December 14, 1906.] 



SCIENCE. 



Ill 



and non-exotic species — from the standpoint 

 of the original author — and the generic term 

 is one generally applied by the ancient Greeks 

 or Romans, the process of elimination is to 

 be restricted to the non-exotic species." This 

 provision was intended to prevent the incon- 

 gruity of applying an ancient Greek or Latin 

 name to species wholly unknown to the an- 

 cients, and thus using it in a grossly inap- 

 plicable sense. This, however, is an unneces- 

 sary provision, inasmuch as one of the funda- 

 mental rules of all modern codes is (A. O. TJ. 

 Code, Canon XXXI.) : " Neither generic nor 

 specific names are to be rejected because of 

 barbarous origin, for faulty construction, for 

 inapplicability of meaning, or for erroneous 

 signification." Since under this rule we tol- 

 erate all sorts of absurdities and inconsist- 

 encies in names, why should we make this 

 single exception to guard against a mild in- 

 congruity ? Why, in other words, * strain at 

 a gnat and swallow a camel ' ? As this pro- 

 vision is open to diversity of construction in 

 regard to what are ' non-exotic species from 

 the standpoint of the author,' it should by all 

 means be eliminated. If enforced in the case 

 of Yultur, gryphus could not be its type. If 

 abrogated, the method of elimination is sim- 

 plicity itself, as is clearly shown in a later 

 paragraph of this paper. The framers of the 

 code were apparently themselves so familiar 

 with the elimination principle that the neces- 

 sity of prescribing rules regarding the method 

 of its use for those less fortunate in this 

 respect did not occur to them. It is, there- 

 fore, not to be wondered at that in inexpert 

 hands dissimilar results follow its faulty ap- 

 plication. Dr. Stiles's rules and suggestions, 

 referred to by Mr, Stone, relate only in small 

 part to the method of elimination; they cover 

 the whole field of the determination of generic 

 types, including the ' four conditions ' enumer- 

 ated above, and relate mainly to a single one 

 of them, being suggestions for the selection 

 of types under the prerogative of the 'first 

 reviser.' 



Much of the perplexity and uncertainty in 

 determining types by the elimination method 

 is unjustly ascribed to it, being due to the 

 lack of conviction on the part of authors as 



to just what groups that have been awarded, 

 by one author or another, the rank of genera 

 are or are not entitled to such recognition, 

 and to the complications of synonymy that 

 necessarily result from this uncertainty. The 

 application in such cases of the first species 

 rule instead of the elimination method does 

 not in the least help the matter, as is obvious 

 from the nature of the case. Yet the onus of 

 the trouble has time and again been saddled 

 on elimination. 



Elimination, properly applied, is an exceed- 

 ingly simple and definitive process. We have 

 a genus, composed originally of several spe- 

 cies, the type of which it is necessary to de- 

 termine. 



(1) Species added subsequent to the found- 

 ing of the genus are excluded from considera- 

 tion. (2) If some or all of the original spe- 

 cies have been made the types of other genera, 

 or are strictly congeneric with such types, 

 they can not be taken as the type of the orig- 

 inal genus, unless all have been so removed, 

 when the last species thus removed becomes 

 the type. (3) If only a part have been re- 

 moved, the type is to be selected by the re- 

 viser from those that remain. (4) If none 

 have been removed, any one of them may be 

 taken as the type, at the discretion of the 

 reviser — either the first species or any other. 



By the first species rule the work of the 

 first reviser is eliminated; hundreds of gen- 

 era which have had their types thus fixed are 

 in current use, and in many cases have been 

 in current use for decades, and to displace 

 them through the introduction of a new rule 

 would cause great and needless confusion. 

 The tendency has been, during recent years, 

 to preserve old names, whether generic or 

 specific, wherever possible. 



By the first species rule, if the first species 

 is unidentifiable in a genus originally contain- 

 ing a number of species, but for which the 

 founder gave no type, the genus is eliminated 

 as having no standing, although the type may 

 have been fixed by some later author, and the 

 genus be in good standing under current rules 

 of nomenclature. 



By the first species rule, where the first 

 species happens to be the same in two or more 



