414 Annals of the Carnegie Museum. 



That anterior vertical and thickened cohnnn of bone which is pres- 

 ent in the keel of nearly all avian sterna is here well developed, but 

 situated at some little distance back from the anterior margin. More- 

 over, it does not descend so far as is usually seen, being apparently 

 interrupted by the muscular line which longitudinally marks the bone. 



The muscular lines of the pectoral aspect are roughly parallel to the 

 costal borders, and remain quite distinct as we proceed towards the 

 xiphoidal extremity, nearly as far as the elliptical fenestra that there 

 occur, one on either side. 



In the shoulder-girdle (Fig. 2) we find a scapula with rather a short 

 blade through a stout one, having the usual sabre-like form with ob- 

 liquely truncated extremity posteriorly. 



It contributes the usual amount of articular surface to the glenoid 

 cavity, but when /// situ does not occupy the entire length of the supe- 

 rior line of the scapular process of the coracoid, nor have any connec- 

 tion with the furculum except through a slight ligamentous attachment. 



It will be observed that in Coiiurus the coracoid has a form that par- 

 takes much of the pattern it assumes among birds generally. (See 

 Figs. 2, 5, and 8 of the i)lates. ) 



Its tuberous summit is inclined slightly forwards and towards the 

 median line, when articulated /;/ situ, and has resting against it the 

 frail clavicular head of that side. The scapular process already al- 

 luded to is well developed, but here chiefly given over to (juite exten- 

 sive ligamentous attachment. 



The coracoidal shaft is strong, comparatively of good length, and 

 subcompressed in the antero-i)Osterior direction, being faintly marked 

 at the usual sites by muscular lines. 



At the sternal extremity of the bone we find the expanded portion, 

 the form of which can best be seen in Fig. 2, where we note that the 

 lateral process at its externo-inferior angle is well marked. 



Many parrots are notorious for having incomplete furcula, as in 

 Calxptorhyiiehus. (See PI. II, Fig. 5). In others the union at their 

 medio-inferior points is very feeble. 



In this particular they resemble some of Strigidcv. Our Carolina 

 paroquet has a complete os furculum : it is, however, a very weak 

 bone, and functionally accomplishes little more than an ossified liga- 

 ment in the same position. Indeed, it reminds one very much of such 

 a structure, for when duly articulated it is but little in advance of the 

 imaginary plane that is tangent to the anterior surfaces of the coracoidal 



