1875.] PROF. W. H. FLOWER ON THE MUSK-DEER. 183 
There is, however, little practical difficulty in deciding, by an ex- 
amination of the molar teeth of any of the existing Ruminants, to 
which section it belongs; and, judged by this test, Moschus is 
decidedly brachyodont, and thereby resembles the Cervine members 
of the group, though in some details, as has already been mentioned, 
it has slight peculiarities of its own. 
The best method, however, of testing the claims of Moschus to a 
definite position will be to take seriatim all the principal characters 
in which it shows variation from the average Pecorine type, and con- 
sider in which direction they severally tend. 
I. The absence of frontal appendages. This is a well-marked ex- 
ternal character, but one the significance of which has been much 
altered by Mr. Swinhoe’s discovery of Hydropotes, which, although 
its anatomy is not yet fully known, I think may he safely assumed 
to bea true Deer. It is certainly less aberrant than Moschus*. 
Even before the existence of other Deer without antlers was known, 
it might have been suspected that such appendages were really only 
of secondary importance in a natural system of classification, as they 
occur among existing Deer in such infinite variety of form and size 
without correlation with other structural modifications ; and as, more- 
over, paleontology teaches us that Deer (7. e. animals having all the 
osteological and dental characters of the group, as Dremotherium) 
abounded before the antlered forms came into being, it is by no 
means unreasonable to suppose that some of the recent members of 
the family might retain this primitive character‘. 
As one or more species of true Deer are without antlers in either 
sex, as all (Tarandus excepted) have noue in the female sex, and as, 
on the other hand, no Bovide are known without frontal appendages 
in the male and nearly all have them in both sexes, it follows 
that a ruminant, like AMoschus, wanting these parts is so far more 
likely to belong to the Cervine than to the Bovine section. The 
absence of antlers is no indication of special relationship to the T’ra- 
gulina any more than it is to the Camels, Pigs, or any of the early 
forms of the order. 
II. Dentition. The brachyodont character of the molar teeth, as 
lately mentioned, is some evidence in favour of Moschus belonging 
to the Cervine section, but not by itself conclusive; for even if we 
knew of no existing Bovine animal in this case, it would be quite 
possible to conceive of some member of the group retaining a character 
once common to all. 
* The still more recently discovered Lophotragus michianus, Swinhoe (P. Z.8. 
1874, p. 452), appears to be another Deer without antlers; but very little is yet 
known of its structure. 
+ Dremotherium is sometimes placed among the Tragulide, or rather the 
artificial group in which those animals as well as Moschus were included ; but 
in the majority of its dental and cranial characters it was a true Deer, of course 
somewhat generalized and in so far approaching the Tragulina. Gelocus was an 
older form, and retained the four premolars of the more primitive types. They 
both appear to belong to the stock from which the Pecora are descended after 
the ancestors of the Tragulina had branched off from it. The latter, as will 
be seen in the tabular view of the classification of the group (p. 189), are the 
lowest and least-modified of all the existing selenodont Artiodactyles. 
