90-24 INDIAN TERTIARY AND POST-TERTIARY VERTEBRATA. 
skull of E. hemionus. The relative proportions of the skulls of the above mention- 
ed species may he estimated by taking the length of the molar series (inclusive of 
the persistent milk-molar) as unity, and seeing how many times this unit is contain- 
ed in the interval separating the last true molar and the foramen magnum. This 
gives the following results : — 
E. onager = 1'07 1'07 
„ hemionus . . =1‘15 1'15 
„ sivalensis . . . .... . . ’. US 1*13 ) „ „ 
gg 1 . 16 j average l‘14c 
„ cahallus = 1'26 ■) 
A distinct trace of a ‘ larmial * cavity is observable in the skull of E. siva- 
lensis, as was first pointed out by Professor Huxley in the passage already cited. 
No living horse shows any trace of such cavity, though it is well developed in the 
hippotlieres. 
The Siwalik skulls are not sufficiently perfect to admit of any closer comparison 
with those of living species ; the points already observed, however, with the excep- 
tion of the doubtful premaxillse, certainly indicate a considerable resemblance 
between the crania of E. sivalensis and E. hemionus , coupled with the retention by 
the former of certain ancestral characters, which have been lost in the latter. 
Mandible. — Two views of the greater portion of the horizontal ramus of the 
right side of the mandible of a Siwalik horse are given in the “ Eauna Anti qua Siva- 
lensis ” (plates LXXXI, fig. 4 ; LXXXII, fig. 2) 1 under the name of Equus siva- 
lensis. As similar jaws are not uncommon in the topmost Siwaliks, where E. 
sivalensis seems to be the commoner species, it is very probable that this reference 
is correct, though it cannot be considered as absolutely certain. 
A very similar specimen, comprising the two rami, is in the collection of the 
Indian Museum (No. C.184), and was formerly in the collection of the Asiatic Society 
of Bengal ; its description by Dr. Ealconer is given in the “ Palaeontological 
Memoirs.” 2 The two foregoing specimens show that the lower jaw of E. sivalensis 
is of great vertical depth, and that the £ diastema 7 is shorter than in the common 
horse, and thereby comes nearer the length of that of the kiang. The following 
table exhibits the chief dimensions of the two Siwalik jaws (a, British Museum, 
b, Indian Museum specimen) and the corresponding dimensions of the jaws of E. 
caballus and E. hemionus ; — 
E. cab. E. siv. E. hemi. 
Length of molar series .... 
Depth „ jaw behind last molars 
„ „ behind last premolar . . 
M „ at commencement of diastema. . 
Length „ diastema 
6 9 7-65 7-1 6 3 
4-35 4-9 5-1 3-7 
3-1 3-85 3-84 32 
22 2-85 2-8 2U5 
3-6 ... 2-6 2-4 
1 In the description of the plates it is not mentioned that both figures are taken from the same specimen ; the- 
second figure is reversed. 
2 Vol. I, p. 186 
