xxvi INTRODUCTION 



Peacock Pheasants and the Argus, each composed of two genera, one less and the other 

 more specialized. 



At the end, in veritable regal isolation, come the Peafowl. 



If the human manipulation of classifications were of paramount interest, a volume 

 might be written on the genera of pheasants alone. A suggested genus could doubtless 

 be found somewhere in ornithological literature for every species which we recognize 

 to-day. And each would represent the personal bias of some author, sincere enough 

 probably, but for the most part handicapped by lack of perspective, and especially by 

 the failure of general application which thorough comparison with all the other genera 

 would yield. No matter what character or characters are chosen as criteria of generic 

 differentiation, consistency should be the test brought constantly to bear. 



The great variation in the apparent generic relationships for a long time gave me 

 infinite trouble. It seemed impossible to make any decision which was not based on 

 personal bias ; to prevent aught but a tentative re-shuffling of the groups. I felt that 

 certain so-called genera were heterogeneous, and yet could find no character of separa- 

 tion which an English or French or German ornithologist would probably accept as 

 logical. At last I went outside the birds themselves, and utilized a factor even more 

 novel than the tail moult character of the subfamilies. This is geographic distribution, 

 and I found that the purposes of taxonomy were consistently fulfilled by refusing to 

 include, in any single genus, species whose ranges coincided or overlapped. The results 

 were not startling, but the rearrangement showed a conservative breaking-up or 

 coalescing of certain genera whose status had long been in dispute. 



Any definite genealogical tree has been absolutely impossible in the light of past 

 knowledge or from the more thorough research which I could bring to bear. From 

 osteological to plumage characters all show such variation that any gradual transition 

 from genus to genus seems hopeless. Evolution, as I have already said, appears to 

 have been radial, and in every species the most generalized characters are found 

 intermingled with those of the utmost specialization. Like the human brain correlated 

 with a pentadactyl hand and foot, we find extremes in almost every species of 

 pheasant. 



This condition of things bears directly upon a phase of evolution which has been 

 almost neglected and yet which may ultimately prove to be one of the most important 

 aspects of the subject. This is the inter-relation of factors within the individual species, 

 which I shall only mention here. When we are perturbed over the comparative status, 

 either higher or lower, of any pheasant, let us remember that it is an organism composed 

 of a vast number of characters, of varying importance, each of which is evolving, either 

 degenerating, holding its own or becoming more specialized. And our confusion 

 over correctly orienting the bird as a species is explainable when we realize that of this 

 vast number of characters, all are evolving in their particular manner, perhaps by 

 continuous variation, by mutation, or by some method of which we as yet know 

 nothing. In pondering problems of evolution, it seems to me that we shall arrive at 

 fundamental conclusions sooner by thinking less of our subject as such-and-such a 

 pheasant, either specifically or sub-specifically related to some other one, than if we 

 consider it, more abstractly, as an organism, composed of a vast, intricately related 

 plexus of characters. 



