878 



SCIENCE. 



[N. S. Vol. V. No. 127. 



not well aiford to wait for any necessary recti- 

 fications of this sort rather than to adopt any 

 alternative thus far suggested. I am heartily 

 in sympathy with any effort to improve the 

 present somewhat unsatisfactory method, and 

 am glad Dr. Merriam has raised this important 

 question for discussion, respecting which the 

 comparison of opinions of experts in this line 

 cannot fail to be interesting and profitable. I 

 am also glad that Mr. Roosevelt has presented 

 'A Layman's Views on Scientific Nomenclature' 

 (Science, N. S , Vol. V., No. 122, April 30, 

 1897), believing that such discussions have value 

 in rendering clear the ' reason of things ' under 

 our modern phase of systematic zoology, as it 

 has drawn forth Dr. Merriam' s admirable expo- 

 sition of the 'other side.' 



I cannot, however, qviite subscribe to Dr. 

 Merriam's proposed remedy for this ' incurable 

 inconsistency ' and these ' inevitable changes ' 

 contingent on increase of knowledge. It is 

 proposed that we base our recognition of species 

 and subspecies on ' the degree of difference be- 

 tween related forms,' principally on the ground 

 that "a knowledge of this is infinitely more 

 important than a knowledge of whether or not 

 the intermediate links connecting such forms 

 happen to be living or extinct." In other 

 words, it would be more useful ' ' if the terms 

 species and subspecies were so used as to indi- 

 cate degree of difference, rather than the au- 

 thor's opinion as to the existence or non- 

 existence of intergrades." On this point there 

 is obviously room for difference of opinion. 

 This phase of the subject, however, may be 

 waived as aside from the main point, which I 

 take to be the feasibility or non-feasibility of 

 adopting the ' degree of difference ' standard for 

 species and subspecies. Yet I would like to 

 add, in passing, that to me it is of far greater 

 interest to know that the connecting links be- 

 tween quite unlike forms still exist, and that we 

 have thus positive evidence of their genetic rela- 

 tionships, than to know that these forms, in 

 their extreme phases, have become so far differ- 

 entiated as to present differences as great as 

 ordinarily characterize closely related species. 



The real difficulty with the degree of differ- 

 ence principle is its elasticity ; it enlarges to the 

 widest possible extent the personal equation 



element, which is one of the alleged sources of 

 dissatisfaction with our present system. Dr. 

 Merriam's paper, taken as a whole, so far 

 shows the wide influence of ' personal equation ' 

 in such matters that little need be added on 

 this point, except by way of further illustration. 

 The diversity of opinion respecting the amount 

 of difference required to distinguish genera and 

 subgenera is notorious ; is it likely to be any 

 less in the case of species and subspecies, in case 

 degree of difference is taken as the basis of 

 their recognition? In reply to this, reference 

 may be made to the treatment of North Amer- 

 ican birds by the authors of the ' British Museum 

 Catalogue of Birds,' on the one hand, and of the 

 'A. O. U. Check List,' on the other. In the 

 former work some of the most worthless sub- 

 species are given the rank of full species, 

 while, on the other hand, many of the most 

 strongly marked subspecies, and even some 

 species, are wholly ignored, being reduced to 

 synonyms, with often not a word of comment. 

 And this is done not in one group, nor by one 

 author, but in all groups and by each of the 

 half-dozen or more eminent ornithologists who 

 have contributed to this monumental work. 

 This does not foreshadow any ' narrowing of 

 bounds' of the personal equation element, nor 

 give much hope of agreement on any ' degree of 

 difference ' standard for the basis of species and 

 subspecies. 



As is well known, not only a great deal de- 

 pends on 'the point of view,' but also on the 

 quantity and character of the material different 

 authors may have before them, in relation to 

 their conclusions on identical questions. The 

 point of view, expertness and amount and kind 

 of material are thus factors in the case, so that, 

 whether we adopt the intergradation test or the 

 degree of difference test, we are not likely to 

 reach unanimity of opinion on such matters for 

 a long time to come. 



But there are other points that demand con- 

 sideration. The advocates of the ' intergrada- 

 tion ' test claim that it is based on a philosophic 

 principle, and that the use of a binomial term 

 means one thing and the use of a trinomial term 

 means another and a very different thing. 



Binomial names are given only to forms 

 known or supposed to be non-intergrading — to 



