7®' 



THE OSPREY. 



niinution of the native birds. And last, but not 

 least, the pestiferous mongoose is finishing- oft" 

 what the others have left. 



Interesting- as this class of representative 

 forms is the species which are represented in all 

 the islands by forms supposed to be identical 

 throug-hout the archipelag-o are hai"dly less so. 

 since the ([uestion arises as to the cause of their 

 uniformitj' in contradistinction to the differen- 

 tiation of the other forms. Why did not the 

 causes which brought about the crystallization 

 of the insular forms in the other genera also 

 bring- it about in all? And, so far as I can see, 

 no satisfactory answer has been forthcoming- 

 yet. It cannot be lack of time, for all those 

 birds must date back to practically the same 

 period. It cannot be lack of isolation, for they 

 seem all to live under the same conditions; the 

 undifferentiated species are not able to cross 

 from island to island any more than the others. 

 It can scarcely' be lack of inherent variability' 

 or plasticity, since all these g-enera are pretty 

 closely related. Let us look a little closer to 

 them; they are only three species, each belong- 

 ing to a separate genus, a fact in itself quite 

 sug-g-estive, viz: Vcsfiaria coccinea, Hiuiafione 

 sauffuiuca and Psitiirostra psittacea. 



The adults of the two former are nearly uni- 

 formly red, the latter belong to the green- 

 colored group of Hawaiian passerine birds 

 which all show tendency to break up when 

 present on more than one island. Notwith- 

 standing the fact chat the distinguished orni- 

 thologists in Cambridge, who have shown such 

 a keen power of discrimination on many of the 

 other forms, have failed to detect any tangible 

 difference in the Psiffirostra from the various 

 island*, nevertheless I believe that they do 

 exist and that the failure to establish this fact 

 is due to a lack of sufficientl)' large and well- 

 prepared material. With the two other species 

 the case is somewhat different. As stated 

 above, both are red, and it may be added colored 

 radically different from the j-oung birds. It is 

 quite possible that the superimposed red color- 

 ing may be indistinguishable in specimens 

 from the various islands, and as the bills are 

 highly specialized, and consequently of great 

 individual variability no constant character 

 may be furnished by this part, but the possi- 

 bilitj' remains that the coloration of the imma- 

 ture birds may prove diff'erent in the different 

 islands. Such cases are not without parallel in 

 other localities and other groups of birds, and I 

 venture to say that notwithstanding the com- 

 paratively large number of specimens of Ves- 

 tiaria and Hiniatione in the museums there is 

 nowhere a sufficient material of immature birds 

 to decide the above question. 



The next category of passerine Hawaiian 

 birds are those genera which are represented 

 only in one island each. Most of them being so 

 closely related that they must have sprung from 

 a common root and so old in the islands as to 

 have assumed almost the dignity of a separate 

 family consisting of no less than seventeen well 

 defined genera, how is it that a great number of 

 them are confined to a single island. Corviis 

 hawaiiensis and Chcetoptila aiigustipluma being 

 outside the Drepanine family and not verj' dis- 



tantly related to other Corvine and Meliphagine 

 birds may be comparatively recent arrivals hav- 

 ing had no chance to spread beyond the island 

 of Hawaii, but whA' have the Drepanine genera 

 Ciridops Viridoiiia, Loxioidcs, Rhodacamhis, 

 and Chloridops which are confined to Hawaii, 

 as well as Rothschildia peculiar to Kauai and 

 Psriidouesfor not yet found outside of Maui, why 

 have they no representatives in the other 

 islands? Some of them ma}' possibly yet be dis- 

 covered elsewhere or have become exterminated, 

 for instance Rothschildia, others may have been 

 developed generically under conditions found 

 only in the one island they inhabit, in which 

 case, in spite of their enormous modifications 

 they maj' be of comparatively recent origin. 

 That the modifying- "conditions" are chiefly 

 those of food is probable from the fact that 

 nearly all these forms, viz: Psi'u'oucstoi\ 

 Lo.vioidfs, Rhodacanthis and Chloridops run to 

 bills after the fashion of the Bullfinches and 

 the Hawfinches. It is probably also significant 

 that most of them hail from the island of 

 Hawaii where the climate varies horizontally 

 between that of the snow-covered mountain tops 

 to the palm-clad tropical lowland. 



A very important fact is brought out forcibly 

 by Professor Newton in the Introduction, viz: 

 that the Hawaiian Islands, ornithologically 

 speaking, do not belong to the "Australian 

 Region". The relationship of the birds to those 

 of Polynesia is very slim. Apart from the mi- 

 gratory birds visiting the archipelago in winter 

 mostly from Alaska the remainder of the avi- 

 fauna is overwhelmingly American. The resi- 

 dent birds fall in two categories, those only dis- 

 tantly related to species or genera of other 

 countries, and those very closely allied. Of the 

 latter kind we have Nycticora.v nycticorax nceviiis 

 indistinguishable from the American form. 

 The Asio accipifriiius although nearly cosmo- 

 politan is probably in this case derived from 

 America. Galliniila sandvicensis is hardly sub- 

 specifically distinct from G. galcata; Piilica 

 alai is closely allied to P^. a»icyicami and Hxinan- 

 fopus k)iudse)u not more removed from H. mexi- 

 cantis. These are probably all comparatively 

 recent immigrants from America. There is an 

 endemic crow in Hawaii, the only Passerine 

 bird in the island belonging to a genus of wide 

 distribution. Its precise relations are not made 

 out yet. but they are just as likely American as 

 not. Of the other Passerine genera that of 

 P/icForuis, as I have shown on a former occasion, 

 is closeh' related to the American IMyadestes 

 belonging with the latter to a separate group of 

 the Turdido' entirely peculiar to America. 



The great bulk of the Hawaiian Passeres, viz., 

 38 species in 17 genera, and constituting'- the 

 so-called family Drepanidce, belong to the 

 second category' of birds whose relationship to 

 those of other lands are rather reinote. But as 

 will be mentioned fui-ther on, it is now pretty 

 conclusively shown that their nearest allies are 

 the Ccrrebidcp, a groupe exclusively' American 

 in its distribution. 



The remaining Passeres, belonging to three 

 genera, are the only ones of distinctly Polyne- 

 sian affinities. Of these Chasicnipis, of the 

 Muscicapidcs, is closely allied to Rhipidura, 



