In the genus Nimravus the dental formula is still further reduced. Another premolar of the lower jaw has gone, the formula being thus I 3/3 C 1/1 Pm 3/2 M 1/2. Nimravus gomphodus was a Carnivore about the size of a Panther. It has no third trochanter upon the femur, which process is present in the corresponding bone of Dinictis. Pogonodon was an equally large animal in which the premolars were three in each jaw, but the molars have become reduced to one in the lower, as they have in this and other genera in the upper
jaw. Finally, Hoplophoneus has acquired the dentition of existing Cats.
The Machaerodons, however, show examples with a yet more reduced dentition than that of the most reduced existing Cat, viz. the Lynx, which has only two premolars in each jaw and one molar. In Eusmilus the molar in both jaws is single, and there is but one premolar in the lower jaw.
The genus Machaerodus itself, which appears to include Smilodon, is referred by Cope to the true Cats, and not to the Nimravidae, as he terms the family which we have called here the Machaerodontidae. These creatures are known as "Sabre-toothed Tigers," and were of very wide distribution, occurring in South America as well as in Europe and North America. "As nothing," remarks Professor Cope, "but the characters of the canine teeth distinguished these from typical felines, it is to these that we must look for the cause of their failure to continue. Professor Flower's suggestion appears to be a good one, viz. that the length of these teeth became an inconvenience and a hindrance to their possessors. I think there can be no doubt that the huge canines in the Smilodons must have prevented the biting off of flesh from large pieces, so as to greatly interfere with feeding, and to keep the animals in poor condition. The size of the canines is such as to prevent their use as cutting instruments excepting with the mouth closed; for the latter could not have been opened sufficiently to allow any object to enter it from the front. Even when it opens so far as to allow the mandible to pass behind the apices of the canines, there would appear to be some risk of the latter being caught on the point of one or the other canine, and forced to remain open, causing early starvation. Such may have been the fate of the fine individual of the S. neogaeus, Lund, whose skull was found in Brazil by Lund, and which is familiar to us through the figures of de Blainville."
Machaerodus is placed among the Felidae on account of the fact that the condyloid and carotid foramina unite with the foramen lacerum posterius. But as in at least one species, M. palmidens, there is an alisphenoid canal, which, however, has disappeared in the more recent American forms, it seems permissible to retain the genus in the family Machaerodontidae though its existence reduces the differential character of that family to a minimum. The genus goes back to the Eocene.
Fam. 3. Viverridae.—The Civets, Genets, and their kind differ from the Cats in a number of points. They form, however, by no means so uniform an assemblage as do the Cats; so that the difficulty is, as Dr. Mivart has remarked, not to divide them into sub-families, but to avoid making too many. But before proceeding to subdivide the family we shall describe the characters of the family and contrast them with those of the Felidae.
All the Viverridae are comparatively small creatures. The head and body are more elongate than in the Cats. The fingers and toes are generally five; but there are some (e.g. Cynictis) where the formula of the toes is as in the Cats, i.e. four on the hind-foot. In the Suricate the fingers are also reduced to four. The claws are perhaps never completely retractile,[[266]] and often are not at all so. The dental formulae of the genera differ considerably; but in the majority there are more teeth than in the Felidae. The well-known sharp-pointed, conical papillae of the Cat's tongue are not present. The majority have a scent gland beneath the tail, from which the perfume civet is derived. There are a number of osteological characters which differentiate the two families; thus the alisphenoid canal is sometimes present. The bulla is divided, as in the Cats, but is externally constricted.
It seems clear from some at any rate of the characters, i.e. the more complete dentition, the five-fingered hands and feet, the non-retractile claws, etc., that the Civets are on a lower level of specialisation than are the Cats.
Sub-Fam. 1. Euplerinae.—The genus Eupleres is in many ways the most aberrant type of Viverrid, and is placed in a sub-family, Euplerinae. Its salient feature is the very peculiar dentition: peculiar in the small size of the canines, the canine-like character of the anterior premolars, and the resemblance of the premolars to molars. In some of the characters of the teeth, Eupleres is Insectivore-like, and was formerly grouped with that family. There are four premolars and two molars in each jaw on each side. It has five toes upon both fore- and hind-limbs; the skull is very slender. It has no alisphenoid canal. The only species, E. goudotii, is of an olive-grey colour, with dark bands across the shoulders in the young. The nose and upper lip are grooved. There are no scent glands. It appears to burrow in the ground, and possibly contents itself with a diet of worms. Eupleres is a