12 R. W. ShurELDT, 
and so on, than it is when so compared with the same view of the 
skull in the Swans, Geese, or most ofthe Fuligulinae. In nearly 
all these forms, when the skull is so viewed, the backward-exten- 
ding median processes of the premaxillaries are clearly defined on 
a small area in the cranio-facial region by more or less open slitlets. 
This character is generally entirely absent in Netta rufina, in most 
of the genus Mar:la, and in many Swans and Geese. It is a 
prominent feature on the upper view of the skull of such a species 
as Chloephaga hybrida. Dendrocygna and most species of the genus 
Anas agree in this particular. 
In some of the smaller and medium-sized ducks, the frontal 
region on the superior aspect of the skull, between the upper mar- 
gins of the orbits, is extremely narrow, as is the case in the Teals 
(Querquedula), the Widgeons (Mareca), and in some of the Scaups 
(Marila). For instance, Maria marila has a skull 90 mm long; 
while the shortest transverse diameter in the same skull measures 
but 6 mm between the peripheries of the orbits on the superior 
frontal region. In fact, in this species the fairly well-marked 
supraorbital depressions, as exemplified in this particular skull, are 
separated mesially by the longitudinal furrow, which here is less 
than a millimeter wide. In this particular, again, Dendrocygna agrees 
with the more typical Anatinae, that is, the genus Anas. 
This is likewise true with respect to the posterior aspect 
of the skull, upon which, in the Tree-Ducks, in the typical species 
of the genus Anas, Marila valisineria, Chaulelasmus, and allied forms, 
we find a largely developed ‘supraoceipital prominence‘, with a rather 
large foramen opening into the brain-case on either side ofit. The 
size of these foramina are, as a rule, larger in Anas than in Dendro- 
cygna, in which latter genus they may occasionally be quite small; 
while in some swans (Cygnus olor) they are absent, and always 
very small in proportion in Chen and BDranta. 
The erotaphyte fossae never meet superiorly in the median 
line in any ofthe Anatidae, while the oceipital area is invariably 
clearly defined by a more or less raised and sharp limiting crest. 
This crest is best marked in any of the true River Ducks, and 
perhaps the least defined in Chen and such species as Chloephaga 
hybrida and its nearest congeners. 
In Chenonetta jubata of Australia, the foramen magnum is 
unusually large, as compared with the size of the bird; the supra- 
oceipital prominence small; the foramen on either side of it large 
