246 BULLETIN OF THE 



doubtfully referable to one of the binomial forms rather than to another. 

 Another important objection may be urged against giving binomial 

 names to intergrading forms. In faunal and nominal lists of the 

 species of a large or continental area, scarcely distinguishable forms 

 take equal rank with the most distinct congeneric species. For in- 

 stance, in a list of the birds of North America, Tardus Alicia and 

 Tardus Swainsoni, Tardus Auduboni and Tardus Pullasi, stand side 

 by side with Tardus mustelinas and Tardus fascescens, though in the 

 former cases Tardus Alicice and T. Auduboni are founded at best on 

 slight, and in the one case on inconstant individual or local differences, 

 while in the latter no two congeneric species need be more distinct. In 

 the one case only experts can distinguish the forms, and frequently they 

 only by an actual comparison of specimens, and then too frequently but 

 doubtfully, while in the other case a casual observer need not mistake 

 them. The names alone give no clew to their real character, and are 

 hence in a great measure meaningless when separated from the most ex- 

 plicit diagnoses, and whose affinities can frequently only be settled by the 

 arbitrary criterion of locality. But it is urged that cognizance should 

 in some way be taken of these differences ; and " How can they be better 

 recognized," it may be asked, " than in the way proposed ?" 



As already shown, and as I trust a large proportion of ornithologists 

 are willing to admit, these local forms occur in accordance with recog- 

 nizable laws of climatic variation, similar variations with locality occur- 

 ring, to a greater or less extent, in all species having nearly the same 

 geographical range. Eventually, then, will not the recognition of these 

 laws be sufficient, and should not a statement of the tendencies to varia- 

 tion with locality, and the degree to which it is developed, be embraced in 

 the specific diagnosis of each species as a part of its specific description ? 

 Is not this, in fact, actually essential to the proper characterization of a 

 species ? The average characters being give*n, a line or two would 

 suffice for a statement of its variations, both geographical and individual. 

 Then only in one case where now there are hundreds would there be 

 instances of doubtful identification. Till within a very recent period, 

 perhaps, no other course could have been pursued than that of giving 

 binomial names to each apparently distinct form, however slightly it 

 may have differed from others previously known. In many cases, 

 indeed, the differences between strictly intergrading geographical forms 

 are very great, — greater, indeed, if they were not thus serially con- 



