( 388 ) 



The genus COLOEUS Kaup. 



By many authors united with Comts. The following characters, however, may 

 be sutlicient excuse for its separation. Bill somewhat short, high, hardly as long as 

 the head, upper mandible rather straight. Feathers on crown and neck decomposed. 

 Tail straight, tips of rectrices not rounded, but square. Nests in holes, eggs lighter, 

 with fewer and larger spots than in C'orvus. Smell ])eculiar, musl<y, different from 

 that of the true Corvi. 



Only Coloeus-monedula, with its subspecies 0. in. dauuncus, and C. ne/jledus — 

 the latter probably not a species, but a dark aberration — belong to this genus. The 

 American Corvus ossifrofjus has none of the characteristics of Coloeus, Also the 

 eggs of C. of<sifragus differ in no way from the type of the genus Con'iis, and might 

 be taken for small rooks' eggs. They do not at all resemble jackdaws' e^gs. 



The genera of PALAEARCTIC MUSCICAPIDAE. 



Kowhere, perhaps, do we find more unuecessarj' genera than among the 

 Muscieapidae. If it would only he understood that zoologists have invented the 

 so-called genera in order that we may find our way through the vast multitude of 

 species, and that we may, by this eminently practical method, group together the 

 most closely allied forms, thus expressing their affinities* in the names by which 

 we know them, and that in nature only species and subspecies have evolved, but no 

 such things as genera ! 



The generally' adopted genera of Miiscicapuhie are alleged to be based on 

 certain " structural " differences — i.e. length of vibrissae at gape, width and length 

 of lull ; but if wo examine tliese supposed characters we find it impossible to follow 

 the customary arrangement, and we shall soon perceive that really the genera were 

 separated by colour and an unlucky attemjit made afterwards to find structural 

 differences. Thus I cannot separate the genera Muacicapa, Hemichelidon, Ficedula 

 (or rather Hedymela), Sijjhin, Zanthopyi/ia (generally altered into Xanihopjygia, 

 but originally deliberately spelt Zanthojiyfjia), Arizelmnyia, and others. On the 

 other band, I .should have separated '^Hemichelidon," on account of its short first 

 jiriinary, if it was not for the fact that sibirica has as short a first jirimary as 

 i/riseisticta, while the southern representative of sibinca, i.e. Muscicapa sibirica 

 fulifjinosa, has a longer first primary, this being as long as, or sometimes even a 

 little longer than, the primary coverts. 



On TCHITREA PRINCEPS ILLEX. 



]Mr. Outram Bangs {Bull. Mas- Comp. Zool., Harvard Coll. xxxvi. 8, p. 264), 

 described as a " new species " a " Terpsiphone illex." He had only two specimens, a 

 male and a female, both from Ishigaki. He declares that they have a different wing- 



* One o£ the objections made to mojein soieatifio trinomial nomenclature is, that its adherents 

 strive to express afllnities in tlicir names, and that this is against our iirincijiU-s of nomenclature, which 

 should only serve practical purposes, and should not be based on rescarclics of afliiiitics and development. 

 I do not know who first invented the dopnia that names should not express affinities, but it seems to mo 

 to be only the personal idea of some of our ornitholoyital friends, and by no means a generally adopted 

 view, though repeated parrot-like by souk; writers. Moreover, it is at) erroneous and objectionable view, 

 because, first of all, our r.ld uni\ersally arloptcd nomenclature already expresses most emphatically the 

 aflinities of species. Everybody will admit that in a genus — unless it is a wrontrly constructed one — the 

 nearest allies are comprised, in opposition to other groups which are more distantly related. Not only is 

 there no reason why we should not express affinities in trinomi.als in a similar way as wc do in binomial 

 names, but the more wc do so the more valuable will our nomenclature be. 



