CLASSIFICATION 87 
are known, do not fall into either of these groups, but are in some 
respects annectant forms, we have placed them provisionally, at 
least, in a third group by themselves, named Archeoceti. There 
is nothing known at present to connect the Cetacea with any 
other order of Mammals; but it is quite as likely that they are 
offsets of a primitive Ungulate as of a Carnivorous type, or perhaps 
of a still more generalised mammalian stock. 
The remaining Eutherian mammals are clearly united by the 
characters of their teeth, being all heterodont and diphyodont, with 
their dental system reducible to a common formula. 
Although older views of, the relationship of Ungulate mammals 
expressed by the terms Pachydermata, Runvinantia, and so forth, still 
linger in some corners of zoological literature, no single point in 
zoological classification can be considered so firmly established as the 
distinction between the Perissodactyle and Artiodactyle Ungulates ; 
both being in the existing fauna of the world perfectly natural 
and distinctly circumscribed groups. The breaking-up of the latter 
into four equivalent sections, the Pecora, Tylopoda, Tragulina, and 
Suina, is equally in accordance with all known facts. Less certain, 
however, is the association of the Proboscidea and the Hyracoidea 
with the true Ungulates. By many zoologists they are each, 
although containing so very few existing species, made into distinct 
orders; and much is to be said in favour of this view. The 
discovery, however, of a vast number of extinct species of Ungu- 
lates which cannot be brought under the definition of either Perisso- 
dactyla or Artiodactyla, and yet are evidently allied to both, and 
to a certain extent bridge over the interval between them and the 
isolated groups just mentioned, make it necessary either to intro- 
duce a number of new and ill-defined ordinal divisions, or so to 
widen the scope of the original order as to embrace them all, 
considering the Elephants and the Hyraces as representing sub- 
orders equivalent to the great Perissodactyle and Artiodactyle groups. 
It is the latter alternative that we have adopted. 
The Rodentia, although generally presenting a low grade of 
development, are a very specialised and distinct group. The 
position here assigned to them would accord with apparent relation- 
ships with the Ungulates, through the Elephant on the one hand 
and the extinct Typotherium on the other. 
In the present state of the fauna of the earth, the Carnivora 
form a very distinct order, though naturally subdivided into two 
groups, the members of the one being more typical, while those of 
the other (the Pinnipedia) are aberrant, having the whole of their 
organisation specially modified for living habitually in the water. 
The Insectivora comprise various lowly organised and generalised 
forms, exhibiting considerable divergence of character, and ap- 
parently connected through transitional extinct species with the 
