College and Research Libraries CHARLES C. WADDINGTON The Location of a Library's Science Collection The arguments for and against centralization of a library's science col- lections-the weight of which seems to favor centralization-are re- viewed. Considerations of importance in deciding location include size, use, proximity, function, and comprehensiveness of .the collection. Office collections grow into departmental libraries, which in turn grow into science divisional libraries. Until book collections become formal libraries, however, they can be used to supplement formal library ser- vice. Increasing emphasis upon multidisciplinary research argues for greater centralization of science libraries. THE ARGUMENTS FOR AND AGAINST cen- tralization or decentralization of a li- brary's science collections have been con- sidered before. Miller1 and Shera2 argue rather successfully for centralization from the library administrator's point of view. Nicholson3 summarizes the factors for centralization very nicely as follows: Factors in favor of centralization of an in- stitution's library resources include econ- omy (the avoidance of some salary costs, of certain duplications of books, of multiple card catalogs, of separate physical quarters, equipment, and maintenance), the con- venience to users where subject interests cut 1 Robert A. Miller, "Centralization vs. Decentraliza- tion," BuUetin of the American Library Association, XXXIII ( 1939), 75-9, 134-5. % Jesse H. Shera, "How Much Is a Physicist's In- ertia Worth?" Physics Today, XV (August 1961), 42-3. a Natalie N. Nicholson, "Centralization of Science Libraries · at Johns Hopkins University," In Studies in Library Administrative Problems (New Brunswick N.J.: Graduate School of Library Service, Rutgers, The State University, 1960). pp. 134-5. Mr. Waddington is Coordinator of Sci- ence Collections, Dartmouth College Li- braries. across several disciplines, better preserva- tion of materials, more complete reference an increase in total resources available to all users alike, at a more reasonable cost. On the other hand, arguments for de- services, and greater _efficiency in certain operative procedures: 'In essence, it provides centralization seem not to be as well for-.. mulated. Wells4 bases his case for de- centralization on convenience to the user, and Marron5 recognizes the fact that economics favor centralization and sci- entists' reading habits favor decentraliza- tion. The compromise of duplication in microforms is not yet a workable pos- sibility because of lack of sufficient titles in these media. The need for close prox- imity between laboratory and the library in the scientific disciplines, however, can best be shown as an out-growth of three factors. First, some laboratory work re- quires almost constant supervision so 4 D. A. Wells, "Individual Department Libraries vs. Consolidated Science Libraries." Physics Today, XIV (May 1961), 40-41. 5 Harvey Marron, "Science Libraries Consolidat- ed/Departmental?" Physics Today, XVI (July 1963), 34-9. /395 396 I College & Research Libraries • September 1965 that a research worker can leave his work only for very short periods. Second, research work in the laboratory some- times requires consultation of the litera- ture during the progress of an experi- ment when leaving the building for a prolonged period of time could ruin the experiment and months of previous work. Finally, the pace of modern science is so fast that, in general, the scientist will have great need to use the periodical literature which because of its bulk (multiple bound volumes) and its nature is best left stored in a library situation with its corresponding abstracting and indexing services. It should also be noted that the advocates of centralization al- ways assume that the central library situation is efficient. In cases where this is not so, it could be much easier to run an efficient branch library or to improve its efficiency than to wait for such im- provement to take place at the central library. The key to a solution to this problem lies in an analysis of the library's func- tions. The library must be viewed first as part of a total system which considers both the book collections and the users of these collections as part of a whole. Five factors seem to be of significance when considering the location of science collections: frequency of use, size, prox- imity to the labora~9ries, kind of use, and completeness. One important measure of a book's value to the user or to the in- stitution is the frequency with which it is used. To arrive at a reasonable utiliza- tion of library materials, the books and journals must be used a certain number of times within a given unit of time. Two variables play a role when trying to in- crease this frequency of use per item: ( 1 ) the size of the collection, ( 2) the distance or proximity of the book collec- tion from the related. scientific laboratory or the chief group of users. The former favors centralization; the larger the size the more economic the library unit. The latter favors decentralization; the closer the collection is to the users the greater the possible use of the collection. Two other factors also play a role in trying to decide the best location for any given item-the type of use made of the book or journal and the completeness of the collection. Certain kinds of books and journals are only useful or find their best use in conjunction with laboratory ex- periments. For example, Index kewensis plantarum phanerogamarum nomina . . . ( Oxonii, eprelo Clarendoniano, 1895); Handbook of Chemistry and Physics (Cleveland, Chemical Rubber Co., cur- rented.); International Tables for X-ray Crystallography (Birmingham, England, International Union of Crystallography, Kynoch Press, 1952-62). In a library situation the size of the collection must be planned within cer- tain limits. To provide minimum service, specifically eighty-eight to ninety hours a week of circulation and reference cov- erage, a branch library must reach a cer- tain critical size. This size-necessary to justify a staff large enough to provide the above coverage and yet provide a sufficient amount of work-is not capable of precise definition. It will vary with the nature of the collections and the program it serves but may minimally be about ten thousand to twenty thousand volumes. For example, libraries with holdings in the applied sciences (e.g., engineering) can be on the smaller end of the spread since considerably more reference work is usually required and performed. Personnel for such a min- imum library would consist of one pro- fessional librarian and one and one-half or two clerical assistants. The type of services included would be both of the housekeeping kind and of a reference nature. The former makes the material within the library more accessible through circulation control, prompt re- shelving, periodic inventories, recall of overdue items, a card catalog, a period- ical check-in file, etc., and provides a permanent ,Point of contact with other Location of Library's Science Collection I 391 libraries in the system. References would brary and encourage patrons to use de- include guidance in locating materials partmental and individual-owned col- within this library and the rest of the li- lections as a supplement. Once science brary system complex through use of the collections reach a critical size, location card catalog; abstracting and indexing of this material in a separate central sci- services; and reference books. Literature ence library, again supplemented by de- searches and instruction in the use of partmental or individual collections, can the library and the literature could be be considered. Finally, in very large in- provided as requested. Faculty members stitutions, the main science collections benefit from prompt attention in making should continue to be centrally located selections, ordering, and processing of in a science library. Departmental read- items in areas of their individual special- ing room collections, justified by fre- ties. quency of use, which reach a critical Consideration must also be given to - size, then can be incorporated into the li- recent changes in scientific research brary system as branch libraries covering trends. Most departmental libraries are one specific discipline. These libraries the product of a former more precise di- for the most part probably would in- vision between scientific disciplines and elude only duplicate collections of items the need for close proximity of books already in the science library. A host of and laboratories. The growth of inter- devices and procedures must be included disciplinary research, 6 however, the ra- in any library system's over-all plans to pidity with which changes occur, the provide greater flexibility aimed at maxi- great increase in the quantity of litera- mum convenience for the user within the ture being published, the construction of economic limits of the institution. Mate- interdisciplinary laboratories,7 and the rial of use only within a laboratory situa- growing application of computers to li- tion should be owned by the department brary routines seem to indicate the need and should be added to the library's col- for re-evaluation of the usefulness and lection only when necessary to complete economy of departmental libraries to one the collection. oriented subject. The larger the depart- Material of a nature that can or must ment involved, the greater is the pressure be used in both the laboratory and the to duplicate material in related subject library, or that is used by two or more areas already in other libraries on cam- disciplines when located in the library pus. It is no longer possible for an in- must be accessible to users through dividual discipline to claim exclusive xerography, telephone reference service, need and use of library materials in one and a messenger service. An adequate subject area. These developments tend number of individual studies and carrels to favor greater centralization. should be available to facilitate use of One scheme for determining the best books and journals during extensive peri- location for library materials might be ods of research in the library. Material based upon consideration as to what located in the library must be supervised books and journals are properly library during the time the library is open, and material and what should be of primar- library hours should be tailored to the ily departmental or individual respon- needs of all users. Researchers having sibility. Libraries that have science col- special and justified needs should have lections below the critical size might cen- the possibility of access to the science tralize all their material in a central li- collections for limited time periods out- side of library hours. For example, a re- s Philip H. Abelson, "Trends of Scientific Research ," searcher wishing to work during a holi- Science, CXLIII ( 1964), 218-23. 7 EFL CoUeue Newsletter, No.4, May 1964. day wecl