1911] Composition of Taxonomic Papers 211 



should be prevented from writing at all! At the bottom of 

 every insufficient index is not carelessness, but downright 

 laziness!" 



To set the standard for indices is not very difficult ; but the 

 standard varies with the contents of books and papers. Here 

 is the criterion: Since the aim of an index is to make the con- 

 tents of a volume accessible to the reader, it should be so con- 

 structed that it will permit access to the greatest possible 

 number of references in the least possible time. In other words 

 an index is a medium of saving time. Hence an index should 

 not be merely a carelessly jumbled summary of the contents, 

 but a carefully arranged alphabetic list of all names, facts and 

 captions in the volume. This includes technical as well as 

 popular names, generic as well as specific names. 



There is such a thing as over-indexing. The author must 

 use his judgment as to the amount of detail he desires to index. 

 Also, unnecessary repetition should be avoided. One fact, how- 

 ever, is patent; that if the author wishes to see his work con- 

 sidered at all as a work of reference, he must supply it with a 

 good index. I, for one, do not care to use poorly indexed books, 

 and consult such as rarely as possible. To say the truth, 

 I consider it a personal affront, when upon purchasing a book, 

 I find myself maltreated to several hundred pages of facts and 

 names, and a two-page index. The author has no cause to 

 treat his readers as if their brains were ware-houses; that they 

 need but read his book and file away the contents together with 

 the exact page number, etc., for future reference. By pur- 

 chasing and reading a book I am doing the author a twofold 

 service. And if I remember some of the statements and quote 

 the book as an authority, the acme of the author's expectations 

 is then reached. More he has no right to demand. But a 

 starved index is inimical to progress, since few men will care 

 to quote when they are unable to find the passages from an 

 insufficient index. 



When is an index desirable? One friend has stated this 

 succinctly: "Any taxonomic paper citing more than fifty 

 names should have an index of its own. ' ' This seems reasonable 

 to me. An index of fifty names, run in two columns, eight point 

 on a ten point base, would occupy less than the ordinary four 

 by seven page of our journals. Because of the practice of 

 societies and institutions to send reprints to an author for pri- 



