1871.] 253 [Hyatt 



The shoulder-girdle is not less diagnostic in the expansion of the 

 scapula, which is irregularly clubbed and almost spatulate. The 

 clavicles are very broad, flat from side to side, and strongly curved 

 backward and downward. The coracoid is long and strong, less con- 

 spicuously flattened, and developes an apophysis at the sternal ex- 

 tremity on the mesiad side. 



All the bones of the anterior extremity are flattened, and the distal 

 end of the laminar humerus has an oblique truncated articular sur- 

 face — a condition only elsewhere seen in Alca impennis. There is 

 no free radial digit — a state of things that might have been in- 

 ferred from the pterylosis of the wing. The wrist preserves two free 

 carpals, as usual, but the ulnare has an immense laminar expansion, 

 not found outside this family. The elbow has two sesamoids (a 

 feature shared, however, by the Guillemots), interesting in relation 

 to the unusually large patella, which in these birds ossifies from two 

 centres (Owen). There is a persistent, free ossicle in the ankle of 

 Aptenodytes, apparently a sesamoid 1 ; this is peculiar, so far as I 

 know. The tibia is very long, bat not specially remarkable ; it does not 

 develope the long apophysis of the Colymbidce. The pelvico-sacral 

 connections are said to be looser than in other birds. Notwithstanding 

 the small size and sessile and elevated condition of the hallux, this has 

 two bones, as usual, besides the accessory metatarsal; and the phalan- 

 ges of the other digits are of normal number (3-4-5, from inner to 

 outer anterior toe). The ribs have articulated accessory processes. 

 The bodies of the hinder dorsal vertebrae are strongly compressed 

 with median hypapophyses; the anterior dorsals have broad divergent 

 laminar parapophyses in Aptenodytes, and most of the dorsals are 

 opisthocoelian (Owen). The palate has the schizognathous structure; 

 there are no basipterygoid processes, and the pterygoids share the 

 flattening that marks so many other bones. In its general configura- 

 tion, and many minor details of structure, the skull shows three, if 

 not four, strongly marked patterns, corresponding with and incontest- 

 ably substantiating three, at least, of the genera that Prof. Hyatt 

 has successfully established upon external characters. No one has 

 hitherto shown us what groups are probably generic, and what are 

 purely arbitrary. 



1 Certainly not a true tarsal ossicle, if the tarsus of Spheniscldce agrees in de- 

 velopment and structure with that of other birds; but this remains to be seen. 

 Cf. Morse's beautiful and invaluable researches into the carpus and tarsus of birds. 

 Ann. Lye. Nat. Hist., N. Y., 1871. 



