Osteulogical Characters in Ornithology. 39 



To prove that true affinities indicated by the sternum are also 

 in most cases exhibited in external characters, it is only necessary 

 to refer to the paper above quoted, in which the relation of the 

 Hummers to the Swifts, and the separation of the Hornbills, 

 the Rollers, the Musophagida, and the Parrots from the Pas- 

 seres, were pointed out from the consideration of such characters 

 alone. In that paper, however, I made two important errors, 

 namely, putting the Todies with the Passeres (from the descrip- 

 tions given of their habits), and including the Swallows among 

 the Swifts. The character of the sternum is undoubtedly of 

 great importance in finally settling such points as these. 



I also at that time included the Psittaci among the Scansores ; 

 but I am now quite convinced that they deserve to rank as a 

 primary division of the class of Birds, a rank to which the great 

 peculiarity of the sternum, the large brain-cavity, and highly 

 organized cranium fully entitle them. 



With regard to M. Blanchard's determination of affinities 

 from the body of the sternum only, without its appendages, 1 

 must remark that it often leads to erroneous results. For 

 example, he says that the sterna of Merops and Tamatia do not 

 differ enough to deserve a separate description ; and he includes 

 Megalcema with Tamatia in one section, as having the same form 

 of sternum. He notices some differences in the Picidce, but 

 remarks on their resemblance to Megaliema and to the Toucans. 

 Now in all these points an examination of the entire sternum, 

 with the furcula, coracoids, and clavicles attached, leads me to 

 very different results. The sterna of Merops and Nyctiornis, 

 compared with those of two species of Megalcema, seem to me 

 to show no resemblance whatever : in almost every part they 

 present important differences of form, surface-texture, and pro- 

 portions, while the furcula and coracoids are so different in the 

 two, that I should unhesitatingly place them far apart, in at 

 least different tribes or primary divisions of the Passeres. On 

 the other hand, the sternum of the Toucans {Pteroglossus) re- 

 sembles that of Megalcema most closely in every particular, and 

 especially in the extreme weakness and complete separation of 

 the two arms of the furcula — a character which I am not aware 

 exists in any other families of birds. The stertium of the Picidce 



