PHYLOGENETIC 47 



flight. Gastornis may also have retained this power : at any 

 rate it possessed a complete furcula. 



We must turn now to a broad survey of the " Carinate" or, 

 as we prefer to call them, " Neognathine " birds — a formidable 

 array, inasmuch as the remainder of the Class Aves is included 

 in this review. 



By the older systematists these were fancifully arranged so 

 as to form a linear series. That is to say, they believed it 

 possible to group the forms with which we are now concerned 

 in a continuous series, the several types being supposed to pass 

 the one into the other. But in such schemes, it must be re- 

 marked, the Struthious types were generally also included. 

 Such schemes were of course founded on external characters 

 alone. 



More philosophical methods of investigation have shown, 

 however, that no such continuity exists ; but that we are dealing, 

 not with a linear series, but with an intricate and at present 

 tangled system of branching. To trace out these ramifications 

 endless patience and research is necessary ; indeed, the classifica- 

 tion of the Class Aves on phylogenetic lines is one of the most 

 difficult tasks which the ornithologist can be called upon to 

 undertake. The attempt made by Dr. Gadow, based on, and 

 inspired by, the monumental work of Professor Max Fiirbringer, 

 has much to recommend it ; and no less valuable is the later 

 work of Dr. Chalmers Mitchell. Though founded on a single 

 set of characters — the convolutions of the intestines — and not 

 offered as a solution of the problem of descent, this work shows 

 a very remarkable grasp of facts and power of interpretation. 



The writer of these pages has endeavoured to blend the 

 labours of these investigators in setting forth the main prin- 

 ciples of the classification of the Neognathae, a classification 

 which it is believed expresses more or less truthfully the 

 phylogeny of the several groups. 



Briefly the Neognathai may be regarded as divisible into 

 two great branches, traceable to a common stock. Each of 

 these two branches again divides into two, giving us on the one 

 hand what we must describe as the Colymbo- and Pelargo- 

 morphine branches, and on the other the Alectoro- and Coracio- 

 morphine branches. 



What the ancestral stock may have been like which gave 



