— 305 — 
them, reproduced in half-tone, are shown :n plates II, Ill, and IV. The 
casts were made by Mr. Otto Falkenbach of the American Museum 
of Natural History. 
It was obvious, almost at the first glance, that the material 
comprised the remains of specimens, apparently all of one species, of 
asmall aquatic reptile. The resemblance to Stereosternum tumidum, 
Cope, was noted and it was supposed that the present remains were 
assignable to that species, until careful study revealed a number of 
differences, the most noteworthy being in the number of presacral 
vertebrae, the present form having but 29, while Stereosternum has 34. 
Further careful observation demonstrated divergence from Ste- 
reosternum tumidum in the following characters : 
1). The skull is relatively larger, being longer than the neck instead 
of equal in length. 
2). There are only 11 cervical vertebrae instead of 12, and only 
18 thoraco-lumbars instead of 22, thus five less presacrals. 
3). The pubic foramen of Sfereosternun is here represented by a 
deep notch, as in the South African genus Mesosaurus (Broom, 1904), 
and since there is also agreement with that form in the number of cer- 
vicals, and no known structure on which to base generic separation, 
I place the present form in the genus dlesosaurus and propose the 
specific name brasiliensis. 
The general form of Azesosaurus brasilicnsts is well shown in 
text-figure 1, and Plate I. is an attempt at a restoration. The jaws 
and dentition, and especially the limbs, bear indubitable evidence of 
the aquatic habit of A/esosaurus, and the very numerous long needle- 
like teeth are well-adapted to the capture of small fishes and other active 
aquatic animals, though not adapted to crushing hard-shelled mol- 
luses. The position of several of the fossilized skeletons shows that 
the neck, though not extremely long, was very flexible. 
Comparison of the positions of the limbs in many examples of 
Mesosaurus and Stervosternum shows that in these Proganosaurians 
the greatest mobility was at the shoulder joint, while the elbow is 
usually not at all or but slightly flexed, the limbs standing out 
straight at rigbt angles to the Jody or more commonly directed 
packward along the sides. (Pl. IL, fig. 2.) 
In every case tle small manus is placed with the palm downward. 
It is probable, I think, that the chief function of the fore-limbs in 
5560 39 
