1902.] OSTEOLOGY OF THE FALCONIFORMES. 301 



we have already shown. With the exception of Serpentarius, 

 the furcula never articulates with the carina in Falconiformes. 

 In the Striges, on the contraiy, the furcula is always attached 

 to the carina. Furthermore it is a much more slender bone 

 than in the Falconiformes, and not bent upon itself. 



vi. The Pelvic Girdle. 



The pelvic girdle of the Falconiformes, through the more 

 aberrant members of the gi'ovip, beai's resemblances on the one 

 hand to that of the Gruidse, and on the other to that of the 

 Ciconiidae ; and, through the more specialized forms, to the Striges. 

 The innominates ai-e never fi'ee. 



Serpentarius presents seveial Gruine chaiacters, the most 

 noticeable of which are the pocket-like ctvvities (iliac recesses) of 

 the postacetabular ilium, and the genei-al contour of the dorsal 

 aspect of the pelvis as a whole. 



The pelvis of Serpentarius is, however, distinguishable from 

 the similar Gruine and Ciconiine pelves by the great height of 

 the supra-trochanteric process, and the enormous size of the ilio- 

 ischiadic foramen. Fui"thei'more, the ischium terminates poste- 

 riorly in a rounded or rather conical border pi'ojecting beyond the 

 postacetabular ilium ; whilst the pubis, which is long and slendei-, 

 sends up a conical process immediately below the projecting 

 extremity of the ischium, which serves more or less effectually to 

 close the obturator fissui'e posteriorly. 



The preacetabular ilia meet one another in the mid- dorsal 

 line, and there is no tiace of the sutui'e between the postace- 

 tabular ilium and the tiunsverse pi-ocesses of the synsacrum. 

 The synsacral fossse lying between the neural spines aiad the 

 postacetabular ilium are roofed over by a thin plate of bone. 

 The obturator fissure is not separated fi-om the foiamen. 



The pectineal process is wanting, not only in Serpentarius but 

 in all the pelves herein desciibed. 



In the Oathartje the pelvic girdle is, extei-naUy, distinctly 

 Ciconiine in chaiacter. The resemblances are especially notice- 

 able in the pelves of Pseudogryphus (text-fig. 35, p. 302), Catharisies 

 (text-fig. 36, p. 303), and Gypagus. 



The presence of iliac pockets, howevei-, at once distinguishes 

 these pelves from those of the Oiconise. Another Stork-like 

 feature is the deep notch in the hinder border of the innominate, 

 marking the division between the now- fused ilium and ischium. 

 In SarcorhMmphus, Pseudogryphus (text-fig. 35, p. 302), and 

 Gypagus the inferior limb of this notch is produced backwards 

 for a very considerable distance beyond the postacetabular ilium 

 to form a long spine. In the degree of development, and in the 

 position of the supra-trochanteric process, the innominate of the 

 Cathartae is Gruine. 



The pelvis of Cathartes differs markedly from that of the other 

 genera in this : That whilst in the genera just discussed the 



