Osteology of Aramus scolopaceus. 37 



outline. On the same aspect of the skull the quadrates are 

 completely visible, as is also the case with the quadrates of 

 the Crane Balearica. In Grus, on the other hand, there is 

 a slight projecting bar ot' bone belonging to the squamosal, 

 which has grown over the articular end of the quadrate 

 between its two heads ; this to some extent conceals the 

 actual articulation of the quadrate when the skull is placed 

 ill an accurately horizontal position and viewed from the 

 side. The same small bar of bone, it is true, does exist in 

 the case of the two genera Aramus and Balearica ; but it 

 is much smaller and is not effectual in concealing the 

 quadrate articulation. 



§ Vertebral Column. 



Aramus has altogether 17 cervical vertebrse, thus dis- 

 agreeing with at least many other Gruine birds, where 

 19 or 20 is the prevailing nunaber ; Psophia, however, and 

 Eupodotis agree with Aramus. As the cervical vertebrae 

 are apt to be variously reckoned by those who deal with the 

 skeletons of birds, it may be as well to state that in the 

 foregoing enumerations I have considered as '' cervicals^'' all 

 those vertebrse which lie in front of that which bears the 

 first complete rib : while by complete rib, I understand one 

 that articulates with the sternum. The atlas vertebra, as in 

 Grus, is merely notched for the odontoid process of the axis. 

 There is not a complete foramen. 



On the cervical vertebrae the catapophyses, as is well 

 known, are often, and indeed generally, of considerable use in 

 distinguishing or uniting allied birds j they often vary very 

 characteristically from family to family or it may be from 

 genus to genus. In Aramus the arrangement of these ventral 

 processes of the centra is as follows : — The catapophyses are 

 first recognisable as quite independent on the 6th vertebra 

 (counting in, of course, the atlas). From that vertebra as 

 far as the 13th the catapophyses are clear and distinct and 

 approach somewhat, being nearer to each other on the 

 last three vertebrse of the series referred to ; the approxima- 

 t;ion, however, is not clearly marked, and they are far indeed 



