Total provision of art literature almost certainly requires international co-operation. It depends too on effective and systematic coverage at the national level, not only of traditional printed books and periodicals but also audiovisual media, microforms, manuscripts and ‘grey literature’. National provision must allow for geographical, political and economic factors and patterns of usage. Various recommendations are made to stimulate discussion of a possible British system.

Art libraries differ from other academic libraries in their stock provision and in the way they are used. The disadvantages of multi-site libraries are outweighed by the difficulties of fitting the art library into a large central library, and by the advantages to the user of co-ordinated decentralisation. The implications for art libraries of centralised cataloguing systems require careful consideration; other problems include staff structures, the use (and abuse) of quantification, and the degree and type of user education. (Text of a paper presented to the course for senior art librarians on ‘Managing the Arts’, organised by ARLIS and the Polytechnic of North London in December).

Picture libraries exist in many shapes and sizes, within the broad categories of public collections, private and usually specialised collections, and photoagencies. Users of picture libraries range from students to professional picture researchers, though even the latter may need guidance. The picture librarian combines the roles of collectors and custodian while also making the collection available for use. The Mary Evans Picture Library, which comprises some two million prints mounted and filed under subject headings, provides ‘visual documentation of the past’ up to the late 1920s, with an emphasis on social conditions. Cross-references and photographic duplicates (retained for permanent reference) are interfiled with the prints, most of which are available for loan and any of which may be copied. The organisation of the library corresponds to its users’ needs rather than to a rigid system.

History of Design is a very new discipline and there are many problems associated with defining the boundaries of the subject and developing the appropriate methodologies for research and teaching purposes. One of the major problems facing historians of design, and students, is the location and availability of source material, and it is in order to help librarians help those faced with this problem that this composite article has been written. The contributors are all members of the recently formed Design History Research Group.