48 
Records of the Geological Surveg of India. [voL. xn 
have been the two middle premolars, because no such discrepancy in size occurs 
between the last premolar and first true molar as occurs between these two teeth, 
while the smaller tooth is of too small dimensions to have been the last true 
molar. I may add that the tooth represented on Plate VI, figure 2 of the above 
referred to volume of the “ Palseontologia Indica,” as the first true molar of 
A.perimense, is really the penultimate premolar; and that the unnamed tooth 
from Sind, represented on figure 6 of the same plate, seems to be the antepenulti¬ 
mate upper premolar of the same species. 
On a future occasion I shall hope to be able to give figures of the almost 
complete upper and lower dentition of the present species ; and I cannot but regret 
that I have previously published figures of such very imperfect specimens. It is 
interesting to observe that A. perimense agreed with the European A. incisivum, 
in being hornless, and in being furnished with a single pair of very large upper 
and lower incisors, clearly showing that the absence of one weapon of offence or 
defence is compensated for by the greater development of another. 
I may here mention that we seem to be gradually obtaining evidence that 
the mammalian fauna of the Punjab and Sind forms a connecting link between 
the fauna of Perim Island on the one hand and of the more eastern Siwaliks on 
the other. Thus, as will be gathered from a perusal of this and my previous 
papers in the Kecords, we have in the Siwaliks of the Punjab and Sind the 
following Perim Island mammals, which were not known to Falconer fix)m the 
more easterly Siwaliks, viz. : — 
Dinotherluui indicum. P.; I. P. S. 
Mastodon pandionis. P.; I. P, Deccan (?) 
Mastodon perimensis. P.; I. P. S. (?) 
Hyotherium sindieuse. P .; I. S. 
Acerotherium perimense. P.; I. P. S. (?) 
Hippotherium tlieobaldi. P.; I. P. 
All these mammals belong to old forms, and seem to indicate that the Perim Island 
deposits and the zone in which they occur in the Punjab (position unknown) 
are low down in the series and correspond to the older Sind Siwaliks. 
Distribution of genera of SiwaliJc Mammals. 
Since the publication of my paper on the “ Fossil Mammalian Fauna of 
India and Burma, ” * several new genera have been added to these fauna, and 
the distribution of the previously known genera has been further elucidated. I 
have therefore compiled the following table of the distribution of the mammalian 
genera in the Siwalik and the other tertiaries below the Nerbudda group, which 
must be taken as superseding the tables given on pages 90-92 of my above 
quoted memoir. 
* Rec. Gcol. Surv. Iiid., Vol. IX, pt. III. 
