“Tree-Ducks” of the genus Dendrocygna. 41 
FÜRBRINGER calls the process on the superior celavicular border 
I refer to above, the processus acrocoracoideus clavi- 
culae, and figures it for Mergus. The long, pointed, free end of 
the clavicle he designates as the processus acromialis clavi- 
ceulae. 
For the Anseres, the free end of either clavicular limb, when 
articulated as in life, rests its outer surface at a point about oppo- 
site the processus acrocoracoideus claviculae, while the 
processus acromlalis claviculae rests for a more or less 
distance upon the supero-median border of the anterior extremity of 
the corresponding scapula. As thus articulated, a most perfect 
foramen triosseum is formed for the passage of the tendon of 
the pectoralis secundus muscle‘) 
This is the mode of articulation of the os furculain Dendro- 
cygna bicolor and autumnalis, and in the Anseres generally. 
The position of the os furcula, with reference to the distance 
it is from the coracoids and sternum, when normally articulated as 
in life, varies to some extent among Ducks, Geese and Swans. For 
example, it is rather close to these bones in Üereopsis novae-hollandiae; 
while in such a species as Hymenolaemus malacorhynchus it is far 
removed from them (Fig. 95, Pl. 12 and Fig. 96, Pl. 12). 
Always large and well developed in the Anseres generally, 
the scapula in Dendrocygna is a long, narrow, arched bone, its 
chord having an average length of some 64 millimeters. It articu- 
lates with the os furcula, as described in the foregoing paragraph, 
and with the entire width of the precoracoid process of the 
coracoid. Distally, it is acutely truncate, with the distal apex formed 
into a rounded little nib. For the first two-thirds of its length, its 
external border is rounded, — the remaining distance to the apex 
being sharp, as is also the internal margin for its distal half, the 
rest being somewhat rounded. The shaft of the blade curves out- 
ward, thus causing the outer border to be concave, and the inner 
correspondingly convex. 
So far as I have examined them, the bones of the pectoral arch 
in the Anatinae and the Merginae are allnon-pneumatic; 
in the Cygninae the scapula alone is non-pneumatic; while in 
Chen and Cereopsis all the bones of this arch are completely 
l) SHUFELDT, R. W., The myology of the Raven, London 1890, 
p- 72, and fig. 32 on p. 89. 
