SAniOKNITIIES 631 



lowest number, fourteen being more general. This fact 

 may, however, have some significance, especially when it is 

 remembered that fourteen is also found in the penguins, and 

 when the remarks on p. 164 are taken into consideration as to 

 the possible low position among birds of the Pico-Passeres. 



The vertebrae were apparently amphicoelous. There are 

 only two sacral vertebrae, and, as already stated, the tail is 

 long and composed of a long series of elongated vertebrae, to 

 each of which a pair of rectrices are attached. 



The ribs seem to have had no uncinate processes, but these 

 may have been present and cartilaginous ; another remark- 

 able feature about them is the fact that they had, as in many 

 reptiles, but one articulation. There are also a number of 

 abdominal ribs which might be supposed to be merely the 

 sternal parts of the vertebral ribs, were it not for the close 

 approximation of the V-shaped pairs. 



The skull is toothed to the very end of the jaws, thus 

 rendering improbable the presence of a beak. The nostrils 

 are definitely holorhinal, and are divided into two holes by an 

 alinasal growth, as in some living birds. Or else the sup- 

 posed posterior part of the nasals is really the antorbital 

 space present in so many birds. It does not seem certain 

 whether in the latter event the nostrils are bounded behind 

 by the posterior division of the nasal bone or whether, as in 

 pterodactyles, a process of the maxilla rises up to join the 

 nasal. The space for the eye, which has a ring of bones 

 in the sclerotic, is completed below, as in certain parrots, 

 by a bony arch. 



Concerning the sternum we must, I suppose, agree with 

 Hurst, who has observed that ' nothing is known, though 

 much has been written.' 



The scapula is eminently bird-like, as is the furcula, with 

 its TJ-shaped meeting of the two ankylosed bones. The 

 coracoid is imperfectly known. The large size of the deltoid 

 crest of the humerus and the apparent absence of the crest 

 for the insertion of the pectoraHs are the two most salient 

 facts in its structure : the latter fact supports those who 

 hold that the sternum, if present, must have been small or 



M M 2 



