828 



SCIENCE. 



[N. S. Vol. IX. No. 233. 



will not be the case. The best arranged 

 subject-catalogues cannot embrace refer- 

 ences which may be entirely appropriate to 

 leclinical scientific bibliographies, but do 

 not belong in general reports. The library 

 officials will be overwhelmed by the separate 

 references to articles, a great part of which 

 they do not possess. A survey of that 

 which a given library possesses, and that 

 which is still wanting, must be secured by 

 assorting the cards, and this will require an 

 enormous amount of work, constantly in- 

 creasing with each additional cross-refer- 

 ence. A library catalogue cannot and ought 

 not to give information about the contents 

 of things which are not in the library, un- 

 less it is to increase infinitely the difficulty 

 of determining what new acquisitions are 

 needful. A library is not a repertory of 

 literature. Of course, it should be able to 

 give ample information concerning those 

 things which it does possess ; it must, there- 

 fore, introduce extensively into its cata- 

 logue cross-references, but only such as are 

 of a bibliographic nature. 



Reflection on the problems and needs of 

 libraries and on the possibility of the gen- 

 eral acceptance and introduction of their 

 plans should have protected the Royal So- 

 ciety from another important mistake, from 

 the limitation of their plans to the natural 

 sciences in the broad sense. In the case of 

 80 gigantic an undertaking as the creation 

 not only of an alphabetic authors' catalogue, 

 but also of an alphabetically arranged sub- 

 ject catalogue, it is useful to limit the plan 

 at first to the inauguration of a part of the 

 scientific literature. But the whole plan — 

 the general scheme — -ought under all cir- 

 cumstances to have been extended to the 

 whole realm of knowledge ; first, in order to 

 facilitate — even to render possible — the 

 same arrangement of parts in the literature 

 of other sciences ; secondly, so that the ne- 

 cessity of uniformity might be grasped by 

 the framers of the scheme. But the Royal 



Society purposely avoids uniformity even 

 within the limits which it has drawn. " No 

 attempt has been made to use similar num- 

 bers in a similar way in two or more 

 sciences [one must, therefore, learn the 

 scheme and the signification of the charac- 

 ters employed for each science by itself] , 

 the only instance in which agreement is 

 met with being in the opening section, 

 which in most cases [therefore, not in all] 

 includes the general bibliography of the 

 science " [p. 10 of Report]. But how is it 

 carried out ? Let us take the first scheme 

 of classification : A. Pure Mathematics. The 

 first division contains the heading ' Bibli- 

 ography ' (without number or other designa- 

 tion of the rubric) ; then follow : 



" 0000 Philosophy, 

 0010 History, 

 0020 Biography, 



0030 Dictionaries and text-books, 

 0040 Pedagogy, 



0050 Addresses, lectures, essays, 

 0060 Works on methods." 



What place, what number, does Bibliog- 

 raphy receive here? In the case of ' G. 

 Meteorology ' history is 0020 and Bibliog- 

 raphy 0040, in that of ' J. Geography ' Bi- 

 bliography is 0400. Elsewhere the things 

 which are grouped together under ' Peda- 

 gogy ' recieve, generally, the index 0040, 

 but under 'J. Geogra^yhy ' it bears the num- 

 ber 0500. If, further, one compares with 

 these ' L. Zoology,^ he finds here a Table 

 with 297 sub-divisions (namely, 33 system- 

 atic and nine times these from various 

 standpoints), beginning with '02 General 

 Zoology' '(comprehensive: 0203).' The 

 wonderful division ' 31 ' '■Pedagogic and Eco- 

 nomic ' embraces : ' •' Special text-books and 

 manuals. Preservation of specimens ; Mu- 

 seums ; Zoological Gardens and Aquaria. 

 Relations to plants, injurious insects, etc. 

 Galls. Special products : wax, silk, honey. 

 Animals injurious to man. Bihliograjihical, 



