32 
MESOZOIC VERTEBRATA. 
to five digitiform processes. Two of these are larger and longer than the 
others, and externally on the right side is a shorter one. Outside of this 
is a larger process, whose extremity is recurved so as to be subparallel with 
the longer processes, and which was connected with another bone by an 
articular surface. This information is derived from Professor Newberry’s 
notes made in the field. It is probable that this bone is the sternum, and 
that the articulation mentioned is costal. It is not certain whether the 
longitudinal median line passes through a sinus or a digitation; but a pro¬ 
jection of the surface of the plate, which is probably median, is opposite 
one of the latter. Supposing, then, that the sternum is produced into a 
median posterior process, we find a resemblance to the corresponding ele¬ 
ment in many birds not heretofore known among reptiles. There are in 
that .case three postero-externally directed processes on each side, of which 
the two posterior are free. Another interpretation might be that it is a cora¬ 
coid with anterior digitations. In this case, the articulation above mentioned 
would be anomalous. The number of digitations is too great for this ele¬ 
ment, and the space remaining for contact with the sternum is too small. 
Another large flat bone approximates a right-angled triangle in form, 
the length greatly exceeding the width. The right angle is massive and 
produced, and is evidently the point of connection with the other parts of the 
skeleton. The bone is flat on one side and convex on the other, and can 
only be identified with probability with the scapula of a Dinosaurian reptile. 
The large size of the_ anterior limb, which might be inferred from this 
scapula, is justified by the humerus, which is preserved in almost perfect 
condition. This humerus is one of tlie longest, and is distally the most con¬ 
tracted known in the Dinosauria. The proximal extremity is of the form 
usual in that order. A sliort distance below the head, the section is T- 
shaped, with one end of the transverse limb shorter than the other. The 
ridge of which this limb is a section is almost wanting at the head, wind] 
is thus T-shaped. The limb representing the stem of the T, is stouter than 
the others, and forms the summit of a massive column, which soon sinks 
into the shaft. Its free extremity's obtuse and rounded, and, though repre¬ 
senting the head, does not rise above the level of the other crests or tuber- 
