PAGE This Magazine is edited by the Publications Committee of the Library Association, who are responsible for all Unsigned Contributions. Contributions and communications for the Record should be addressed to The Hon. Secretary, The Publications Committee, The Library Associa- tion, 24 Bloomsbury Square, London, W.c. Contributions are specially invited as follows:- 1. Articles on Library Administration, Bibliography, and kindred subjects. 2. Catalogues, Bulletins, Reports and other Library Publications (these, after notice, are placed in the Association Library.) 3. Books and Periodicals (which should come within the scope of the 4. Notes of Adoptions, Benefactions, New Buildings, Library Practice, "Current Views" should be sent to Dr. E. A. Baker, Public Library, Eltham; the opinions expressed in this section are those of the writers whose signatures follow the paragraphs. Queries on Public Library Law should be addressed to the Hon. Solicitor of the Association, H. W. Fovargue, Esq., Town Hall, Eastbourne, who will send replies direct to correspondents on the understanding that both TO LIBRARIANS-IMPORTANT. B. RILEY & & CO., LTD., SUPPLY THE BEST VALUE OBTAINABLE IN LIBRARY BINDING. Patent Sewing. Best Material and Workmanship. Replacements from the Sheets. Catalogues, Price Lists, Samples and Estimates per return post. New York Public Library, U.S.A. ENTRANCE EXAMINATIONS, JUNE 11, 1912. One-Year Course for Certificate. Two-Year Course for Diploma. Send to H. W. PLUMMER, Principal, 476 Fifth Avenue, for descriptive circular. The Library Association Record. 15 APRIL, 1912. IN CURRENT VIEWS. Correspondence Classes for the General N the Middle Ages when Universities were forming they were composed of small and select numbers of students round a library as a centre. The modern library movement has placed collections of books within the reach of the multitude. Some are real students, where many are readers; but they need Reader. help and direction. At present formal lecture courses are out of the question; but a series of correspondence classes, each in the charge of a competent graduate conductor, would be the means of bringing these scattered students into closer union with the libraries; would help them and probably discover much genius that only wants fostering to unfold; and would enable the Library Association to do valuable work supplementing both University Extension and the other popular education movements at present in existence. Thus the public library would realize a definite educational purpose as an informal University. The fee charged should be moderate; and work might go on from the beginning of September till the middle of July. If a beginning were made in the History and Criticism of Literature, it might be supplemented by another class in History and Sociology, by another in Philosophy, and by another in the History of Art. In this way the subjects that tend to be overlooked from utilitarian points of view would receive careful treatment, and it is probable that in every class there would be a sufficient number of students (if each library published details of the scheme) to make this feature of the public library movement a real force; and if the papers and work were sent out through the central office at 24 Bloomsbury Square, the prestige of the Library Association would be increased, and their funds also. W. E. STEBBING. PRINCIPLES OF BOOK CLASSIFICATION. BY E. WYNDHAM HULME. Chapter V.-Principles of Co-Ordination. TRICTLY speaking, the present chapter should be pre of importance in book classification, viz. the formal and geographical division of subject-matter. It will, however, be more convenient to defer the discussion of these points to a later stage. Here it will be sufficient to state that as all headings require a formal division and many a geographical division as well, some means must be taken to secure uniformity of definition and order in these divisions. This is effected by the construction of composite auxiliary tables of literary forms and local areas, which applied to headings in succession, secure the requisite uniformity of order and definition. This formal and geographical division of headings necessarily takes precedence of their real subdivision, to the discussion of which we may now proceed without further preamble. All logical systems of classification are characterized by continuity of principle. The fundamental principles which determine the primary division of their classification are repeated with additions at each successive stage of their extension, so that the characteristics of a class can be read off with ease and certainty merely by recapitulating the specific attributes of that class together with those of the classes preceding it in the scheme of classification. In this way a logical classification becomes a sort of memoria technica or ready method of memorizing the most important features of the classes composing it. But this property is only attainable where the classifier has absolute freedom in selecting the definition of his classes. In Book Classification, where the composition of classes is practically settled beforehand by the author and publisher, anything in the shape of a logical classification is an impossibility, while a high mnemonic property in any given classification may be generally taken to indicate that a considerable tampering with the positive areas of classes as revealed by the literary survey has taken place. And this is a serious defect in practice, for all deviation from the positive basis upon which the definition of classes rests, is accompanied, as we have already shown, by a corresponding loss of anatomical truth in definition as well as of economy in registration. But because, in Co-ordination, we cannot be philosophical there is no reason why we may not strive to be methodical. For without method all systems of book classification must remain upon their present arbitrary basis. Upon what method then, is the co-ordination of book classes to be based? Obviously the particular method must be selected with reference to the function which book classification is intended to perform. This function having been already discussed in our introductory chapter, we may restate our question in the following terms, viz. What arrangement of classes is best calculated to further the discovery of subject-matter in literature? And to this question we submit there is only one consistent answer, viz. an arrangement in the order of their common subjectmatter. For admitting that other principles, philosophical, philological, geographical, historical, and alphabetical, are capable of being pressed into the service-and admitting that each one of these principles is calculated to introduce its peculiar efficiency—it is nevertheless clear that the orders resulting therefrom will not directly meet the requirements of the investigator, who, having exhausted one or more classes of subject-matter, still requires further information on the same topic. Other orders may serve to open out fresh avenues of research-may, by the familiarity of their sequence, facilitate the rapid comprehension of a classification and thus enhance its interest and usefulness; but the genetic principle, i.e. the principle which assembles classes in the order of their common subject-matter, alone is competent to prescribe |