xxiv INTRODUCTION 



presented an incongruous exception. Of the first type of effort we may instance 

 Dr. Elliot's grouping in his monograph published over forty years ago, and of the 

 second the selection of the relative length of the flight feathers, to the logical 

 application of which the most typical and important genus Phasianus offers an 

 insuperable exception. 



We hear constantly of the artificiality of classification, of the makeshifts of genera 

 and species, but seldom is the absolute truth of this realized until we have an intensive 

 study of some limited group of organisms. As regards importance of definitive 

 characters in family, genus and species of Pheasants, we can in many cases 

 compare them respectively only with genera, species and subspecies of mammals or 

 reptiles. 



As I have said, there has been little attempt, and that of a wholly artificial character, 

 at classification of the Pheasants on any scale higher than genera, and the most thorough 

 search I could make revealed no consistent differences in either bones or muscles, length 

 or number or colour of feathers. Finally, I came upon a character, external and transient, 

 but in rhythmical return and in consistency well worthy of consideration. I found that 

 each year there was a regular sequence in the moulting of the tail-feathers and that 

 this held good throughout the life of the bird, the individuals of the species and the 

 members of the genus. In addition, it agreed closely with the assumed relationships 

 which had heretofore been taken for granted. 



Thus I have distinguished the following groups — 



Subfamily PERDICINAE (Quail-like) (_,. , _ . , . . 



[Tail moult centrifugal, from the central feathers i ? lood Partrid S es .... Itkagenes. 

 outward.] JTragopans Tragopan. 



i Eared-Pheasants Crossoptilon. 



Impeyans Lophophorus. 



Kaleege and Silvers .... Gennaeus. 



Crestless Firebacks Acomus. 



Crested Firebacks ..... Lophura. 



White-tailed Pheasant .... Lobiophasis. 



Junglefowl Gallus. 



Koklass Pheasants Pucrasia. 



Cheer Pheasants Catreus. 



True Pheasants Phasianus. 



Long-tailed Pheasants .... Syrmaticus. 



Golden and Amherst .... Chrysolophus. 



„,„.,. /A ,., v f Bronze-tailed Peacock Pheasants . . Chalcurus 



Subfamily ARGUS^NINAE (Argus-hke) _ J Peacock Pheasants Polyplectron. 



[Tail moults 3rd from the central pair outward j 0ceUated Rheinardius 



Subfamily PHASIANINAE (Pheasant-like) 



[Tail moult centripetal, from the outer feathers 

 inward.] 



and inward.] 



Subfamily PAVONINAE (Peafowl-like) 



[Tail moults 6th from the central pair outward.] 



[ Argus Pheasants Argusianus. 



\ Peafowl . Pavo. 



There is a certain amount of variation in the moults of the Argusianinae and Pavoninae which tends to 

 amalgamate these two groups, but for the present I shall keep them separate. 



Further than this it is impossible to go in linear classification. Whatever may 

 have been the generalized ancestor of the pheasants, the nineteen groups in which we 

 are interested have evolved more or less radially, and, considered as terminal living 

 foliage on the tree of evolution, all are equally distant from that common ancestor. In 



