ADDITIONAL SIWALIK PERISSODACTYLA AND PROBOSCIDIA. 3 
At the entrance to the same valley there is a large, blunt tubercle {g) attached 
exclusively to the posterior collis (^). The latter gives off a blunt projection into 
the median valley, placed a short distance internally to the ante-crochet (/), and 
more externally a distinct, though small crochet (e) : there is no combing-plate at 
the extremity of the median valley. The posterior valley (^) is elongated antero- 
posteriorly, and is much less deep than the median valley ; the descent from its 
outer wall being gradual. Tlie cingulum forming the anterior valley (left of a) does 
not extend beyond the inner half of the tooth ; and in one specimen (fig. 2) stops 
short of the inner surface of the anterjor collis, but in the other extends a short 
distance on to this part. 
In the last molar (m. 3, fig. 1) the general characters are the same as in the 
preceding tooth : the anterior cingulum extends, however, still more on to the inner 
face of the anterior collis (a), and the tubercle at the entrance to the median valley 
is larger. The much Avorn first molar (m. 1, fig. 1) shows the median valley all but 
divided by the aboAm-mentioned obstruction into tA^x) distinct portions, the inner of 
which is triangular, and the outer sub-elliptical in shape. An equally Avorn m. 1 from 
Gandoi represented in plate II., fig. 2, exhibits these characters still more clearly. 
Ug)pcr cheek-teeth of smaller race. — In plate II., fig. 4, there is represented the left 
maxilla of a smaller rhinoceros from Gandoi, containing all the permanent cheek- 
teeth, Avith the exception of pm. 1 , in a medium condition of Avear ; tlie first and last 
of the series being someAvhat damaged. With the exception of their considerably 
smaller size the true molars of this specimen agree precisel}" Avith the teeth described 
above ; and since there is a considerable variation in the size of different races of 
some of the existing species of rhinoceros, there appears no good reason for regarding 
the Gandoi specimen otherwise than as belonging to a small race of the species to 
Avhich the larger teeth belong. 
The premolars of this specimen have their inner halves surrounded by a 
complete, sharp-edged cingulum ; whose free edge extends much higher ujd on the 
crown of the posterior (h) than on that of the anterior collis (a). The latter 
extends considerably more inwardly than tlie former ; which is, so to speak, buried 
in the cingulum. With the exception tliat the ante-crochet is less developed, the 
other general characters of the premolars are similar to those of the true molars. 
In pm. 4 there is a small combing-plate.^ 
Comparisons loith other Siiualik rhinoceroses. — A comparison of the figures of the 
specimens described above Avith the small maxilla rej^resented in vol. II., pi. VI., 
fig. 1, will shoAV pretty clearly that all belong to the same sjDecies, the Punjab 
specimen being intermediate in point of size betAveen those figured in this volume. 
Accepting this identification, it remains to shoAV in what respects these specimens 
differ from Rhinoceros palmindicus, with Avhich, in the perhaps over anxiety to avoid 
making unnecessary species, the first specimen was provisionally classed. In the 
1 There is another specimen of the same size from Gandoi in the Indian Museum (Xo. 0. 266), exhibiting the three true 
molars of the right side in a much damaged condition. 
