88-22 INDIAN TERTIARY AND POST-TERTIARY VERTEBRATA. 
Siwaliks near the village of Padri in the Punjab. The teeth of this specimen are 
in an intermediate condition of wear, and the first premolar (pm. 1) is seen to be 
persistent. These teeth agree with the specimens last noticed in the comparative 
shortness of the grinding surface of the anterior ‘ pillars 5 of the premolars, i.e., they 
are not longer than the corresponding surface in the second true molar (m. 2) ; the 
specimen is accordingly referred to the present species. 
In figure 1 of the same plate there are represented the five last teeth of the 
right upper molar series of a horse obtained by Mr. Theobald from the higher 
Siwaliks of the village of Rupur in the Punjab. These teeth are somewhat less 
worn than those of the previous specimen, and exhibit the characteristic shortness 
of the grinding surface of the anterior c pillars ’ of the premolars, this being 
especially noticeable in this specimen, where this surface in the last premolar (pm. 4) 
is shorter than in the first true molar (m. I). 
The last specimen of the upper molar dentition of this species to be noticed is 
contained in a fragment of the right maxilla : another of Mr. Theobald’s numerous 
Siwalik specimens (No. C180, Ind. Mus.). In this specimen, which has not been 
figured, the four last teeth of the molar series are exhibited, the last true molar 
having only just come into wear : this specimen, therefore, belongs to a younger 
individual than either of the others. In it the length of the grinding surface of the 
anterior ‘ pillar ’ of the last premolar is considerably shorter than the corresponding 
surface of the first true molar, the respective lengths being 04 and 0 51 inch. 
Taking, therefore, the five specimens above mentioned, which exhibit the teeth 
at all stages of wear, we find it to be a constant character of the upper molars of 
Equus sivalensis that the grinding surfaces of the anterior ‘ pillars ’ of the premo- 
lars are not longer than those of the later true molars, and are frequently shorter 
than the corresponding surface of the first true molar. The number of specimens 
examined, and their different ages, leave little doubt as to the trustworthiness of 
this character. It may be noticed that a considerable difference occurs in the size 
of the molars of the two specimens figured in this volume, but it does not appear to 
me that this can be considered as more than an individual or sexual character. In the 
only specimen exhibiting the complete molar series, the first premolar (milk-molar) 
is persistent, and of considerably larger size than the same tooth in living horses. 
Comparison. — The difficulty of arriving at any satisfactory conclusion merely 
from a comparisons of the teeth of the horse-family is well instanced by the numerous 
so-called species which have been made from the European fossil remains of the 
domestic horse {Equus coballus). The teeth of many of the different living 
species are, indeed, so much alike that it would, I think, be impossible to distin- 
guish many of them by the characters of the molar teeth alone. In the case 
of the fossil Siwalik horse, from the materials at my command, my comparisons 
must, perforce, be limited to the living Asiatic species. 
In Equus caballus I cannot discover any instances where the anterior ‘ pillar 5 
is as small as it frequently is in Equus sivalensis. In the former the second pre- 
