314 VALUE OF COLOR CHARACTERS IN BIRDS. 



no doubt a complicated one, into which different elements enter in varying 

 proportions. The striking contrasts and resemblances, however, with which we 

 meet not only in closely allied groups but also in those obviously not intimately 

 related, offer an attractive field for investigation and one which is likely to lead 



to important results. 



I have been interested in considering especially one phase of this question, 

 namely : how far coloration may be used in various groups of birds as a clue to the 

 phylogeny of the species. In certain genera the dominant color is an obvious 

 and excellent group character, as for instance red in the genus Cardinalis or 

 green in the genus Chloropsis. So too, similar color patterns are found to run 

 through all, or nearly all, species of certain large groups, as for instance the me- 

 tallic blue or green speculum in the ducks, or the blue-and-black wing patch in the 

 true jays; while in other groups certain colors are conspicuous by their absence, 

 as the lack of red among the jays or green among the thrushes. 



Reverting to Cardinalis, we find the allied genus Pyrrhuloxia exhibiting the 

 same heavy bill and conspicuous red coloration but differing clearly in the 

 shape of the bill and in the great admixture of the gray in the plumage. Here 

 we see differentiation both in structure and coloration by which Cardinalis and 

 Pyrrhuloxia have departed from a common ancestral type. In many cases, 



however, differentiation has taken place in only one set of characters. For 

 instance in the genus Merops we have a green and yellow color type, a red and 

 green type and a green and chestnut type, but none of the species show any 

 corresponding structural differences. In the genus Tyrannus also we have gray 

 and white breasted species and others with sulphur yellow breasts, but no parallel 

 variation in the proportions of the bill or other structural characters. On the 

 other hand we have groups like Geospiza of the Galapagos Islands, where all man- 

 ner of variation in the size and proportions of the bill have taken place while the 

 coloration of the plumage remains the same throughout. Attempts have been 

 made to divide the species into several genera according to bill structure but such 

 an unbroken series of forms exists connecting one extreme with the other that 

 it seems well-nigh impossible. A still more remarkable illustration is seen in 

 the East Indian cuckoos of the essentially monotypic genera Dryococcyx from 

 Celebes, Rhinococcyx from Java, and Urococcyx from Sumatra and the Malay 

 peninsula. These birds differ strikingly in the location and character of the 



► 



in the hand without examining the bill. (PL XXVII, figs. 1,3, and 6). 



The same differentiation in bill structure is found sometimes in two forms 

 inhabiting the same region as in the Chilian parrots Enicognathus leptorhynchus 

 and Microsittace ferrugineus, both of which are green, with red tails and red on 

 the lores, forehead and abdomen, but which differ materially in the structure of 

 the upper mandible (PI. XXVII, figs. 7 and 10). 



In such cases it seems obvious that differentiation in bill structure has taken 



