korstian's classification of forestry literature 209 



of them and bring the number down (especially if they were using 

 Dewey's scheme and making forestry subordinate to horticulture!). 



For a further suggestion, if a commitee be appointed to prepare such 

 a classification, why not do as the committee of the American Libraries 

 Association on correct practice for agricultural libraries is doing — send 

 a circular letter to every library and to every person likely to be 

 interested, asking for copies of any scheme now in use and for sug- 

 gestions? If the suggestions of a few help to improve a thing, why 

 would not the suggestions of all be still better? 



By B. B. Fernozv — 



The reading of the foregoing discussion suggests the need of simpli- 

 fying the procedure in classifying material for a smaller library of a 

 technical subject not intended for general public use, but for a tech- 

 nical clientele, and, indeed, to have classifications for various purposes. 



Some 40 years ago, when the Dewey system had just made its appear- 

 ance, the writer organized the technical library of a friend according 

 to the new system ; but the friend never used it, because it was too 

 circumstantial, and by merely relying on a catchword arrangement he 

 could help himself much more readily. 



Even the most logical arrangement of the subject-matter of such a 

 varied material as a complete forestry library contains cannot dis- 

 pense with cross-referencing, for the same material viewed from differ- 

 ent points of view falls into different classes. No two men would 

 probably agree as to the logical position of many articles in a given 

 scheme. Moreover, logical classification of the contents of a large sub- 

 ject answers one purpose, but the practical needs of a library for quickly 

 finding a given publication may not be served by it. 



The specialist, the professional user of a specialized library, does 

 not need the logical scheme and can rely readily on a catchword system 

 if properly cross-referenced, and a very considerable amount of library 

 material can be handled under such a system, which, to be sure, re- 

 quires the knowledge of the "key," which consists mainly of a list of 

 the catchwords used. 



The difference in the ease of use is as between a table of contents and 

 an index. The search in a library, as in a book, is for subject-matter ; 

 it is a nuisance to first have to classify the subject instead of steering 

 directly for it. When the library becomes too large for the catchword 

 system, then a simple classification into broad classes may be added. 



We come to the conclusion that for different purposes different 

 classications are called for, and that it is doubtful whether a com- 

 mittee of the Society would be able to satisfy these various needs. 



