ADDITIONAL SIWALIK PERISSODACTYLA AND PROBOSCIDIA. 25 
whicli this tootli differs from m. I of M. iiandionis will be alluded to under the head 
of that species. 
The still more worn hinder portion of a right lower molar represented in pi. V., 
fif>'. 7, from its being narrower and somewhat smaller than the second true molar 
described above, may in all probability be likewise regarded as a first true molar. 
This tooth is of a shorter type than the last specimen ; but otherwise agrees with it. 
It is of very nearly the same size as an homologous upper tooth figured by H. von 
MeyeP ; and might have belonged to an individual of the same size as that to which 
the tooth represented in pi. IV., fig. 1 belonged. 
Fourth lower milJc-molar. — The perfect tooth represented in pi. IV., fig. 5, 
evidently belongs to the same species as the preceding specimens ; and from its 
small size must be the last milk-molar, probably belonging to an individual of the 
size of that to which the first true molar represented in fig. 8 of the same plate 
belonged. It is of nearly the same size as a last milk-molar of M. angustidens figured 
by Kaup,^ but rather smaller than one figured by MM. Lortet and Chantre.® The 
specimen is in a middle stage of wear ; and is almost a miniature facsimile of the 
tooth represented in fig. 7 of the same plate. It agrees with European specimens of 
mm. 4 of M. angustidens^ with the_ exception of the somewhat larger size of the 
posterior accessory columns (a). The cingulum and the tubercles at the outer 
extremities of the transverse valle 3 ^s [tiC) are particularly well displayed. The 
length of this tooth is 2*4; its greatest width D37 ; and the height of the worn 
inner column of the third ridge O’ 78 inch. The last lower milk-molar of M. 
pandionis"^ has a higher crown ; and a length of 3 ’4, with a width of 2’3 inches. 
The corresponding tooth of M. falconeri is unknown. 
Last upper joremolar. — At least two types of last upper premolars of trilophodont 
mastodons are known from the Siwaliks ; but as these have never been found 
associated with the true molars there is very great difficulty in assigning them to 
their respective species ; and their reference must accordingly be considered more or 
less provisional. In pi. V., figs. 2, 2a, there is represented a tooth from Dera Bugti, 
which, from certain reasons noted below,^ and from its resemblance to pm. 4 of 
M. angustidens^ is provisionally referred to the present form. The tooth is in germ ; 
the crown not having become attached to the root. The sub-quadrate shape of the 
crown indicates that it should be referred to the upper premolar series, of which it 
must be the last; and the forward inclination of the columns indicates that it 
1 Op. cit., pi. III., fig. 5: the specimen is called the last milk-molar, hut is evidently m. 1 : many of the earlier teeth 
figured hy Meyer are classed one place too early in the series 
2 Op. cit., pt. 3, pi. I., fig. 2. 3 Op. cit., pi XV., fig. 4. 4 Supra , vol. I , p 220. 
5 Since describing the remains of mastodons in the first volume the writer has found out that worn teeth of the premolar 
series may generally he distinguished from those of the milk-molar series bj" the fact of the hinder ridge being more worn 
than the front one ; the reverse condition generally prevailing in the milk-molars. Thus the specimen represented in pi. V., 
fig. 6, from the forward inclination of the ridges is referred to the right side ; and consequently from the hind ridge being 
the most worn, is probably a premolar : the squared form of the crown .shows that it belongs to the upper jaw. The cause of 
the earlier wear of the hind ridge in the prcmolars is that this part of the tooth comes in contact with the already fully 
protruded milk, or true molar immediately behind it in the dental series of the opposing jaw. 
G 
