization's name, usually initial letters but conflicts with existing codes may require some other choice. Examples include: "MSaE" for Essex Institute in Salem, Mass., "MSaP" for the Peabody Museum in the same city, "PPiUSM" for the U.S. Bureau of Mines in Pittsburgh, Pa., "PPiUS" for the U.S. Steel corporation, also in Pittsburgh, and "FrPURD" for the Université René Descartes in Paris, France.
- Fourth Subunit (Dashes) - A fourth subunit is sometimes added to a code to identify a subdivision of a larger organization. If the base code is not already seven (7) characters long, it may be separated from the preceding subunits by a dash ( - ), for example, "ICU-L" for the Law Library of the University of Chicago; "DN-Ob" for the Naval Observatory, a division of the U.S. Department of the Navy in Washington, D.C. The fourth subunit also begins with an uppercase letter and is composed of one or more upper or lowercase letters. The letters are taken from words in the name of the subdivision. The dash is counted as one of the characters toward the maximum of eight characters. The code string, without the dash must be unique against the other valid codes in the list.
Exceptions to Standard Structure
The judgment and work of many persons has gone into the creation and maintenance of the institutional identifiers over a number of years. The following special conventions were used in the early assignment of codes for simplicity. While no longer applied to newly-assigned codes, familiarity with these older practices helps to understand why a particular combination of letters was assigned to an organization.
- Official U.S. state libraries are often identified by the state code alone, e.g., "Mo".
- U.S. state universities are identified by an uppercase letter "U" following the code for the state, e.g., "TxU", "InU".
- U.S. city, state, and Canadian provincial historical societies are represented by the unit "Hi" added to the appropriate geographic unit, e.g., "WHi", "MnHi".
- A small number of libraries which participated in the earliest bibliographic projects have retained codes that were assigned to them before the standard structure was established, e.g., "MH" for Harvard and "CtY" for Yale (city subunits omitted for both).
- Older codes assigned to the public libraries of U.S. and Canadian cities and towns often consist of the subunit for the U.S. state (or Canadian province) and city alone. Codes assigned after 1992 do not follow this practice and routinely include a subunit representing the name of the organization, even if the name is simply "public library" (represented as "PL").
- For commercial enterprises, older codes consist of a code based on the firm's name alone, e.g., "BMI" for Bay Microfilms, Inc., "UnM" for University Microfilms, and "EpG" for the EP Group of Companies. These identifiers are considered a single subunit, that for the institution, which for most other codes is the third unit. Although some of these older codes lacking geographic prefixes are harder to decipher, they have been retained without change as much as possible.
- In most cases, the code for an organization is changed when its location or name is changed. In a few instances, organizations prefer to retain an older code, to minimize confusion (for example: some state universities retain letters reminiscent of earlier names or status such as "T" for "Teacher" or "C" for "College").
The organization codes in this list are designed to be used wherever the recording of the full name of an organization is not possible or desirable. There is no restriction on their use outside the MARC or bibliographic environment. Because the codes are unique regardless of capitalization and hyphenation, they can be recorded as all lower case letters or all upper case letters, with or without hyphens. The MARC 21 formats prescribe the use of codes from this list in a number of data elements (for example, field 040 (Cataloging Source)). It is recommended that the capitalization and hyphenation shown in this list be followed, as much as possible, to facilitate the legibility and intelligibility of these codes.
Although use of obsolete codes is permitted in the MARC records, their use in newly-created records is discouraged, as obsolete codes are often non-unique or in some way less desirable than the valid codes for organizations. References from obsolete codes to the valid MARC organization codes are included in this list.
Codes for Organizations Outside the U.S.
Codes for organizations outside the U.S. are generally structured like U.S. codes except that the first subunit is for the country. As stated previously, the first subunit of the code is from the ISO 3166-1 country code list followed by a hyphen (for example "FR-" for France). The second subunit represents the local jurisdiction, usually city or town. The final unit represents the actual name of the organization and may be absent in some cases, as has been described already. For Canada, the subunit for the country is followed by a subunit representing the province, and a third subunit which represents the city or town. The fourth subunit represents the name of the organization. The Library and Archives Canada currently assigns codes for Canadian organizations.
Names and Addresses
Names and addresses of organizations listed in this publication are intended to be complete enough for mailing purposes. Since most local, state, college, and university libraries can be located without a street address, often only the city, state, postal code are given for these entities. An effort has been made to include street addresses if they were available at the time this publication was sent to press. For foreign organizations, the name of the country is given in all cases.
The web site provides access to the MARC Code List for Organizations by code and organization name. The database address has changed as of 24 April 2007.
Please note: The Library of Congress does not provide search capabilities on codes for countries that have their own assignment agency. [Other assignment agencies]
Searches by MARC organization code can be done using both valid and invalid codes. Since a normalized version of the valid MARC code without hyphens or capitalization is provided, it is not necessary to key in the correct combination of upper and lowercase letters when searching. Valid and invalid codes are clearly marked in the MARC organization code database.
It is also possible to search the MARC organization code list by name of organization. Both keyword and phrase searches are supported. It is not necessary to use uppercase letters or diacritical marks in searching. Multiple name entries are provided for organizations with variant names, or foreign organizations whose names appear in other languages. In all cases, names are represented using the Latin alphabet only. Beginning with the 2000 Edition of the MARC Code List for Organizations names entries for for Chinese organizations use Pinyin romanization This enhancement coincides with the adoption, by American libraries, of Pinyin as the preferred romanization scheme for Chinese. The transliteration of names in other non-roman scripts generally follows the guidelines in the widely used ALA-LC Romanization Tables.
Search the online Organization Codes database first to be sure a code has not already been assigned. If you don't find a code for your organization, click on one of these links to jump to an easy-to-use online request form in the language of your choice. (Database address has changed as of 24 April 2007.)
Once you have applied for a new code, it generally takes up to 15 working days for processing. The MARC Code List for Organizations is very dynamic. New codes are added on a regular basis. Since information in the requests must be verified and massaged into the MARC Organization Codes database, a short delay between time of request and appearance of a newly-assigned code in the database should be expected.
Names and addresses are revised when changes are reported to us. You can use the online request form to request a code for one organization. If you need codes for more than one organization, you can fill out the Web form for each organization or send a request via email including a list of the organization names, their addresses (including street, city, state, postal code, and country), and contact information to the following email address: NDMSO@LOC.GOV
NOTE: For security reasons, attachments to email messages (such as MS Word files) are not accepted.
Requests for the assignment of new codes or changes involving organizations in Canada, United Kingdom, or Germany are processed by Library and Archives Canada, The British Library, and Staatsbibliothek zu Berlin, respectively.
Requests for the assignment of new MARC organization codes or
for changes to existing names, and/or addresses, can also be
sent via FAX to: +1-202-707-0115 or by surface mail to:
Library of Congress
Network Development and MARC Standards Office
101 Independence Avenue, S.E.
Washington, D.C. 20540-4402, U.S.A.
Please note that requests received by surface mail will take longer to process than those received via email or FAX.