In the South American Quadrumana, the number of teeth is increased to thirty-six (art. QuAnatimANA,Vol. IV. p.210.; by an addition of one tooth to the molar series on each side of both jaws. It might be concluded, d priori, that as three is the typical number of true molars in the placental Mani malia with two sets of teeth, the additional tooth in the Cebince would be a premolar, and form one step to the resumption of the normal number (four) of that kind of teeth. The proof of the accuracy of this inference is given by the state of the dentition in the young Cebus in fig. 589., which corresponds with that of the human child in fig. 58.1., i. e. the whole of the deciduous dentition is retained, and the first true molar (m. 1) is in place on each side of both jaws. The germs of the other teeth of the permanent series are exposed in the upper jaw ; and the crown of a premolar is found above the third molar in place, as well as above the second and first. As regards number, therefore, the molar series, in Cebus, is interme diate between that of Mnstela (fig. 580., IV.) and Felis (ib. V.) ; the little premolar p. i. in illustela tells plainly enough which of the four is wanting to complete the typical number in the South American Monkey, and which is the additional premolar distinguishing its dental formula from that of the Old World monkeys and man. By reference to Prof. Vro lik's article (QUADRUMANA) it will be seen that the eighth genus, including the little Marmoset monkeys (Hapale, Ouistiti), only the same number of teeth as the monkeys1-1 of the Old World, viz. 32, — 1-1' 5-5 „ But the difference is much greater than this numerical conformity would intimate. In a young Jacek= penicillatus I find that there are three deciduous molars displaced by three premolars, as in the other South Ame rican Quadrumana, and that it is the last true molar, m. 3, the development of which is sup pressed, not the premolar p. 2, and thus these diminutive squirrel-like monkeys actually differ from the Old World Quadrumana more than the Cebida. i.e. they differ not only in having 1 four teeth (p. 2 F. which the monkeys of the Old World do not possess, but also by 1-1 wanting four teeth (ni. 3 which those monkeys, as well as the Cebidce, actually have. It is thus that the investigation of the exact homologies of parts leads to a recognition of the true characters indicative of zoological affinity.
3-3 Most of the Lenlurinee have p. at.
3-3 toaether with remarkable modifications 3-3' 6 of their incisive and canine teeth, of which an extreme example is shown in the pectinated tooth (fig. 556.) of the Galeopithecus. The inferior incisors slope forwards in all, and the canines also, which are contiguous to them, and very similar in shape. In the Chirogaleus these canines are entered as in cisors in the dental formula of the genus ( Vol. IV. p. 215), and the laniariform premolar (p. 2) is entered as a canine ; M. Vrolik also describes four teeth on each side of the upper jaw, and four on each side of the lower jaw, as true tuberculated molars. They have tuber culated crowns, but the value of shape as a character is too small to permit our accepting so great an anomaly without the requisite proof of their order of development and suc cession.
Even in the hoofed quadrupeds with toes in uneven number (Perissodaetyk), whose premolars, for the most part, repeat both the Jinni and the complex structure of the true molars, such premolars are distinguished by the same character of development as those of the Artiodactyla, or Ungulates with toes in even number ; although in these the premolars arc distinguished also by modifications of size and shape. The complex ridged and tuber
culate crowns of the second, third, and fourth grinders of the Rhinoceros, Hyrax (fig. 590.), tooth be determined, and its proper symbol applied to it.
In pl. 136, fig. 5, of my Odontography, the three posterior teeth of the almost uniform grinding series of the horse's dentition are thus proved to be the only ones entitled to the name of " true molars ;" and, if any one should doubt the certainty of the rule of count ing, by which the symbols, p. 4, p. 3, and p. 2, are applied to the three large anterior grinding teeth (ih. fig. 19), which are commonly the only premolars present in each lateral series of the horse's jaws, yet the occasional retention of the diminutive tooth, p. 1 (ib..fig. 6). would establish its accuracy, whether such tooth be regarded as the first of the deciduous series unusually long retained, or the unusually small and speedily lost successor (p. 1) of an abortive d 1.
The law of development, so beautiful for its instructiveness and constancy in the pla cental Dip/ayodunts, is here illustrated in the little Hyrax (fig. 591.), in which the d. 1 is normally developed and succeeded by a per manent p. 1, differing from the rest only by a graduated inferiority of size, which, in regard to the last premolar, ceases to be a distinction between it and the first true molar.
The elephant, which by its digital characters belongs to the odd-toed, or perissodactyle, group of Pachyderms, also resembles them in the close agreement in form and structure of the grinding teeth representing the premolars, and horse, no more prove them to be true molars, than the trenchant shape of the lower carnassials of the lion proves them to be false molars. It is by development alone that the primary division of the series of grinding teeth can he established, and by that character only can the homologies of each individual with those that answer to the true molars of the Hyrax, Tapir, and Rhinoceros. The gi gantic Proboscidian Pachyderms of Asia and Africa present, however, so many peculiarities of structure, as to have led to their being located in a particular family in the Systematic Mammalogics. And this seems to be justified by no character more than by the singular seeming exception which they present to the Diphyodont rule which governs the dentition of other hoofed quadrupeds. In fact, the elephant, like the Dugong, sheds and replaces vertically only its incisors, which are also two in number, very long, and of constant growth, forming tusks, with an analogous sexual dif ercnce in this respect in the female of the Asiatic species. The molars, also, are suc cessively lost, are not vertically replaced, and are reduced finally to one on each side of both jaws, which is larger than any of its pre decessors. These analogies are interesting and suggestive in connection with the other approximations in the " Sirenia" to the pa chydermal type, which I have pointed out in the " Proceedings of the Zoological Society." * The dentition of the genus Elephas, the sole existing modification of the once numerous and varied Proboscidian family, includes two long tusks (fig. 592.), one in each of the lntermaxillary bones, and large and complex molars (ib. in. 3, 4, and 5) in both jaws : of the latter there is never more than one wholly, or two partially, in place and use on each side at any given time; for, like the molars of the Mastodons, the series is continually in progress of formation and destruction, of shedding and replacement ; and in the ele phants all the grinders succeed one another like true molars horizontally, from behind forwards.