INTRODUCTORY OBSERVATIONS. 
XIX. 
This distribution is of peculiar interest in regard to the apparent relationship of M. 
cautleiji to the European hi. longirostris, and of its cousin M. latidens to Elephs clifti^ 
and thus to the higher stegodonts, which are especially characteristic of the typical 
Siwalik Hills and the countries to the eastward {vide infra, p. 18). 
With regard to the milk-molars figured in plate XVII. of the present volume 
under the name of M. perimensis, it is very difficult to determine whether they 
belong to that species as now restricted, or to M. caatleyi. The small hind talon of 
the specimen represented in figs. 3, 3a (which Falconer had labelled M. latidens^) ^ 
coupled with the comparative clearness of its valleys, and the absence of any 
distinctly alternate arrangement of its columns, render it, however, not improbable 
that this specimen should be referred to M. cautleyi. The writer has at present been 
unable to refer any lower teeth to the latter form.^ 
Mastodon angustidens. — On page 32 of the present volume two teeth were 
described and provisionally regarded as upper milk-molars of M. falconeri ; the more 
perfect example being figured in the woodcut on the same page (reproduced in fig* 
Fig. 9. 2Iastodo}i angustidens var. palccindicus. 
The first and second ridges of the first right 
lower true molar, in a half-worn condition ; 
from the Lower Siwaliks of Sin 1 or the regions 
to the north or west, p British Rluseum. 
(No. 40788). 
sists of the second and third ridges of the homologous tooth, in a somewhat 
earlier stage of wear. It is not improbable that both specimens were obtained from 
the Biigti Hills or Dera Bugti {vide infra page 1). 
Mastodon pandionis. — It may be advisable to note here that the tooth figured in 
vol. I. pi. XXXII. fig. 4 under the name of M. falconeri been referred on page 
30 of the present volume to M. pandionis. 
A penultimate lower milk-molar of a Mastodon from Perim Island in the 
collection of the British Museum which has recently come under the writer’s notice 
affords some clue to the vexed question of the specific determination of the milk- 
molars figured in vol. I. pis. XXXII. figs. 2, 3, and XXXIII. fig. 2, of which 
1 Vide infra. -p. 151. 
2 It is not improbable that one of the specimens of mandibular symphyses figured in vol. 1. pi. XLIII. under the name 
of M. perimensis may belong to M. cautleyi, as both differ somewhat from the mandible of the former described on p. 245 of 
the same volume. 
9). A re-examination of these specimens has, 
however, convinced the writer that he was 
mistaken in regarding them as complete teeth, 
and that they are really imperfect examples of 
the first right lower true molar of M. angustidens 
var. palceindicus. The figured specimen com- 
prises the first and second ridges, and its 
identity with ni. 1 of that species may be seen 
by comparing the woodcut with the specimen 
represented in plate IV. fig. 8, which belongs 
to the ojiposite side of the jaw, and is in a 
somewhat more worn condition. The second 
specimen (British Museum. No. 32503) con- 
