398 ME. W. p. PTCEAFT ON THE 



The interorbital septum is largely fenestrated. The mes- 

 ethmoid does not extend as far forwards as the free end of 

 the premaxillary limb of the lachrymal. There is no ossified 

 antorbital plate, and no nasal septum. The anterior nares are 

 of considerable size — larger than in Somateria. 



The hyoid bones differ but little from those of typical ducks. 

 The basi-hyal is of great size, shows traces of its paired origin, 

 and is pierced at the base by a large cordiform fenestra. The 

 1st and 2nd basi-branchials are fused. 



The skull of Biziura may be distinguished from that of all 

 other Anserine birds by the curiously truncated form of the 

 upper jaw ; the free lachrymal, which is further remarkable for 

 the great length of the descending processes and the shortness of 

 its horizontal limbs ; the short blunt postorbital process ; the 

 peculiar shape of the grooves for lachrymal gland, which do not 

 meet in the middle line, nor extend backward beyond the post- 

 orbital processes ; the great size of the anterior nares ; and the 

 perforate interorbital septum. 



In the curiously forward tilted position of the quadrate, and 

 the angulated quadrato-jugal, Biziura resembles Somateria and 

 the Mergansers, but in Biziura this feature is most marked. 



Although Biziura at the present day appears to derive much 

 of its support from the sea, it would seem that it is essentially a 

 freshwater species. And this may be inferred from the small 

 size of the supra-orbital groove for the lachrymal glands, which, 

 in Somateria and (Edemia for example, extend far backwards as 

 shallow depressions over the roof of the skull, and in Tachyeres 

 almost touch the lambdoidal ridge. In Tachyeres and Somateria 

 this glandular area is increased by the development of the orbital 

 processes of the lachrymal, which project outwards, upwards, and 

 backwards after the fashion of a pair of horns. Deep supra- 

 orbital depressions, more sharply defined, occur in the Sphenisci 

 and the Colymbi, the Petrels and Gulls, and, less well defined, in 

 the Plovers, which swim and dive only on relatively rare occasions. 

 Thus there seem to be good grounds for associating the develop- 

 ment of this groove with the necessity for a large supra-orbital 

 gland in marine diving birds. This seems to be supported by 

 the fact that in the Mergansers, which are freshwater diving 

 ducks, this groove is wanting and the gland only feebly developed. 



