468 PROF. HUXLEY ON THE CLASSIFICATION OF BIRDS. [Apr. 1 1, 



Sundevall* proposed to break it up into the three distinct groups of 

 Pici, PsiTTACi, and Coccyges, — the first to contain Picus and 

 Yiinx ; the last Pogonias, Bucco, Crotophayus, Phoenicophceus, 

 Coccyzus, Centrop^iSy Cuculus, Galbula, Dacelo, Merops, Colaris, 

 Trogon, and Caprimulgus. 



Sundevall calls these groups "orders;" but, leaving the question 

 of taxonomic rank aside, the first two exactly correspond with the 

 Celeomorphse and Psittacomorphse of the present essay ; while the 

 third nearly answers to my Coccygomorphse, — a coincidence which 

 I the more desire to signalize, as the Swedish naturalist attends only 

 to external characters, while I have, almost exclusively, been guided 

 by the skeleton. 



Kesslerf takes very much the same view as Sundevall, though he 

 is inclined to put Bucco along with the Woodpeckers, instead of ar- 

 ranging it, as Sundevall more justly does, with the Cuckoos. 



Not that the resemblances pointed out by Kessler do not exist ; 

 they are genuine enough, just as are others which might be pointed 

 out between the Woodpeckers and the Hornbills and other Coccy- 

 gomorphse ; but the structure of the skull affords a very definite 

 and complete distinction between the latter and any of the Gecino- 

 morphse. 



The Woodpeckers, in fact, are not Desmognathous, the palate in 

 these birds exhibiting rather a degradation and simplification of the 

 iEgithognathous structure. The vomers retain throughout life the 

 condition which is transitory in the Coracomorphse. With the latter 

 the Celeomorphse have in common the shortness of the wing-coverts, 

 the conical scapulce accessoiice, the bifiircate manubrium of the ster- 

 num, the multiperforate backward process of the tarso-metatarsus, 

 and the brevity of the basal phalanges of the toes as compared with 

 the penultimate. 



Thus I conceive that the Celeomorphse are intermediate between 

 the Coracomorphse and the Coccygomorphse, and that they may be 

 best associated with the former as an aberrant group of the ^Egitho- 

 gnathce, tending towards the Coccygomorphee as the Cypselomorphse 

 do in another way. 



The other iEciTHOGNATHiE are divisible into two groups, the 

 CypselomorphjE and the Coracomorph^. 



"V'The Cypselomorph^, like the Gecinomorphse, are annectent 

 forms between the Coracomorjjhse and the Coccygomorphas. 



The vomer is truncated at the anterior end, and the maxillo- 

 palatines slender and disposed nearly as in the typical Coracomorphae 

 (? Trochi/us). 



The sternum is broad and is devoid of a forked manubrium. Its pos- 

 terior edge may be entire, or may have two excavations on each side. 



The furcula has no backwardly directed median process, or only a 



* Ornithologibkt System af C. J. Sundevall, Kongl. Vetensk. Akad. Handlingar, 

 1835, p. 68. 



t " Belirfige zur Naturgeschichte der Spechte," Bulletin de la Sooete Impe- 

 riale des Naturalistes de Moscou, 1844, pp. 331-340. 



