November 12, 1915] 



SCIENCE 



691 



This trivial saving may, I venture to think, 

 be disregarded, and the question decided purely 

 in the interests of the reader. Ifow the reader 

 vcants one or all of three things : &st, a speedy 

 reference from the quoted statement to the 

 authority; secondly, a name and a date that 

 will appeal at once to his historical sense and 

 furnish him with some idea of the present 

 value of the statement; thirdly, a conspectus 

 of the so-called " literature " arranged in some 

 logical order. Whatever the merits of Mr. 

 Scudder's proposals, they provide the reader 

 with none of these things. Their merits are 

 twofold : they get rid of references at the foot, 

 which are expensive and encourage the vicious 

 habit of putting matter into footnotes because 

 the author will not be at the pains to rehandle 

 his text; and they do away with loc. cit., which 

 is not merely wasteful, but more often than not 

 erroneously used in place of op. cit. or torn. cit. 



In offering counter-proposals it is advisable 

 to distinguish between two classes of papers : 

 first, brief articles in which the references are 

 correspondingly few and rarely repeated; 

 secondly, long articles or memoirs in which 

 the references are correspondingly nmnerous 

 and frequently repeated. In articles of the 

 first class, references may quite easily be 

 worked into the text, and can be repeated by 

 giving the cited author's name, with a dis- 

 tinguishing date when more than one of his 

 works has been mentioned. For memoirs of 

 the second class it is certainly convenient for 

 both author and reader to have a " list of 

 works referred to " at the end (or sometimes 

 at the beginning) of the memoir. But though 

 it may save trouble to the author to number 

 these works in the order of their citation, 

 this will save nothing to the reader, for that 

 order has often no meaning apart from the 

 text. Here is an actual example: 1. Tegner, 

 1880. 2. Jespersen, 1913. 3. Johannsen, 1913. 

 4. Anon., no date. 5. Hoffding, 1910. 6. 

 Hoffding, 1914. 7. Goethe, 1858. 8. (Another 

 page of 3.) 9. Anon., 1873. 10. Eadl, 1913. 

 11. Bernard, 186Y, and so on for nearly 100 

 items. To use such a list as a guide, or to look 

 up an author in it, is difficult enough as it is, 

 but would be more so if the items were sepa- 



rated only by a 5 mm. space (a mutton-head, 

 as our printers call it). The most convenient 

 plan for subsequent reference is to give the 

 authors in alphabetical order, with the papers 

 by each in chronological order. The refer- 

 ences in the text will then be simply : Tegner 

 (1880), Goethe (1858, p. 279), Claude Ber- 

 nard (1867), H. M. Bernard (1896, p. 53). 

 Such a mode of reference gives the historical 

 perspective, and is of itself enough to save a 

 reader familiar with the subject from repeated 

 application to the list at the end. 



So far as I can see, the methods here out- 

 lined (which have no pretensions to novelty) 

 would meet all Mr. Scudder's requirements 

 and need not cost more in either time or 

 money. F. A. Bather 



British Museum of Natdkal History, 

 London, S.W., 

 October 13, 1915 



In a recent number of Soiencei Heyward 

 Scudder, in an article with the above heading, 

 calls attention to the fact that from one half 

 to one per cent, of the space in the majority of 

 scientific journals giving many references is 

 wasted by the faulty position and arrangement 

 of the references. He recommends, as a means 

 of saving this space, that each reference be 

 given a number (the numbers to run consecu- 

 tively) and that all references be printed at 

 the end of the article, leaving an extra wide 

 spacing between the period at the end of one 

 number and the next number, in order to catch 

 the eye. 



It is quite possible that the method sug- 

 gested would effect a small saving in space. It 

 would seem, however, that the desirability of 

 this method of giving references is open to 

 discussion. 



It must be conceded at the outset that the 

 matter is largely one of personal opinion, and 

 that one of the hardest tasks of a conscientious 

 editor is to edit consistently the references of 

 his journal. Furthermore, no two journals, 

 unless published under the same supervision, 

 have the same system of references. Certainly 

 no two papers, unless by the same author, will 

 give references in exactly the same way and 



1 Science, 1915, XLII., 454, October 1. 



