“Tree-Ducks’” of the genus Dendrocygna. 25 
sphenotic process in Cygnus olor; though, as a matter of fact, the 
lower free portion of the lacrymal in the Swan is more like what 
we find in species of the genus Anas than it is in Dendrocygna, 
where this part is narrow, long and somewhat slender. 
With respect to the mandible in the Anatidae, I have 
already contributed, in my “Osteology of Birds”, cited above, a brief 
account of this bone as it occurs in a number of anserine forms. 
Illustrations will also be found there, giving the mandibles of Spatula 
clypeata and COlangula islandica, seen upon upper view (p. 280, 281, 
figs. 19 and 20). 
There will be found upon the Plates in the present contribution 
numerous figures of the mandible, as we find it among the ducks, 
geese and swans. These are so clear that the bone hardly requires 
any further description. 
Dendrocygna has a lower mandible agreeing, in its several 
characters, with any of the typical ducks, and especially with those 
of the genus Anas (Compare Fig. 12, Pl.3; Fig 110, Pl. 16; Fig. 27, 
Pl. 5; Fig. 38, Pl. 6, and others). On the other hand, the man- 
dible of Dendrocygna autumnalis (455) has identically the same 
character, in every detail, as that bone has as we find it in Olor 
columbianus (Fig. 40, Pl.7; Figs. 42, 43, Pl. 7). In the swan, however, 
the vascular longitudinal groove on the outer side of the ramus, from 
the juncture of the middle and posterior thirds to the symphysis, is, 
comparatively, much deeper than it is in the “Tree-Duck”. But this 
is not invariably the case, for this groove is deep in specimens of 
D. autumnalis. It is also very well marked in Mallards and Canvas- | 
backs, and, indeed, in the majority of the Anseres. They nearly 
all, too, have a deep and extensive fossa, that opens to the inner 
side of either backward-extending angular process, which fossa 
underlies the entire articular facet (and beyond) for the quadrate. 
This cavity is small in Dranta, Chloephaga, Tachyeres cinereus, and 
is not present in Cereopsis. 
Branta c. hutchensi, Cereopsis, Chloephaga poliocephala, Tachyeres 
cinereus and some other species have the backward-extending pro- 
cesses of the articular ends of the mandible sabre-shaped, and only 
gently curved (Figs. 11, 14 and others of Plate 3). Polysticta stelleri 
has these processes of a most remarkable form, each being a long, 
deep, quadrilateral plate, scarcely turned up at all at the posterior 
ends. I know of no other duck that has them of this form (Fig. 87, 
