24 MR. W. H. FLOWER ON THE [Jan. 14, 
outer division, in which is the opening of the eustachian tube (e), and 
an inner one, of which the anterior end is a cul-de-sac, as in Felis ; 
but this septum only extends through about one-fourth or one-third 
of the entire cavity, so that the two chambers communicate most 
freely. In the hinder part of the inner chamber are a few irregular 
projecting bony ridges. The bulla is developed as in the Cats, from 
two parts, an outer true tympanic, and an inner cartilaginous portion. 
At birth ossification has not commenced in the latter, and appears 
in the former only as the horseshoe-shaped tympanic ring. 
The carotid canal (car) is complete, and of tolerable dimensions ; 
but its external opening is not visible on the surface of the bulla, being 
deep in the foramen lacerum posticum (/); the course of the artery 
is similar to that which it takes in the Bears. The paroccipital pro- 
cess (p) is long and prominent; and its anterior surface is applied 
closely to the back part of the bulla, but to a less extent than in the 
Cats, as the process is more compressed. The mastoid (m) is dis- 
tinct, but slightly developed. The condyloid foramen (c) is con- 
spicuously situated on the ridge passing from the paroccipital to the 
condyle, and is quite distinct from the foramen lacerum posticum 
(1). The glenoid foramen (g) is very large. The alisphenoid 
canal (a a’) is present. 
It is clear that, with these cranial characters, which are found dis- 
tinctly developed (with some modifications to be noticed hereafter) 
in all the members of the family Canidae, the Dogs cannot be placed 
in either of the great primary groups as above defined. They are in 
fact as nearly as possible intermediate between the two. The general 
form of the bulla and the presence of a septum, though imperfect, 
incline to the Aluroid type; while the position and development of 
the carotid canal, of the condyloid foramen, and of the glenoid fora- 
men are Arctoid characters. The form of the paroccipital and 
mastoid-processes, and the length of the meatal lip, are intermediate. 
Though I agree with Mr. Turner in making the Dogs a distinct 
primary group, which might be termed Cynorpea, I differ from him 
in placing them at one end of the series and the Bears at the other. 
The Dog appears to me to be the most central or generalized form of 
the whole order as at present existing, at least as far as the structure 
of the cranium is concerned. The comparative length of the folded 
czecum is a special peculiarity*. The reproductive organs belong 
neither to the Arctoid nor to the Ailuroid type, but partake of some 
of the characters of each. In the absence of Cowper’s glands, and 
* The length of the caecum in the Dogs, among other characters, induced 
De Blainville to place them at the end of the series of which the Bears were at the 
head, the Cats, with moderate cecum, intervening. But may not the long cecum 
be rather a retention of general mammalian structure, which has been completely 
or partially deviated from in the Bears on the one hand, and the Cats on the other, 
both more specialized groups ? 
It should be remarked that the Dogs retain more nearly the typical number of 
teeth than any other Carnivores, and that some of the earliest known forms of the 
order appear, as far as can be gathered from their fragmentary remains, to belong 
to the group. Thus the Early Miocene Amphicyon had the teeth of a modern 
Dog, with an additional upper molar, completing the typical mammalian dentition. 
