NEW RUMINANTS FROM THE SIWALIKS. 
7—111 
specimen represented in fig. 4. In the same tooth the external enamel-fold extends 
as near down to the root as in the latter specimen ; and there is accordingly a great 
probability tliat the two specimens belong to the same species. The young cranium 
is very similar in general contour to the cranium of H. hirsutirostris and II. 
cristata, having the characteristic convex profile ; which is wanting in the smaller 
porcupines. 
Distinctness and a^nities. — The foregoing comparisons lead to the conclusion 
that the Siwalik porcupine is specifically distinct from all the living species of the 
genus, while it is probably also distinct from the described fossil forms, though 
apparently coming nearest, in one respect, to II. refossa, and perhaps to 71. veniista. 
The characters in which the Siwalik form differs from the existing species are 
precisely those which might be expected to occur in an early, or generalized, form, 
and it is therefore quite possible that it may be the ancestor of the large existing 
Indian species. The provisional name 77. sivalensis may be retained for the Siwalik 
species. 
Distribution. — If both the specimens described above belong to the same species, 
the range of that species extended from the typical Siwalik Hills to the Punjab. 
Order: UNGULATA. Sub-Order: ARTIODACTYLA. 
Section : RUMINANTIA. 
Introductory. — In the preceding volumes of this worlP most of the Siwalik 
ruminants have been described and figured. The present part contains descriptions 
of a few new forms, and a revision of some of those previously described. 
The retention of the generally discarded term Ruminantia, as comprehending 
the modern Pecora, Tylopoda, and Tragulina, is adopted in view of the difficulty of 
referring many of the fossil forms to one or other of those groups. That the term 
Ruminantia is incapable of definition the writer is fully aware ; but as all divisions 
in palseontological zoology must be arbitrary in cases where a large number of forms 
belonging to allied groups are known, and as it is highly desirable to divide the 
Artiodactyla into a more specialized and a less specialized group, no valid objection 
to the use of the term can be drawn from this source. 
Family : BO VIDjB.^ 
Genus : OREAS, Desmarest. 
Distribution. — The genus is confined at the present day to south and tropical 
Africa, and includes only two species. It has not hitherto been recorded in a fossil 
condition. 
Species : Oeeas (?) lattdens, nobis. 
Syn. Cervus latidens, nobis. 
History. — In the first volume of the present work,® an upper and lower molar 
1 Vol. I., pta. 2, 3, 4. Vol. II., pts. 4, 5. 
2 In accordance witR Prof. Flower’s classification this family is taken to include all the cavicorn ruminants except 
Antilocapra. 3 Page 65, plate VIII., figs. 4, 6, 7, 10 (1876). 
