ADDITIONAL SIWALIK PERISSODACTYLA AND PROBOSCIDIA. 31 
Fig. 5. (?) Mastedon patidionis, 
Falc. Last right upper premolar, 
from DeraBugti: Indian Museum 
In all the above-mentioned characters this specimen agrees with the teeth of 
M. pancUoiiis ; and as it has precisely the same dimensions as the right m. i of that 
species represented in plate XXXV., fig. 1, of the first volume, it may safely be 
regarded as the homologous tooth of the opposite side.^ 
Tipper premolar. — The tooth represented in figs. 6, 6aof plate V., and the one in the 
woodcut (fig. 5), are two of three similar specimens from Dera 
Biigti. The former has lost the greater part of the cingulum, 
and the inner face of the first inner column, and is less worn 
than the latter specimen. The sub-quadrate form of the 
crown shows that the specimens belong to the upper jaw; 
while tlie forward inclination of the larger columns (fig. 6a) 
indicates that they are from the right side. The absence of 
a disc of pressure on the anterior side, the rounded angles of 
the crown, and the more worn condition of the hinder as 
compared with the front columns, indicates that the teeth 
(No. A. 432). 1. belong to the premolar series, of which they are the last. 
From the strongly-marked vertical corrugations in the enamel, the forward inclination 
of the columns, and the obstructed transverse valleys, it is inferred that these 
specimens probably belong to the present species. Their distinction from the 
corresponding tooth referred to M. angustidens (pi. V., figs. 2, 2a) will be sufficiently 
apparent from a comparison of the figures ; and from their complex crowns, as well 
as from the absence of M. falconcri from Dera Biigti, it is imjirobable that they 
belong to that species. The present teeth are of relatively smaller size when 
compared with the true molars of If. pandionis^ than is the case with the corresjiond- 
ing teeth of ilf. angustidens. 
Affinities . — The affinities of the present species have been already alluded to in 
the first volume,^ but a few remarks may be added. That the species is allied to 
M. angustidens is pretty evident from the general form of the mandible, and the 
structure of the molars. The greater complexity of the latter, and the addition of 
cement, indicates, however, that the present species is a more specialized form, and 
may possibly have sprung from the progenitors of the Indian race of M. angustidens.^ 
which has been shown to diverge somewhat from the typical race in the direction of 
the present species. The absence of M. pandionis from the Eastern Siwaliks indicates 
the probability of its being an immigrant into India from the west ; which confirms 
its relationship to M. angustidens. From the coexistence of the two forms in the 
lower Siwaliks of Dera Bugti their genetic connection probably took place in the 
regions to the westward of India. i¥. pandionis appears to have survived longer, and 
penetrated further into India than AI. angustidens, since its remains occur abundantly 
in the upper Siwaliks of the Punjab to the westward of the river Jhelam. 
The circumstance that the last molars of the upper Siwalik M. sivalensis have a 
1 The specimen figured in the first volume is narrower posteriorly, hut this is probably due to its imperfect protrusion. 
2 P. 225, et. seq. 
