1393.] Fossil Mammalia of Southern Patagonia. 447 
and four are back molars. The first have but a single cusp, 
but they have a posterior heel, well-developed in the last. 
The first three true molars increase regularly in size from the 
first to the third, and have the anterior internal cusp atrophied, 
which gives them a cutting form, especially noticeable in 
in the third. The fourth is much reduced, and occupies the 
same position as in the genera Thylacynus, Prothylacynus and 
Amphiproviverra. Perhaps Dynamictis, of which only the 
lower jaw is known, is a synonym for Borhyæna. In any case, 
the latter animal was a formidable Carnivore, as dangerous as 
our tigers, and which it is hard to classify. It may belong 
with the placental Carnivores, or among the marsupial flesh- 
eaters (Dasyures), or with the ambiguous forms known as 
Creodonts. 
It would take too long to speak of the numerous fossil Eden- 
tates of the Santacruzian formation and of the many peculiari- 
ties they present. However, I cannot resist mentioning one of 
the greatest novelties of the new family of the Metopotheriide, 
which comprises the genera Metopotherium, Pelecyodon, and 
Zamicrus. In this group, each of the horizontal branches of 
the mandibula is formed of two distinct bony parts, one in 
front, the other behind. These two pieces are united by a 
suture which commences at the alveolar border, very nearly 
in the middle of the dental series, and is directed obliquely 
back and down, terminating in the lower border, almost just 
below the last tooth. This is evidently a character inherited 
from the reptiles. The same suture can be seen, but not 
quite so plainly, in the mandible of some of the Orthotheriide. 
I remarked, in the beginning of this article, that I had 
found some forms characteristic of the Puerco of North Amer- 
ica in the beds much more recent than those of the Santa- 
cruzian epoch. In fact, I have received from the Parana, a 
fragment of a upper maxillary, which shows several tooth 
_ sockets and one tooth in place. The latter corresponds to a 
third premolar. It is impossible to separate this specimen 
generically from Periptychus (Cope). The tooth resembles 
perfectly the third upper premolar of Periptychus rhabdodon 
and shows the same grooves or wrinkles in the enamel which 
