262 PROF. HUXLEY ON THE CRANIAL AND L^^P'- 6, 



with small frontal sinuses, are so slightly separated from the lower 

 Alopecoids, that it is hard to say whether we have any right to look 

 for a Thooid representative of Otocyon or not. It is quite as reason- 

 able to suppose that Oiocyon is the nearest living representative of 

 the primitive type of the Canidee, whence all the rest have been de- 

 rived, in the first place, by the differentiation of the Thooid from the 

 Alopecoid series, and, in the second, by the occurrence of corre- 

 sponding series of modifications leading up to the Eox on the one 

 hand and to the Wolf on the other. 



11 . If this view of the facts is correct, the key to the morphological 

 relations of the whole of the Canidse must lie in the determination of 

 the affinities of Otocyon. The facts hitherto considered primarily 

 appear to me to suggest looking in two directions — in the first place 

 towards the Procyouidse, and in the second towards the Didelphia. 



In studying the Canidae it is impossible not to be struck by the 

 wonderful persistency of the fundamental patterns of the sectorial 

 teeth and of those which follow them. This singular uniformity can 

 hardly be accounted for by adaptation to similar modes of life ; for 

 the pattern is as distinctly marked in C.jubatus and C. procyonoides, 

 which live largely upon fruits and roots and never attack large animals, 

 as in any of the more purely carnivorous Canidse. It must therefore 

 be regarded as a morphological fact of fundamental importance, and 

 the best guide to the immediate affinities of this group of animals. 



Now, in Bassaris we have a procyonine form, the teeth of which 

 are extraordinarily similar to those of C. zerda, if we suppose the little 

 posterior lower molar of the Fennec suppressed. The posterior 

 margin of the bony palate is on a level with the hindermost molar 

 teeth, and therefore does not extend further back than in the ordi- 

 nary Canidae. There are no frontal sinuses ; and the ethmoid is high. 

 In ^lurus, again, the patterns of the teeth are essentially canine, 

 though inclining in some respects towards the Bears : the frontal 

 sinuses are large, the ethmoid low, and the cranial cavity has a com- 

 pletely Thooid contour. In this genus, as Prof. Flower has pointed 

 out, an alisphenoid canal is present. The small flattened bulla, with 

 its long meatus, is unlike that of the Dogs ; on the other hand, the 

 carotid canal is long, and its posterior aperture opens into a depres- 

 sion common to it and the foramen lacerum posterius. The bony 

 palate extends considerably further back than in any existing canine 

 animal. 



In I^asua the fourth premolar above is triangular ; but a small 

 second inner cusp is beginning to appear behind the large one. In 

 Procyon this cusp has increased so much that the crown of the tooth 

 is quadrangular. In this genus there is a small cingulum on the inner 

 side of the first and second molars, which thus retain a resemblance 

 to those of the Dog. In Nasua, however, it is no longer visible. 

 In both these genera a line joining the inner and outer cusps of the 

 lower sectorial teeth is almost transverse to the axis of the tooth, 

 and the inner cusp is higher than the outer, as in Otocyon. 



I find the proportional lengths of the teeth in Nasua and Procyon 

 to be as follows : — 



