n regarded as the direct ancestor of the sabre-toothed
cats, or _Machaerodontinae_ (see MACHAERODUS); but it is possible that
_Palaeonictis_ may be off the direct line, and that the _Felidae_ are
sprung from _Amphictis_. Be this as it may, from another group of
creodonts, represented by _Vulpavus_ (_Miacis_), _Viverravus_
(_Didymictis_), and _Uintacyon_, is probably derived the Oligocene
_Cynodictis_, with a dental formula like that of _Canis_ or _Cyon_, a
perforation to the humerus, and an apparently undivided auditory
bulla; and from _Cynodictis_ the transition is easy to the _Canidae_.
It should be mentioned, however, that there is a group of North
American Oligocene dog-like animals, such as _Daphaenus_,
_Protemnocyon_, and _Temnocyon_, which agree with _Cyon_ in the
shortness of the jaws, and with that genus and _Speothos_ in the
cutting-heel of the lower sectorial. Possibly these genera may be
nearly related to _Cyon_. Other dog-like North American types are
_Oligohinis_, _Enhydrocyon_ and _Hyaenocyon_.
By means of the _Amphicyonidae_, as represented by the Middle Tertiary
genera _Proamphicyon, Pseudamphicyon_, and _Amphicyon_, in which there
were three upper molars, we have a transition from the
_Cynodictis_-type to the bear-group; one of the later intermediate
forms being the Lower Pliocene Old World _Hyaenarctus_, in which the
two upper molars are squared and foreshadow those of _Ursus_ itself.
In some unknown manner _Hyaenarctus_ appears to be related to
_Aeluropus_. An allied type is found in _Arctotherium_ of the South
American Pleistocene.
By the loss of the third lower molar and certain modifications of the
other teeth and skull, the Miocene genus _Plesictis_ may be derived
from _Cynodictis_, its dental formula being i. 3/3, c. 1/1, p. 4/4, m.
(1 or 2)/2. Now _Plesictis_ is nothing more than a generalized
representative of the _Mustelidae_. We have thus traced three out of
the four modern arctoid families to the _Cynodictis_-type. The
_Procyonidae_, or fourth family (apart from the Asiatic _Aelurus_ and
_Aeluropus_) are connected with the last-named genus through the North
American Oligocene _Phlaeocyon_, which is stated to be in almost every
respect intermediate between _Procyon_ and _Cynodictis_ while the
living _Bassariscus_ is stated to show closer signs of affinity with
_Cynodictis_ than with _Phlaeocyon_.
To deal with fos
|