Cypselidoe, Caprimulgidas, and Podargidae. 363 



also acknowledge this fact in our systematic treatment 

 of such forms, and recognize it in our nomenclature, or 

 both our work and its nomenclature will be inadequate and 

 insufficient. To my mind it is of just as much, or perhaps of 

 more interest and importance to see that a species grades, for 

 example, from a small and dark western form into a large 

 and pale eastern form, so that, if the extremes only were 

 known, they would be readily recognized as "good species,'^ 

 as to discover that a genus contains, for example, ten species 

 instead of nine. A mistake, however, in my opinion, is gene- 

 rally made in speaking of so many species and so many 

 subspecies, as if the latter existed besides the former, while, 

 in fact, the subspecies are subdivisions of the species ; 

 and it would be more correct to say that there are so many 

 species with so many subspecies, as if the latter formed parts 

 of the species. Thus, instead of enumerating (1) Acredula 

 caudata, (2) A. caudata rosea, as if they were two species, 

 we should speak of the Long-tailed Tit, Acredula caudata, 

 and we should divide this into (a) A. caudata typica and 

 (b) A. caudata rosea. To name the subspecies, i. e., such 

 forms which cannot rank as full species, is as important and 

 as necessary for our convenience as it is to name the species, 

 for they must be quoted, and it is out of the question to give 

 diagnoses of them whenever they are spoken of. In the case 

 of Acredula, for example, there seems no difficulty whatever in 

 naming the second form (b) A. caudata rosea. The trinomial 

 is, for such cases, most convenient, and undoubtedly the 

 shortest way. To name the subspecies in the same binomial 

 way as the species is unquestionably wrong, since we do not 

 consider them to be species ! If our learned friend. Dr. 

 Sharpe, writes Corethrura mcAe?^o^y^, subsp. nov., then he does 

 the same that the " trinoraialists " do : it is naming a form 

 which is not a species, and yet he names it exactly like a species. 

 If then it is quoted afterwards with his bino'lnial name, it is 

 impossible to recognize that it is not meant to be a species. 

 To term it, on the other hand, Corethrura pulchra reichenowi 

 at once simplifies matters and shows what the form in 

 question is considered to be. There cannot be a simpler and 



8ER. VIl. VOL. II. 2d 



