470 Recent Literature. [oct. 



The change of method in the treatment of composite or polytypic species 

 introduced in the third edition has added greatly to the number of tri- 

 nomial names. In former editions all specific groups which were repre- 

 sented in the Check-List fauna by only a single form were designated by 

 binomial names, the additional subspecies by trinomial. Thus Uria 

 troille of former editions is now Uria troille troille, to render the name 

 distinctive of the particular form of the group referred to, while the species 

 name stands as a heading or caption for its subspecies as a group, whether, 

 as in other cases, the group consists of two subspecies or more. This 

 method is also applied to composite species represented in the Check-List 

 by a single subspecies, as in the case of Colymbus nigricollis, where the 

 earliest-named, or "typical," subspecies is extralimital ; but it is not 

 introduced where the group is represented by a single subspecies which 

 happened to be the first-described form of the group,i as in cases like 

 Spinus pinus, etc. The introduction of this method has increased the 

 number of trinomials over previous editions by the addition of 232. In 

 65 of these cases the 'type' subspecies is "North American" (in the sense 

 of the Check-List), and in 65 cases extralimital. (In the preceding table, 

 under subspecies, these 232 trinomials are not included.) 



Of the 802 species in the third edition, 611 are monotypic and 191 poly- 

 typic, as regards their representation in the Check-List. Of the 191 poly- 

 typic species, 102 have each only a single additional subspecies, 47 have 

 2 each, 15 have 3 each, and 27 have 4 or more each. The following have 

 the highest numbers : Lagopus rupestris, 6 ; Otus asio, 8 ; Dryobates villosus, 

 7; D. pubescens, 5; Otocoris alpestris, 13; Agelaius phoeniceus, 7; Junco 

 hyemalis,9; Melospiza melodia, 19; Passerella iliaca, 7 . 



The relationships and relative importance of the more than 600 sub- 

 species included in the Check-List is clearly set forth through the use of 

 the trinomial system, since if all the forms which are considered as entitled 

 to a place in the list were given binomial names, as some ornithologists 

 insist is the only satisfactory way of recording them, slightly differentiated 

 intergrading forms, that even the expert finds difficulty in distinguishing, 

 would have the same apparent value as wholly isolated and strongly 

 characterized congeneric species. While it may be difficult for even a 

 committee of experts always to determine with exactness the proper status 

 of certain obscurely differentiated forms, their errors are not likely to be 

 numerous, and are far more than offset by the guidance afforded to the 

 uninitiated in the multitude of other cases that can be determined beyond 

 reasonable doubt. The relatively few changes that have been made in the 

 Check-List in the status of species and subspecies have been mostly in 



1 This Is an inconsistency, due perhaps to oversight, as this innovation was 

 not adopted till the first half of the Check-List was already in galley proof. In 

 cases like Spinus pinus the name should be Spinus pinus pinus, to indicate un- 

 equivocally that the form referred to is only the typical race and does not include 

 Spinus pinus macropterus; and similarly in all parallel eases. 



